Deep Freeze 9

Vendor representations: a solution to the decoupled receiver problem

2012-01-01T11:01:00.000-08:00

Requiring receiver vendors to represent to buyers that their equipment is fit for purpose is a way to avoid cheap receivers from reducing the performance of coexisting systems in the “decoupled receiver” case, i.e. when there isn't a license holder to negotiate with.

Ensuring adequate receiver performance using protection limits

The starting point for any receiver regulation is an interference protection limit that establishes a ceiling on the interference from other systems (co-channel and out-of-band, though the hard problem is out-of-band) that a system operator has to cope with. It defines the radio signal environment rather than tackling the immensely more complex regulatory challenge of specifying receiver performance.

A protection limit sets the entitlement point around which disputes are resolved. For example, if interference is above the limit, there's an objectively verifiable claim of harmful interference. It also facilitates negotiation: the receiver operator can pay other systems to reduce their delivered energy if it wants to build cheaper receivers; or the receiver operator can be paid by a neighbor (and they could use the money to improve their front-end filters, say) if the neighbor wants to increase their filters. (See "Receiver protection limits: a better way to manage interference than receiver standards" and follow-on posts list there for details.)

Licenses are tied to transmitters. When a licensee controls the receivers in their system, the licensee is the one who is able to make claims of harmful interference, and who can negotiate with neighbors about increased or reduced interference. The licensee has the means and incentive to tailor the performance of their receivers.

The decoupled receiver problem

However, this isn't possible in the decoupled receiver case where receivers are not controlled by a license holder. Anybody can put a lousy GPS or TV set into the market, say, and there's no effective way to hold the manufacturer accountable. An end user who suffers interference, even though the energy from neighboring systems is below the protection limit, can only complain to Congress, with all the messiness that entails (cf. the notorious “garage door opener” case).

Now, given a protection limit, the FCC could theoretically devise a receiver standard and impose it on all manufacturers – but as I’ve argued, that's not only bad policy, it's also a non-starter practically speaking. So what else can one do?

I’ve proposed accrediting a limited number of entities, the holders of the right to protection against interference, who can then authorize any number of manufacturers to put products into the market (see my blog post "Stamps and stampholders: a third way to regulate radio operation"). However, this solution may draw objections from those who cannot countenance any restriction whatsoever on the number of players in the market for receivers (even though it would not limit the number of potential manufacturers).

The “vendor representation” solution

The subject of this post is a third possibility: The FCC sets protection limits, and then requires vendors to warrant to customers that their equipment is fit for purpose, given that the interference environment as defined by the receiver protection limit, and the intended use of the device.
It’s up to the vendor to define the purpose of their device and decide the performance level that will be sufficient. The vendor determines how to respond to the protection limit. This keeps the FCC out of the impossible business of defining and updating receiver performance standards.

Enforcement would happen through trade law, i.e. customers, the FCC or FTC could bring a deceptive advertising case if a receiver was so poor that it could serve its purpose even for signal strength below the protection limit. Enforcement would be against the company selling the device, not the (perhaps off-shore) manufacturer. If and when such a case arose, a tribunal (e.g. the FTC) would have to consider receiver performance requirements, but it would be in a particular, well-defined context. The FCC wouldn’t be defining receiver standards speculatively, trying to cover all possible eventualities.

To see how this helps, imagine that (a) the FCC decides that a certain field strength ceiling in the Lower 10 MHz (1526-1536 MHz) would protect GPS receivers tuned to 1575.42 MHz from harmful interference by LightSquared, and (b) a user buys a GPS device that fails when LightSquared starts operating in their vicinity. With current standard low Q filter, a consumer navigation device would work fine, and the user won't suffer service degradation. However, if someone tried to sell a precision surveying system using only a low Q filter, it would fail. Therefore, if a vendor wanted to sell a satnav system and claim that it allowed highly precise location determination, it would have to use a high Q filter, or ensure adequate precision in some other way. In this way, the representation ensures that receivers are designed to the required quality, but keeps the FCC out of the business of defining standards that are either unnecessarily strict (e.g. a single standard for all GPS receivers), or parsing devices into pretty arbitrary categories and defining different receiver standards for each (e.g. consumer navigation vs. high precision surveying).

From spectrum efficiency metrics to parameter spaces

2011-12-30T10:50:00.000-08:00

In my post FCC white paper shows that “spectrum efficiency” is meaningless I argued that spectrum efficiency metrics are not very helpful.

They won’t go away, though, because engineers and economists instinctively characterize systems numerically. Both tribes strive to separate a problem into smaller independent parts, each described quantitatively; metrics are just a symptom. The goal is to convert a complex mess into a problem amenable to objective analysis, yielding an incontrovertible answer. No more messy politics!

Since politicians always look for cover behind engineers and economists, simplistic metrics will always be with us – not least in radio regulation. Given that reality, I’m going to dig into spectrum metrics a little more. I conclude that it could be more productive to define a series of axes in a parameter space than a single metric.

Any metric is a model, i.e. an abstraction. It leaves a lot of stuff out in order to provide a useful thumbnail. As George Box said, “All models are wrong, but some are useful.” For example, to quote from the TAC “spectrum efficiency metrics” white paper (DOC), “communications systems must often meet basic user needs in a number of quality of service (QoS) measures, including latency/access time, coverage/reliability, information error rates, and peak-loading requirements. Maintaining this service level or even improving it in some of these areas may have a negative impact on spectral efficiency metrics, but may be required for particular system applications. ”

There are lots of ways to measure radio operation. It’s worth distinguishing between rulers and ratios:

Rulers: a measure of a quantity of interest. Broadly speaking of two kinds:

Inputs: e.g. MHz, sq. miles, POPs, $ infrastructure investment, $ backhaul cost, $ device cost, maximum transmit power, maximum PFD, antenna size, consumed field of view or orbital arc or geographic region (for satellites); combinations of these like MHz*POP
Outputs: e.g. bps, $ surplus, system response time, bit error rate, device size, coverage, peak loading requirement. Note that some important goods that flow from radio operation cannot be quantified, e.g. public safety, distributional equity, basic knowledge.

Ratios: built using rulers, e.g. $/MHz*POP, bps/Hz, $ surplus / $ investment.

Rather than picking any single one of these, it could be more productive to define a series of axes in a parameter space. One could then plot different services in this space; rather than requiring that all of them are in the "top right" of the space, one could look at the various clusters and try to move services withing clusters out to the leading edge in parameter space.

A parameter space is a bunch of axes or dimensions that can be rulers or ratios. The paradigmatic one in wireless is electrospace (Hinchman 1969; Matheson 2003 et seq.) that defines radio field strength in a seven-dimensional space {frequency, time, spatial location, direction-of-travel}; it’s defined using input rulers as axes.

A given ratio metric is a surface in a parameters space. For example, a given body mass index (weight divided by height squared) parameterizes a parabola if you plot weight against height. A bunch of BMI’s define boundaries between regions that represent being underweight, normal, overweight and obese. If you then plot BMI against age, you get growth curves for BMI-for-age percentiles (Wikipedia).

In order to get a flavor of how this might work in radio regulation, here are the metrics in the TAC white paper (omitting the “Additional Efficiency Considerations”). I’ve translated the dimensions into SI units; they vary quite a bit in the document.

Satellite broadcast systems: bits / (second * Hz) within each common program area
Point-to-point satellite systems: bits / (second * Hz * meter^2)
Terrestrial Broadcast Systems: bits / (second * Hz) within each common geographic area * the average number of users simultaneously served
Personal Communications Systems (aka PCS): bits / (second * Hz * meter^2)
Point-to-Point Terrestrial Systems: (transmitted distance) * bits / (second * Hz* meter^2)
Hybrid Terrestrial Systems: (info bits / second / Hz) * meter^2 * the average number of users simultaneously served.

(Note that the area in Hybrid Terrestrial Systems seems to be in the numerator, not the denominator; this probably relates to the fact that for public safety systems, similar to terrestrial broadcasting systems, as the number of users increases, the spectrum efficiency improves when compared to point-to-point systems where each additional user consumes additional capacity.)

In could represent all these metrics as surfaces in a parameter space with these dimensions:

bits / (second * Hz); this could be split into two dimensions, bits/second and Hz
area (meter^2)
transmitted distance (meters)
average number of users simultaneously served

Note that the surfaces depend on the kind of service. Thus, for PCS systems, the figure of merit goes as bits/(second*Hz) / meter^2, whereas for terrestrial broadcast and public safety it’s bits/(second*Hz) * meter^2. Increasing the figure of merit for PCS increases the slope, whereas increasing it for broadcast pushes it out towards the top-right. It’s therefore obvious that it’s hard to compare different categories of service with each other.

For systems that aren’t measured on a particular axis (e.g. the satellite broadcast systems metric omits area), the representation of the metric becomes a section of the space, e.g.

Stamps and Stewards: A third way to regulate radio operation

2011-12-27T16:03:00.000-08:00

Radio operation to date has largely been regulated in two ways. The dominant approach has been licensing station operators, whether they’re amateurs, TV broadcasters, or companies operating cellular systems. In the last twenty years or so, device licensing (aka unlicensed in the US, and license exemption in Europe) has also become widely used: if a device has been certified to meet regulatory requirements, anyone can operate a “station” using it without needing a license. [1] In these two approaches, the regulation controls either the system operator (for licensed), or the device manufacturer (for unlicensed).

I’m exploring another way, where the regulator accredits a limited number of “stampholders” who can each an issue an unlimited number of “stamps.” One can see these stampholders as the designated stewards of a "spectrum commons," and the stamps as the mechanism they use for controlling access to a common pool resource. A device may only be sold if it bears the requisite stamp or seal, in addition to any other statutory requirements such as Part 15 certification. Control is exercised at the point of sale through labeling or marks.

This notes builds on the previous posts Licensing radio receivers (Aug 2011) and Licensed Unlicensed (Sep 2011). I learned long ago that if I can think of something, someone’s already done it. However, I haven’t found good precedents yet, and I’m still looking for canonical examples or ringing metaphors. Stamps (in the sense of signet rings and seals) and Stewards is the best analogy I’ve found so far. [2]

Uses

The stampholder method is suitable for both for unlicensed allocations, and decoupled receivers. I think it will be particularly useful in so-called dynamic spectrum access or cognitive radio sharing scenarios, but I'll leave that discussion for later.

Unlicensed applications

The stewards/stamps approach is designed to retain the benefits of unlicensed – e.g. anyone’s ability to deploy a radio network without obtaining station license, by choosing from a wide array competing devices – while addressing its collective action problems: (1) the lack of coordination (e.g. failure of IEEE 802.11 or the Wi-Fi Alliance to agree on unlicensed-unlicensed dynamic frequency selection or transmit power control etiquettes) which leads to inefficient sharing when there are many mutually interfering devices; and (2) the inability to adjust boundaries (e.g. power levels) between unlicensed bands and their neighboring bands by negotiation. A limited number of stampholders would be more likely to reach agreement on sharing etiquettes, and would be less likely to suffer from collective action problems when, say, a neighboring licensee would be willing to make a payment if all unlicensed devices reduced their transmit power, or when the neighbor would accept more interference if all the unlicensed players could band together to come up with an acceptable payment.

Stampholding also provides a mechanism to kill off an allocation if it has proven to be a failure – zombie allocations are a weakness of the current unlicensed approach (see my post Time limiting unlicensed authorizations). For example, the reserve price for an allocation of stampholdings could be set at 80% of the revenue raised at the previous auction. If the reserve price isn’t met by the aggregate of all the bids, no stamps are issued, and no new devices can be sold from then on. Once legacy devices sold have been deprecated, the band can be reallocated.

Managing decoupled receivers

Dale Hatfield has dubbed the situation where receivers are not controlled by a license holder the “decoupled receiver problem.” Examples include television receivers, unlicensed satellite weather receivers, and GPS devices. The problem arises because (1) the regulator has most leverage over license holders, and these devices aren’t licensed; and (2) the unlimited number of receiver manufacturers and operators makes it impossible for a neighboring licensee to come to a negotiated agreement over boundary disputes, since there are too many entities to bargain with – the same collective action problem seen in unlicensed.

I proposed "licensing" receivers as a way to deal with this problem in the August post. The word licensing isn't ideal; it not only makes some people uncomfortable, but the FCC (at least) may not have the authority to license receivers under current statute. The stamps and stewards works in a similar way, suggesting that licensing may have been the wrong way to think about the solution.

Mechanism

As explained in Licensed Unlicensed, the regulator would define the rules and entitlements for any operation in a particular band, but only devices bearing a prescribed mark or stamp could be sold or operated. A limited number of entities – the “stampholders” or "commons stewards" – would be authorized to issue these stamps, but there would be no limit on the number of stamps each could issue.

Since the word “stamp” can mean either the object making a stamp, or the resulting mark (the same is true, probably not coincidentally, for cognate terms like “seal”), I will use the term “stampmaker” where necessary to make it clear that I’m referring to the stampmaking object, not the mark. Thus in this metaphor, the regulator issues a limited number of stampmakers, one to a stampholder.

The limitation on the number of stampholders is intended to overcome a key deficiency of unlicensed as a commons: the lack of control on the number of actors managing the common pool resource. [3]

Depending on one’s ideological preferences, a stampholder can be seen as a rentier deriving income from its control of a stampmaker, or as a steward of a “spectrum commons.” It can charge fees from, and/or impose other behavioral conditions on, companies that want to put a product into the market. The rent it thus obtains is a reward taking the risk of purchasing a stampmaker (there may not be a sufficiently large market in stamps to repay their investment), and for being a joint steward of the commons of radio operations.

Additional conditions imposed by stampholders

In the unlicensed case, all devices would operate under Part 15 rules, e.g. transmit power limits, and requirements not to cause interference to, but accept interference from, other operations. Only manufacturers authorized by stampholders could seek Past 15 certification. A stampholder could impose additional requirements on devices as a condition for receiving a stamp, such as implementing a particular spectrum etiquette or industry standard.

In the case of decoupled receivers, the stampholders would be the holders of rights embodied in the receiver protection limits in this allocation. They would decide how receivers bearing their stamp would take protection limits into account; for example, they might define receiver standards for devices bearing their stamp, or collaborate with other stampholders to agree on standards all their devices would conform to.

Expiration

A stampholding accreditation could, and I think should, have a limited term, say 15 years. Since stampholders are not required to build out network, as cellular licensees are, there is no requirement for massive infrastructure investment that would support arguments for the indefinite terms of the entitlement. At the end of the term, current stampholdings would expire, and the regulator would hold a new auction. At this point, it could change the rules of the game if necessary, e.g. by changing the number of stampholders. (I have proposed elsewhere that regulators should make greater use license renewal to adjust rules, while conversely abstaining from any rule changes during the term of a license; see De Vries & Sieh 2011 http://ssrn.com/abstract=1907813.)

Number of stampholders

Judgment will be required to determine the number of stampholders/stewards to be accredited. There are countervailing goals: facilitating negotiation requires as few as possible, while avoiding market power abuses and increasing innovation by market entry suggests a larger number. It depends in part on whether one is regulating a commodity or differentiated market (Mark Bykowsky, personal communication). If the devices in a particular allocation are commodities, three stampholders may suffice, whereas a variegated, multi-sectoral, rapidly changing scene might indicate ten. The number of stampholders will thus vary from allocation to allocation, and may change over time in a given allocation (e.g. by limiting the term of stampholding, and changing the number at renewal time).

Since one of the considerations when choosing the number of stampholders is the maintenance of market diversity and competition, rules may be required to prevent the aggregation of multiple stampholdings in the same allocation by one entity.

Regulatory restraint

However, the regulator will have to exercise great restraint in not attaching “political” conditions to stampholding. The mechanism should be used merely to designate a limited number of entities; all technical, operational and commercial decisions, with the exception of the most basic transmission permissions and receiver protections, should be delegated to them.

Designating stewards

As I proposed in August, the stewards/stampholders could be identified by auction. Since there would be no limitation on the kind of entity that could become a stampholder, they might be RF chipset vendors (e.g. Broadcom, CSR, Qualcomm), device manufacturers (e.g. Cisco, Nokia, Motorola), service providers (e.g. Google, Microsoft, Verizon), certification providers such as UL, or consortiums such as SpectrumCo. Given the limited number of stampholdings, not every player at every stage in the value chain could be a stampholder. However, a company at one stage in the value chain could sign on with a stampholder at another stage, and so not be beholden to their competitors; so for example, chipset vendors that were not stampholders (Texas Instruments, for argument sake) could affiliate with a service provider (say, Google).

Auction design is rocket science, and I couldn’t even fold a paper airplane… but if I had to venture a mechanism, one might think in terms of a multiple round, ascending bid auction with the auctioneer (e.g. the FCC) declaring lowest bid at the end of every round. The auctioneer declares at the start how many accreditations, say N, will be issued. When bidding ceases, the highest N bids receive accreditation, paying the price they bid.

An auction for unlicensed stewards would differ substantially from the mechanism proposed by Bykowsky, Olson and Sharkey (BOS) for using an auction to decide whether to allocate bands to unlicensed use. [4], [5] For example, in the BOS proposal, winning an unlicensed allocation via an auction doesn’t endow the bidders with any privileges, other than having an unlicensed authorization at all. However, there are resonances between the approaches. For example, the stampholder approach provides additional incentive for participating in the bidding for an unlicensed allocation in a BOS auction, since non-winning bidders do not obtain accreditation. (Note, though, that Bykowsky does not believe that such an incentive is required, since their experimental results suggest that unlicensed bidders can overcome collective action problems.)

Update 7 Jan 2012: Gave a little more emphasis to the "commons steward" metaphor

Changed title from "Stamps and stampholders: a third way to regulate radio operation" to "Stamps and Stewards: a third way to regulate radio operation"
In a few places, replaced "stampholders" by "stewards" or "stewards/stampholders"

Notes

I thank Mark Bykowsky and Susan Tonkin for helpful conversations.

[1] There have been a variety of so-called hybrid or flexible license approaches, such as “light licensing” where anyone may operate in a band, subject only to registration (e.g. 3650-3700 MHz in the US, 5725-5850 MHz in the UK). Whitespace databases that provide ad hoc permission to otherwise unlicensed devices to operate in a particular frequency ranges at a specified place and time combine elements of a licensed and unlicensed regime (Stirling 2010). There are also “license by rule” cases where transmitters operate under a blanket authorization (e.g. walkie-talkies in the Family Radio Service)

[2] Other analogies include licensing taxicab companies not vehicles or drivers, Iceland’s fishing quota system, and privateering commissions. None of them capture all the essentials, and all diverge in important ways from the stampholder/stamp model.

[3] The successful common pool resource (CPR) management cases that Elinor Ostrom has described all seem to have a relatively small number of participants. Many of the eight design principles for CPRs that she describes in Governing the Commons: The evolution of institutions for collective action (1990) presume resource appropriators that know each other - very unlike the case in unlicensed wireless allocations, but quite similar to the traditional interactions between the engineers of adjoining radio licenses. To quote from a few of the principles (Table 3.1, p. 90): #1 "Clearly defined boundaries: Individuals of households who have rights to withdraw resource units from the CPR must be clearly defined"; #3 "Collective-choice arrangements: Most individuals affected by the operational rules can participate in modifying the operational rules"; #6 "Conflict-resolution mechanisms: Appropriators and their officials have rapid access to low-cost local arenas to resolve conflicts".

[4] M. Bykowsky, M. Olson, and W. Sharkey, "A Market-Based approach to establishing licensing rules: Licensed versus unlicensed use of spectrum," FCC OSP, Tech. Rep. 43, Feb. 2008. http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-280523A1.pdf

[5] M. M. Bykowsky, M. Olson, and W. W. Sharkey, "Efficiency gains from using a market approach to spectrum management☆," Information Economics and Policy, vol. 22, no. 1, pp. 73-90, Mar. 2010. http://dx.doi.org/10.1016/j.infoecopol.2009.12.003

Spectrum utilization and a Buddhist perspective on space

2011-12-05T11:37:00.001-08:00

The “Spectrum as Space” metaphor implies that spectrum is a neutral container that can be filled with radio signals, leading to naïve notions of utilization such as empty and full spectrum bands. “Spectrum” is imagined a collection of axes which mark out an abstract space, such as frequency, geography, and time (e.g. Robert Matheson’s “electrospace” concept, cf. Matheson & Morris 2011, The Technical Basis for Spectrum Rights: Policies to Enhance Market Efficiency).

However, that’s not the only way to look at it. Non-spatial models such “Wireless as Trademark” work just as well (see my 2008 paper De-Situating Spectrum: Rethinking Radio Policy Using Non-Spatial Metaphors): by analogy, a trademark stands for both a part of the wireless resource (customarily, frequency band x geographic region x time slot), and signals. The wireless resource is all possible radio operations. In such an approach, one is much less likely to ignore the importance of receivers in this approach than spectrum-as-space, where only transmitters can “fill the space” with signals. Any radio operation includes the use of a receiver, and that receiver-transmitter pair influences what transmissions are possible by third parties.

Curiously, I found a relevant perspective on this problem in a book on ethics – Stephen Batchelor’s Living with the Devil. He writes:

One tends to think of space in terms of physical extension and location. A body “occupies” or “fills” a space. For there to be “no more space” means that nothing more can be fitted into a room or a vehicle or a document. Outer space is that virtually infinite expanse speckled with galaxies and stars separated by inconceivable distances. “Inner space” suggests a formless expanse of mind in which thoughts, mental images, memories, and fantasies rise and pass away. Space seems to be the relatively permanent place where temporal events happen.

Buddhist philosophers see space differently. They define it as the “absence of resistance.” The space in a room is under stood as the absence of anything that would prevent one moving around in it. To cross from one side of the room to the other is possible because nothing gets in your way. Rather than being the place where things happen, space is the absence of what prevents things from happening. The space in the room is nothing in itself; it is just the absence of chairs or tables, glass walls or hidden tripwires that would obstruct movement within it. In encountering no such resistance, we are able to move about freely. [In the footnotes, Batchelor ascribes this approach to the Geluk school of Tibetan Buddhism.]

The customary view that Batchelor outlines is “space as a set of dimensions” that informs the Spectrum as Space metaphor. One can transpose his summary to spectrum as “the relatively permanent place where [radio operations] happen.” The “Buddhist” view, on the other hand, would see spectrum as the absence of factors that would obstruct radio operations. Existing radio operations, including receivers, would provide resistance to new operations, even in quite distant frequency bands. And there is an interaction between the agent that wants to move about and the nature of obstructions: neither a mouse nor a monkey would have no trouble scurrying around in a restaurant, while a person would be obstructed by all the tables and chairs. Likewise, one has to first define the new operation one has in mind before deciding that spectrum is “occupied”; calculating utilization is not a straightforward matter of marking spectrum as “empty” or “full.”

FCC white paper shows that “spectrum efficiency” is meaningless

2011-10-23T12:41:00.000-07:00

The FCC Technical Advisory Council’s (TAC) draft white paper on spectrum efficiency metrics (25 September 2011) is an excellent piece of work. It is authoritative, instructive, and demonstrates decisively that spectrum [1] efficiency metrics are a meaningless concept.

While they don’t say this in so many words, members of the Sharing Working Group perhaps intended this conclusion to be drawn; “spectrum efficiency” is a DC catchphrase that is hard to avoid, and it would probably be unwise to refute it overtly…

The following elements of the paper imply that the “spectrum efficiency” concept is useless:

There is no metric that can be applied across the myriad of different wireless services.
The metrics are incomplete, even within a service.
While the paper suggests metrics for specific services, the taxonomy of services is arbitrary.

Consequently

There is no way to compare the “efficiency” of one radio service (aka one “spectrum use”) to another, denying politicians the pseudo-scientific rationale they dream of for converting a frequency band allocation from one use to another.

Even within a given service type, there is no defensible way to rate one deployment’s performance over another; even if one scored much lower using the relevant efficiency metric, its defenders could invoke any of the long list of “additional efficiency considerations” to deny that the comparison was valid.

The paper also misses an opportunity: It hints at the importance of cost effectiveness rather than mere efficiency, but doesn’t address this broader context.

Follow-up post:

From spectrum efficiency metrics to parameter spaces

There is No One Metric

The draft white paper makes a key admission in the very first paragraph: “Unfortunately, as discussed in more detail herein, there is no single measure of spectrum efficiency that can be applied across the myriad of different services that rely upon wireless systems.”

To get a flavor of just how much how the metrics differ, here's what it came up with for its chosen service classes:

Satellite Broadcast: Information bits per second per Hz of allocated (licensed) spectrum within each common program area, e.g. bits / (second – Hz)

Satellite Point-to-Point: Information bits per second per Hz of allocated (licensed) spectrum per square mile of service area, e.g. bits / (second – Hz – sq. mi.)

Terrestrial Broadcast: Information bits per second per Hz of allocated (licensed) spectrum within each common geographic area, e.g. bits / (second – Hz), times the average number of users simultaneously served.

Terrestrial Personal Communication: Information bits per second per Hz of allocated (licensed) spectrum per geographic service area, e.g. bits / (second – Hz – sq. mi.)

Terrestrial Point-to-Point: Information bits per second per Hz of allocated (licensed) spectrum) x (transmitted distance) per square mile of service area

Terrestrial Hybrid: Information bits per second per Hz of allocated (licensed) spectrum over the geographic area served, e.g. bits / (second – Hz - sq. mi.), times the average number of users simultaneously served

An even more profound qualification emerged as the Working Paper declined to offer a metric for radar, concluding that “commonly applied efficiency measures (such as bps/Hz) are not appropriate for radars since the spectrum efficiency of a radar system cannot be directly compared to the spectrum efficiency of a typical communications system.” For this and other reasons, it was “unable to identify or evaluate suitable spectrum efficiency metrics for radar systems at this time.”

It declined to address spectrum efficiency metrics for “passive” (mostly scientific) uses, and short range systems that typically operate on an unlicensed or “licensed by rule” basis.

Bottom line: there is no way to compare different services with each other. This is a decisive weakness, since the advocates of spectrum efficiency metrics had hoped that that metrics would place arguments over the re-allocation of spectrum from one service to another (because one was supposedly “less efficient”) on a more “scientific” footing.

The Metrics are Incomplete

So, there is no one metric to rule them all – but can’t they be used to choose between operations within the service class for which they can be defined? No, because they are incomplete, even within a service.

The white paper admits that “additional efficiency considerations need to be evaluated in addition to the spectrum efficiency metric so that a comprehensive determination of […] system efficiency is properly made.” Since it focuses most of its attention on satellite systems, most of these considerations are introduced in that context; however, as one can see from the factors considered, they obviously apply to any system. The considerations include:

Device size: Larger antennas increase the spectrum efficiency metric, but decrease user value. As the paper says, “In terms of overall system optimization and user value, it is not always desirable to use the largest possible antenna sizes to achieve the greatest spectrum efficiency.” This applies more generally to device design; for example, increased spectral efficiency may come at the cost of higher power consumption, which requires a larger battery.

Response time: A more responsive system increases user value, but requires reserving a greater percentage of the spectrum (time or frequency) for the signaling associated with allocating access on an as-needed basis.

Quality of service (QoS): Improving QoS elements like latency/access time, coverage/reliability, information error rates, and peak-to-average loading ratios will reduce spectral efficiency metrics. Mission-critical public safety voice communications have higher QoS requirements in some dimensions compared to, say, consumer-grade mobile broadband internet access, meaning that the one cannot measure both merely using the recommended spectral efficiency metric.

System-level cost: The trade-off between spectrum and capital costs, of backhaul, towers, and user equipment, implies that maximizing spectral efficiency may not be the most cost-effective solution.

Even worse: not only does a comprehensive determination of efficiency require the consideration of “additional efficiency considerations,” but no guidance is given on how to weight them relative to each other. There is thus absolutely no rational basis, within this recommendation, for comparing one operation with another, even within a service class.

Division into Services is Arbitrary

Not only are the metrics are incomplete; the “services” within which they are defined are arbitrary categories, and thus extremely vulnerable legally and politically.

Of course, it goes without saying that any taxonomy is arbitrary. However, the white paper generously highlights and proves it in this case by citing the system classes proposed in an earlier report that it found to be important: “Definitions of Efficiency in Spectrum Use” (2008) by Working Group 1 of the Commerce Spectrum Management Advisory Committee (CSMAC):

Broadcast Systems
Personal Communications Systems
Point-to-Point Systems
Radar Systems
Satellite Systems
Passive Listeners (e.g., radio astronomy)
Short Range Systems

The TAC White Paper refines the CSMAC taxonomy by refining the distinction between satellite and terrestrial systems, and adding a “hybrid” class. While there are similarities, there isn’t a 1-1 mapping between the systems classifications

Satellite Broadcast
Satellite Point-to-Point
Terrestrial Broadcast
Terrestrial Personal Communication
Terrestrial Point-to-Point
Terrestrial Hybrid (combination of broadcast and point-to-point)
Radar
“Passive” (mostly scientific)
Short range

The losing operation in a spectrum efficiency shoot-out in a given service will thus not only argue that additional efficiency considerations prove that it’s doing a fine job after all, but also that it has been arbitrarily and unfairly lumped in with the winning operation. The loser will argue that it should be in a different service category, and advocate an even finer distinction that the one the TAC chose; and since the TAC had already refined the CSMAC’s taxonomy, there’s evidently no reason, in principle, to deny this claim.

Note

[1] In the white paper, the term “spectrum” seems to connote the combination of frequency, geography, and time – i.e. not just frequency alone – since it builds its analysis on ITU-R SM.1046-2, which defines Spectrum Utilization Efficiency as a function (undefined!) of the “useful effect” and the “spectrum utilization factor” of a system for that system, where the spectrum utilization factor is the product of the frequency bandwidth, the geometric (geographic) space, and the time denied to other potential users.

The extent of FCC/NTIA frequency sharing

2011-10-17T13:20:00.000-07:00

Take a guess: What percentage of US frequencies are controlled by the Federal government (represented by the NTIA), and what percentage is shared with non-Federal [1] users, who are under FCC jurisdiction? And what’s the remainder, devoted solely to non-Federal users?

My intuition, for what it's worth, was completely wrong. I thought the Fed/non-Fed split was roughly 50/50, with a bit (say 10%) being shared. As I pointed out in my recent post about partitioning Fed and non-Fed allocations, the amount of sharing should be easy enough to establish. It turns out that Peter Tenhula of Shared Spectrum Company has done a lot of work on this [2], and he pointed me to the FCC’s spectrum dashboard where one can download an XML snapshot of the allocation database (currently the API only covers the range 225-3700 MHz).

The answer? It depends on the frequency range and how one counts [3], but very roughly 10% is Federal, 40% shared, and 50% non-Federal (i.e. FCC) only.

Here's a picture (click it to enlarge); an Excel file with my analysis is here.

Update, 21 October 2011: Perhaps the flaw didn't lie with my intuition, but with my interpretation of the data. A senior FCC person has pointed out to me that many supposedly "shared" allocations are, to all intents and purposes, controlled by Federal agencies, and non-Federal (i.e. FCC-managed) services are present only on sufferance, if at all (e.g. 220-2290 MHz); or "sharing" only occurs both Federal and non-Federal entities used the same service (e.g. air traffic or maritime radar). So the question still stands, pending further digging: how much sharing (for various values of "sharing") is really going on?
The implications for my partition proposal is that there’s a lot more opportunity – or, seen another way, a lot more work to be done! – in untangling shared frequencies. The jurisdiction problem means that the intensity of use of frequencies shared by the FCC and NTIA is probably very low: since there’s no-one to arbitrate a fight between them, it’s better not to get into a fight in the first place. The best way to avoid conflict is to be over-conservative about protection rules for services sharing a band.

One might argue that the fact of such extensive shared jurisdiction means that partition isn’t necessary; it’s worked so far, so why change it? I’d respond that we haven’t really seen much real sharing, because utilization has been low, and guard bands large. It’s like a kid on rollerblades and a couple of seniors sharing the boardwalk at two o’clock in the morning; there’s so little traffic, they might as well each be there on their own. As everyone starts using radios for everything, we’re heading into an era analogous to the kid and the couple crowding onto the boardwalk along with thousands of others on a sunny Saturday evening.

The extensive sharing of Federal with non-Federal users also points to an incentive for both communities. Untangling use will give the Feds the capacity for bandwidth-intensive new applications like live streaming video, and will give the FCC the authority to optimize sharing between commercial players without “people will die if you do this” trump card always lurking up the NTIA’s sleeve.

Notes

[1] This is sometimes referred to as “commercial use spectrum,” but that’s a misnomer since the non-Federal frequencies managed by the FCC includes public safety uses.

[2] Peter has compiled an excellent list of sources:

Karl Nebbia, Director, NTIA Office of Spectrum Management, presentation to the Commerce Spectrum Management Advisory Committee (CSMAC), December 9, 2009, available at http://www.ntia.doc.gov/advisory/spectrum/meeting_files/225_3700MHzPresentation.pptx.

Lawrence E. Strickling, Assistant Secretary of Commerce for Communications and Information, keynote speech delivered at the 11th Annual International Symposium on Advanced Radio Technologies (ISART), Boulder, Colorado, July 28, 2010, available at http://www.ntia.doc.gov/presentations/2010/ISARTkeynote_07292010.html.

Spectrum Policy for the 21st Century – The President’s Spectrum Policy Initiative: Report 1 Recommendations of the Federal Government Spectrum Task Force and Spectrum Policy for the 21st Century, National Telecommunications and Information Administration, U.S. Department of Commerce (June 2004), available at http://www.ntia.doc.gov/reports/specpolini/presspecpolini_report1_06242004.htm#_Toc74447274.

Press Release: U.S. Department of Commerce Takes Major Step Towards Unleashing the Wireless Broadband Revolution, available at http://www.ntia.doc.gov/press/2010/SpectrumReports_11152010.html.

Factsheet, available at http://www.ntia.doc.gov/reports/2010/SpectrumFactSheet_11152010.pdf

P.C. Roosa, Jr., Basic Elements of Spectrum Management: How the Spectrum is Shared, August 1992, available at http://www.ntia.doc.gov/osmhome/roosa5.html.

[3] A hertz isn’t a hertz isn’t a hertz; lower frequencies are usually valued more highly, since they offer better propagation and thus coverage. Further, 500 MHz covers pretty much the entire UHF TV band of 300-700 MHz, whereas it’s only a fraction of the 60 GHz band (57-64 GHz, i.e. 7,000 MHz) that’s about to be opened up for multi-gigabit wireless systems. It might thus make sense to weight a hertz at low frequencies more highly than one at high frequencies. I use a "frequency-adjusted bandwidth" of bandwidth divided by center frequency, as used in the 2005 Ofcom Spectrum Framework Review (PDF), "a 1MHz allocation at 100MHz is given equal weighting to a 10MHz allocation at 1GHz". I multiply the ratio by 1,000 for ease of interpretation, so that 1 MHz at 1 GHz counts at 1, not 0.001. For frequencies lower than 1 GHz, a MHz is "counts for more," and “counts for less" above 1 GHz. So for example, 10 MHz at 500 MHz counts for 20 = 1000*10/500, whereas at 5 GHz it counts for only 2 = 1000*10/5000. This is equivalent to plotting the bands on a logarithmic scale, as is done in the allocation chart.

Partition, not sharing: An alternative approach to the Fed/non-Fed spectrum divide

2011-10-10T11:40:00.000-07:00

South Sudan. Serbia/Kosovo. India/Pakistan. Britney Spears and Kevin Federline. Sometimes a clean break is best for everyone, particularly when there are fundamental differences in mindset. Enforced coexistence is not, for many couples, the best way to live.

Sharing between Federal and non-Federal wireless users (aka Fed and non-Fed) is a favored way to realize the FCC’s dream of finding 500 MHz for commercial mobile broadband services; as reported in TheHill.com, “It is unclear where the 500 megahertz of spectrum will come from, but a large portion will likely come from government agencies that do not use the frequencies efficiently.”

Fed/non-Fed sharing can be made to work, and worthy efforts are being made. However, I doubt it’s worth the effort, given the insane difficulty of negotiating band re-allocations, let alone sharing agreements; questions over whether 500 MHz is, in fact, either needed or would make a dent on cellular companies’ problems; and fundamental concerns about jurisdiction (see my August 2011 post No Common Authority: Why spectrum sharing across the Fed/non-Fed boundary is a bad idea).

It would be a better use of time and effort to go in the opposite direction: make the partition between Federal and non-Federal as clean as possible, and let each group of figure out sharing among its own constituents.

The difficulty of re-allocation and sharing is not only a matter of institutional recalcitrance or missing incentives; it’s that the same technologies and scenarios – and vendors! – that are driving the demand for commercial wireless broadband also apply to Federal uses. In addition, some applications of particular interest to government are very bandwidth-hungry. The US Army is buying 1,300 RQ-11B Raven aerial surveillance drones per year (The Economist). These back-packable snap-together drones weigh only two kilos and deliver real-time streaming video, which chews up bandwidth. If such devices were cleared for use in domestic airspace, Federal agencies would line up deploy them – as would state and local public safety departments.

And even if a few dimes-worth of “unused spectrum” were found between the Feds’ sofa cushions, swapping or sharing a tens or even hundreds of MHz between Federal and non-Federal jurisdictions is just not going to be worth the pain. Experience suggests that such re-allocation attempts are contentious, time-consuming, and often fruitless.

Hence, partition: make the Fed/non-Fed divide clearer, not murkier, and let each side get on with improving the value of radio operations in ways that make most sense in its particular institutional context. Three steps come to mind:

The first task is to minimize sharing between Federal and non-Federal services in the same band, since the lack of a common authority complicates dispute resolution as well as negotiation, particularly since the goals and rewards of Fed and non-Fed users are quite distinct (except for state and local public safety agencies which fall under the FCC's "non-Fed" jurisdiction, but are much more similar to Federal operations - more on this below). This means reducing the number of bands where jurisdiction is shared between the NTIA (nominally the manager of Federal radio operations) and FCC. Reducing, not eliminating; some services such as GPS will always be shared, since Fed and non-Fed users depend on the same infrastructure. And in some cases, sharing has been recently imposed and can’t be rolled back again, e.g. Wi-Fi systems coexisting with weather and military radars at 5GHz.

Second, minimizing sharing also requires reducing the number of Fed/non-Fed band boundaries. Radio systems in adjacent bands affect each other because real-world receivers cannot reject all the energy radiated outside their allocated frequencies. This amounts to defragging the US frequency allocation chart. The required band swaps won’t be easy, but should be easier than transferring a band from Federal to non-Federal use with nothing in return. Since the swaps would be local in frequency, the cost of hardware change should be manageable. Some swaps won’t be possible; just as some files on a hard disk can’t be moved, some allocations are fixed by physics (e.g. frequencies used for radio astronomy) or treaty (e.g. agreements reached at the World Radio Conference).

Third, and most aggressive, is the prospect of moving state and local government allocations out of the FCC’s jurisdiction and over to the NTIA. I expect this will require legislation (if it is even possible, constitutionally…) but the similarity of goals, incentives and culture among Federal and state & local users should make for much more harmonious (or at least less acrimonious…) coordination of interests. The failure to share the 700 MHz D (as in “Doomed”) Block between public safety and commercial users demonstrates the difficulties of cross-culture coordination even within the FCC's remit, and is just a taste of the deadlocks that will be the hallmark of attempts to foster extensive Fed/non-Fed sharing.

It’s an empirical question how much sharing occurs in the current band plan, and whether it’s increased or decreased. Appendix B in the September 2011 draft report on Spectrum Efficiency Metrics of the FCC Technological Advisory lists some examples of spectrum sharing, many but not all of them Fed/non-Fed; a more exhaustive inventory probably exists, but I haven’t found it yet. More to come...

This partition proposal does not preclude sharing; in fact, it gives it the best possible chance to succeed. Organizations that are all tasked to work for the greater public good are more likely to develop the trust required for effective wireless sharing among themselves, and likewise, organizations that are all measured by commercial metrics are more likely to find a mutually beneficial sharing solution. Federal and non-Federal users will continue to share some bands, and rub shoulders in others, but fewer fences will make for better neighbors.

Licensed Unlicensed: Having your Coase, and your Commons too

2011-09-29T12:48:00.000-07:00

I lighted on the notion of issuing a handful of receiver licenses in allocations where transmitter licensees don’t control receivers (e.g. TV, GPS) to facilitate negotiations between operators in neighboring bands; details blogged here.

The same idea could be applied to unlicensed allocations, where the unbounded number of operators makes it essentially impossible for Coasian adjustments to be made: a neighbor that would like quieter unlicensed devices has nobody to make a deal with, nor do unlicensed users have an effective way to band together to make a deal if they’d like to increase their own transmit power. This approach also has the benefit, as in the receiver license case, of giving the regulator a tool for changing operating expectations over time, e.g. ratcheting down receiver protections or increasing receiver standards.

The catch-phrase “licensed unlicensed” is obviously a contradiction in terms; it’s shorthand for a regime where non-exclusive operating permissions are issued to a limited number of entities, while retaining the key characteristic that has made unlicensed successful: the ability of end users to choose for themselves what equipment to buy and deploy. These entities can use or sub-license these authorizations to build and/or sell devices to end-users.

Follow-up post

Stamps and stampholders: a third way to regulate radio operation

How it would work

These “unlicensed licenses” would have the same technical rules as unlicensed or license-exempt currently does, e.g. the Part 15 conditions that one may not cause interference to a primary, and must accept interference from anyone else. However, the ownership and transferability would differ: only license holders would be able to seek device certification, though that right could be sub-licensed to other parties so that any number of device designs could be sold by any number of manufacturers.

The regulator could assign these licenses as it chooses, e.g. by auction, lottery, beauty contest, etc.; the current received wisdom would suggest an auction. There would be no limitation on who might obtain them, but I suspect that they would be most sought-after by a handful of chipset companies. For example, while there are dozens (hundreds?) of companies selling Wi-Fi devices, they all pretty much use chips from Atheros-Qualcomm, Broadcom, Marvell, or TI.

This seems so obvious that someone must already have thought of it; and the fact that it hasn’t been done means there were flaws in the idea or the politics. Please let me know if this rings any bells.

Some benefits

This idea has an interesting side effect: it’s a plausible way to raise revenue from unlicensed, i.e. by an auction of (say) six manufacturer certifications. (Standard disclaimer: The purpose of auctions is to identify those who place the highest value on the operating right, not to raise money. Politician’s standard response: Yeah, right.) While I’m also contemplating an auction, this idea differs from the Bykowsky, Olson & Sharkey proposal that companies interested in unlicensed should bid against those interested in licenses; in my case, the allocation to unlicensed has already been decided.

The limited number of parties with skin in the game should make it easier to negotiate coexistence etiquettes. Advocates of unlicensed sharing often invoke standards processes as the way in which coexistence is delivered. However, 802.11 is becoming increasingly unwieldy, I’m told, and the 802.19 coexistence TAG hasn’t made much progress as far as I can tell. A regulatory limit on the number of parties to this negotiation would improve the chances of agreeing to win-win etiquettes.

Some questions

Auction theory questions arise. The dynamics are different from previous wireless auctions, since it’s not an exclusive license; rather, it’s an equal sharing of the “resource” among a limited number of players. Presumably an auction of five co-equal licenses with the same rights will raise less revenue than an auction of an exclusive right; but how much less? Any pointers gratefully received. I’m struggling to think of an analogy where a right to operate an indeterminate number of devices is at issue; neither airline landing slots nor taxicab medallions fit. The best analogy I have so far is seats on the NYSE, thanks to a conversation with John Helm; trouble is, not only has the NYSE now moved to selling one-year licenses to trade directly on the exchange, but the number of seats were very large (Wikipedia).There must be precedents, surely, by the “If I can think of it, someone’s already done it” rule.

There are also questions of competition policy and transaction costs. Just as in the receiver license case, one would want to allocate enough licenses to foster competition and innovation, but not so many that cost-effective negotiations is precluded. My impression is that the literature doesn’t give much if any guidance on picking a number; the answer is “it depends” – one would have to analyze the wireless situation and connect it to the key factors that literature has highlighted.

It’s interesting to compare this with Martin Weiss’s neat idea of “free licensed” (cf. unlicensed = license free). Martin’s only briefly mentioned it to me (9/24/2011 at TPRC), and he’s still developing the concept. As I understand it, the idea is to retain the regulatory leverage of licenses while leaving the manufacture and deployment of devices unconstrained. It’s another way to get the putative benefits of fixed license terms (cf. an earlier post) without giving up the decentralized and concurrent nature of “unlicensed.” I think the FCC’s 3650-3700 MHz light license could be seen as a version of this idea, though the licenses there are of indefinite term. Martin’s Free Licensed differs from Licensed Unlicensed in that I’m proposing a limited number of non-free licenses, whereas Martin’s talking about an unlimited number of free licenses. I suspect Free Licensed would work best for small numbers of relatively high power unlicensed operators in a region (e.g. WISPs – the WAN case), whereas Licensed Unlicensed is better suited to large numbers of low power devices (e.g. PAN/LAN).

The obvious question is, Where could one try this out? One possibility is the 3550-3650 MHz band (cf. Motorola filing), but I think this restriction added on top of the requirement for operators to hold a non-exclusive nationwide license and register their fixed stations may be too onerous. While I’m no fan of unlicensed secondaries “sharing” with licensed primaries, I’d propose this the next time such an opportunity arises as a way of mitigating the primaries’ worries about accountability.

TV white space databases: A bad idea for developing countries

2011-09-02T11:34:00.000-07:00

Now that TV white space rulemakings are in the can in the US and UK, proponents will be pitching the technology to any government that’ll listen, e.g. at the Internet Governance Forum meeting to be held in Nairobi on 27-30 September.

It’s understandable: the more widespread white space database rules, the larger device volumes will be, and thus the lower the equipment cost, leading to wider adoption – a positive feedback loop. However, white space database technology is unnecessary in many countries, particularly developing ones.

Yet it verges on dodgy ethics for companies to hype this technology to countries that don’t need it, particularly since there’s a better solution: dedicating part of the TV frequencies that are freed as a result of the transition to digital TV (the “Digital Dividend”) to unlicensed operation, without the white space bells and whistles.

White space operation is only necessary where the TV frequency bands are filled with many TV stations with many gaps between them. In many countries there are relatively few terrestrial broadcast stations, because there isn’t the capacity to generate many channels of content, and/or because large geographies are more efficiently covered by satellite transmission than a network ground-based towers. Countries that have waited to make the transition to digital TV can also easily opt for technology also allows a TV transmission to use the same frequency on all towers (Single Frequency Networks), allowing the same number of TV programs to be broadcast using fewer channels. Ensuring that all transmitters in an area use the same tower, and other techniques being discussed as part of, would allow broadcasts to be arranged with fewer white spaces, leading to an even larger digital dividend of freed up TV frequencies that can be used for data networks.

It’s worth remembering that the database approach was adopted only as Plan C, after Plan A (a single, contiguous band dedicated unlicensed) and Plan B (each device using its own sensors to determine vacant channels) had failed. Implementing a database solution requires not only collecting and managing data about TV (and other primary) transmitters in a central location, but also managing the on-line service itself. That’s a heavy overhead in places where limited institutional capacity could be better used in other ways. White space databases are also more liable to be used for extortion and excessive social control than a conventional unlicensed band.

The Better Alternative

Places with few broadcast TV channels should forget about white spaces, particularly the database flavor, [1] and simply divide their Digital Dividend into three chunks: one third for licensed, one third for unlicensed, and one third held in reserve for later use (e.g. licensed or unlicensed, depending on which one turned out to be more useful in a particular country).

Countries like South Africa are in the midst of deciding what to do with their Digital Dividend; the cellular companies are hungry, and unlicensed isn’t discussed much if at all. Decisions made now will have long-lasting consequences.

- - - - - -

[1] If a country has remaining white spaces where it would like to authorize unlicensed use, a country should consider using spectrum sensing techniques. A database is, in a way, just a way to sense spectrum, i.e. for a device to know which channels are not being used for TV. Device-based spectrum sensing proved less accurate than databases in FCC testing, but other countries might consider this an acceptable price to pay in their circumstances.

Spectrum “sharing”: the convenient ambiguity of an English verb

2011-08-29T01:20:00.000-07:00

I realized while writing Spectrum Sharing: Not really sharing, and not just spectrum that my confusion over the meaning of spectrum sharing derives from two meanings of the English verb "to share":

(1) to divide and distribute in shares, to apportion;

(2) to use, experience or occupy with others, to have in common.

For example, the first is sharing a bag of peanuts, and the second is sharing a kitchen or an MP3 file. Cellular operators and economists tend to use the word with the first meaning, and Open Spectrum advocates with the second.

But that raises the question: is the double meaning inherent in the concept, or is it just an accident of English vocabulary?

I asked some friends about the regulatory terminology in other languages; so far I have information about Arabic, Chinese and German. If you could shed light on regulatory terminology in other languages, for example French, Japanese or Spanish, please get in touch.

I learned from Dr. Fadel Digham, R&D Director of the National Telecom Regulatory Authority (NTRA) in Egypt, and Mr. Mohamed El-Moghazi, Senior Manager, Spectrum Research & Studies at NTRA, that there are two different words/definitions in Arabic:

mokasama (مقاسمة): distributing things in non-overlapping shares given fixed boundaries. This is the term used in Egyptian regulation when dealing with spectrum; although the exact word is not used, the definition itself is included as the basis of assignment. Fadel Digham observes that in the networking literature, the term time-sharing is used to indicate sharing across time (not simultaneously sharing some resource): TDMA. Likewise, frequency-sharing is synonymous with FDMA. So, both time-sharing and frequency sharing in network and communications literature belong to this denotation.

mosharaka (مشاركة): having common access to a certain resource/utility, e.g. sharing a room. The term is used by NTRA to denote site sharing (the same site used by different operators to install their equipment), duct sharing (laying down different cables in the same duct), and the like. It is not used in the spectrum context; if it were used there, it would imply either concurrent access, or a common pool of frequencies (perhaps something like Wi-Fi).

A Chinese-speaking wireless researcher informs me that there are two popular translations for sharing:

fen(1) xiang(3) (分享): usually used for sharing a file on the internet, for sharing your food, etc. It means sharing a mutual space/item without defining a clear boundary/allocation.

fen(1) pei(4) (分配): allocation; usually used for spectrum, meaning distributing spectrum for exclusive usage across service providers.

This colleague reports that Chinese regulators use the second word (fen(1) pei(4) 分配, exclusive usage) even for the 700Mhz spectrum they plan to release by 2015. Chinese translations of “sharing” always use 分配.

Courtesy of Petri Mähönen at RWTH in Aachen and a German-speaking colleague, I understand that German regulators have a number of different terms for "sharing":

Zuteilen: to slice and point out who retrieves ownership, i.e. "allotment" (see below).

Aufteilen: just 'slicing' spectrum, but does not necessarily imply anything about the ownership of the "slice".

Verteilen: the act of handing objects to a group of people, in English closest would be to "distribute", or "to share".

Teilen: slightly different from verteilen, with the connotation of a lack of hierarchy, a bit of "socialism". In this case there is usually no third party involved that is a decision maker, as when the kids share candies among themselves. (If mother needs to do it, it is more like verteilen.)

Anteilige Nutzung: using a part of the bigger thing, a kind of shared commons use.

Exploring the terminology of the German telecommunications act, I gather that the first step is Frequenzbereichszuweisung, which is the allocation of allowed types of services for a frequency band. Then follows Frequenzzuteilung – here we see the zuteilen part – which is assigning spectrum to a specific use and specific operator. The verteilen is done in the Frequenznutzungsplan, which I understand to be the designation of coexisting uses in a band, e.g. TV transmitters and wireless microphones. A frequency band under Allgemeinzuteilung (allotment to general use, similar to Part 15 in the U.S.) does not need a license. In short, German usage makes a distinction in the terminology of sharing between apportionment (zuteilung) and in-common use (verteilung).

Overall, it seems that the English ambiguity between sharing-apportion and sharing-in-common is unusual, and not seen in some other important regulatory nomenclatures.

Time limiting unlicensed authorizations

2011-08-23T06:40:00.000-07:00

I’m coming around to Tom Hazlett’s view that unlicensed devices in the TV whitespaces are a bad idea because they preclude alternative future uses for the channels now being used for TV. (For his main objections, see “Shooting Blanks on Wireless Policy,” FT.com October 5, 2010 PDF) It’s a figure-ground problem; defining whitespace operating rules on the basis of TV operations reciprocally defines viable operations in the TV “blackspace”.

One could get around the problem and still have unlicensed use, though, by time limiting the unlicensed authorization. [1] Just like build-out conditions on licenses, there would be a fixed time window within which widespread deployment should occur. If it doesn’t, the authorization is revoked.

This approach seems particularly relevant when an authorization holds great promise, but that promise is very uncertain, e.g. when the technology or the market is changing rapidly. “Sunsets” on rules are important since the passage of time invariably invalidates the premises of regulation, even as it entrenches the interests that coalesce around those regulations. [2]

The period of the authorization should be long enough to allow entrepreneurs to make a return on their investment, but short enough that an under-used authorization doesn’t lie fallow for decades. Fifteen years sounds about right. At the end of that period, the authorization would either (1) expire unless renewed, or (2) be automatically renewed unless canceled; even if renewed, the regulator could choose to make changes to operating parameters. I prefer option (1) since it would force proponents to make the case for the value of the authorization.

The criteria for renewal would be set at the time of initial authorization; they would measure the success of the authorization, and would probably include the installed base of devices and the growth rate. The renewal assessment should include not just the value of the use at that time, but also the potential value of an alternative use. Overall, the regulator would make a judgment about the success of its “investment” in this authorization just as a company or financier might judge a start-up they had funded; just like investors, the regulator is likely to be subject to confirmation bias, but at least it will be forced to revisit its decision in the light of new information.

The regulator would be required to make its determination some number of years (say five) before the authorization expires; if the authorization is not to be renewed, sales of equipment would have to cease at this point, e.g. five years before the authorization expires.

If the regulator decides that the unlicensed authorization has failed, it has a number of choices: it could retain the unlicensed designation but change the operating rules dramatically, it could redesignate the band for licensed use, or it could start thinking through the allocation of band again from scratch.

The two obvious historical test cases for such an approach would be the 2.4 GHz ISM allocation, where there was a meteoric take-off in use, and the 1920-1930 MHz Unlicensed PCS (UPCS) allocation, which has languished. Installed base numbers are always hard to find, but the FCC’s Equipment Authorization database should provide some indication, since it lists the devices that were authorized for sale. For example, UPCS saw 36 original grants over the period 3/20/1996 to 9/2/2011, whereas there were 358 grants in the 2400-2483.5 MHz ISM bands over the period 5/10/1994 to 8/11/2011. Investigation of the adoption rates in these two bands should provide a basis for choosing suitable allocation periods. It seems pretty clear from this chart that a decade is enough time to see whether an unlicensed authorization in a band is successful or not.

Notes

[1] Technically, users of Part 15 devices do not have any allocation status in the Commission’s rules; they are simply authorized to operate in terms of Part 15 rules. Instead, the Commission makes spectrum available for Part 15 devices on an unprotected and non-interference basis. See e.g. footnote 11 to para 5, Notice of Proposed Rulemaking in WT Docket No. 06-49, March 2006. I’ll use the terms “allocation” and “authorization” interchangeably here.

[2] See e.g. my “Internet Governance as Forestry: Deriving Policy Principles from Managed Complex Adaptive Systems” (2008), arguing that the more detailed a regulation, the more likely it is to become obsolete in the face of social, technological and commercial innovation; consequently, the more detailed the rules, the more rapidly regulations should expire (or be revisited).

Licensing radio receivers as a way to facilitate negotiation about interference

2011-08-17T11:15:00.000-07:00

It’s a curious fact that, while receivers are just as much responsible for breakdowns in radio operations as transmitters [a], regulation is aimed pretty much exclusively at transmitters [b].

Since one can’t ignore the receivers in practice, arguments over interference almost invariably turn to receiver standards. Even if receiver standards were a good idea (and I don’t think they are - see my post Receiver protection limits: a better way to manage interference than receiver standards), the ability to adjust receiver performance by fiat or negotiation is limited when receivers are operated independently of transmitters.

I suspect that receiver licenses may be necessary to reach the optimum outcome in at least some cases. This post is going to take that idea out for a first test drive.

Regulators evidently have managed without receiver licenses (beyond their use as a way to fund traditional broadcasting) so far. Why introduce them now? I’ll give my usual answer: the dramatically increased demand for wireless is squeezing radio operators of widely varying kinds together to an unprecedented extent, and we no longer have the luxury of the wide gaps that allowed regulators to ignore receiver performance, and ways of managing it.

Follow-up post

Stamps and stampholders: a third way to regulate radio operation

The Problem

I am going to focus on the case where (1) a more efficient economic outcome can be achieved by negotiating a change in receiver performance, and (2) receivers operate independently of transmitters. [c]

Here are three examples where efficient negotiations are impossible today, and where the assignment of a limited number of receiver licensees would give adjacent operators someone to talk to about finding the economic optimum solution:

1. LightSquared would like to build a terrestrial cellular network adjacent to GPS frequencies, but GPS receivers have been built on the assumption that this band is quiet; since anyone can build, sell or buy a GPS receiver, there is an essentially unlimited numbers of parties LightSquared has to negotiate with

2. US regulators are thinking about delivering shifting weather information off a satellite band, freeing it up for other uses. However, no-one knows how many unlicensed satellite weather receivers have been deployed, let alone how to negotiate with them about relocation.

3. If TV sets had better receivers, whitespace devices operating next to them could use higher power, and provide better service. However, there are no standards for TV receivers, and nobody for whitespace device manufacturers or service providers to negotiate with –TV broadcasters don’t control the receivers, and nobody can sign an agreement on behalf of all the receiver manufacturers

The Receiver Licensing Solution

The proposal here is to require receivers to be authorized, just like transmitters are. Licenses would be issued, either by assignment or by rule.

I think licenses would be expressed most effectively in terms of reception protection limits that would be afforded to receivers, although regulators could choose to condition licenses on performance standards. Devices wouldn’t have to operate to these protection limits; manufacturers could choose to build cheap receivers that would fail even when third party signals were below these limits. However, receivers wouldn’t be afforded any protection in such a case.

If an operator in an adjacent band wanted to increase their power above the protection limit, it would then have to compensate the receiver licensee.

The regulator should issue a limited number of licenses: many enough to ensure competition and innovation, and few enough to make negotiation possible by limiting transaction costs. The literature on what makes for a competitive market is exceedingly complicated, and I’m not aware of work that has calculated the number of parties above which negotiation costs become prohibitive; my guess is that half a dozen licenses would probably be sufficient to avoid market concentration problems, and small enough to keep transaction costs low.

Receiver rights (licenses) could be assigned via auction; this would ensure that those who had them valued them most highly. For communications systems like cellular, transmission and reception rights would typically be kept together in a bundle; for broadcast systems like GPS, one would probably typically separate them.

Ideally the regulator should be neutral about who bid for receiver licenses. Thinking about the equipment value chain (chip manufacturer – equipment manufacturer – vendor – operator) helps predict where ownership of reception licenses might end up. The nexus with the fewest players is probably the most efficient place to hold the licenses; thus, if there were 4-8 reception licenses, they’d probably be bought by chip manufacturers.

(Auctioning receiver licenses to manufacturers can be contrasted with the zombie concept of unlicensed band managers or private commons, i.e. that one or more companies like Google and/or Microsoft would buy a license and turn the band over to unlicensed, covering the cost by charging per-device fees to manufacturers selling equipment for this band. I’ve seen no indication that any company is willing to try this; it’s remained a pipe dream of some activists and regulators. Moving the payment up the value chain is more likely to succeed, since manufacturers make money directly; and having a limited number of licenses – or in this case, type approvals – will spread the risk. In fact, doing it this way might (just might) make private commons practicable.)

I was led to think about reception licenses because it enables negotiation when receivers aren’t tied to transmission licenses. However, it also has the benefit of providing a mechanism to implement a “ratchet” in reception protections or receiver standards, i.e. a way to manage change in the conditions under which receivers have to operate. Reception licenses would have fixed terms just like transmission licenses do, and the conditions associated with them could be changed at successive license renewals.

Objections

Receivers are not licensed today (as far as I know), presumably for good reasons. There are surely also not-so-good reasons, e.g. “this is the way we’ve always done things” and “we’ve never needed them before.”

The most serious objection, I believe, is that any new regulation always has unpredicted, unwanted side effects. However, risk is unavoidable if a solution is required to a new problem. A cost-benefit trade-off needs to be done; I don’t know where it would end up.

Some may argue that reception licenses would be a restraint on public discourse: by controlling receiver licenses, a government could control who could listen to broadcasts. I don’t see this as a problem: efficiency requires few licenses, which means that end users will not be subject to licensing, and thus control. And in any case, the government can already control the transmission of broadcast information through existing licenses if it wished.

There will be claims that regulators such as the FCC do not have sufficient statutory authority to issue receiver licenses. This may or may not be so; a colleague with a legal background is helping me research this question. Stay tuned.

There will be the usual arguments from Left and Right that such licenses will Harm Innovation and the Consumer by reducing number of competitors, or by increasing the amount of regulation, respectively. This shouldn’t be a problem if the number of licenses is large enough to protect competition in the value chain segment where they are issued, e.g. to manufacturers; and a great deal of competition and innovation will continue unfettered in other parts of the chain, e.g. in end user applications. If licenses use reception protection limits rather than receiver performance standards, the terms will be broad enough to leave a great deal of discretion in the hands of both operators and manufacturers.

Conclusion

I don’t know if reception licenses are worth introducing. However, I’m surprised that it hasn’t attracted more attention. It may be an idea whose time has come.

Notes

[a] To see why responsibility is shared, imagine the case of a two systems in adjacent frequency bands that are coexisting successfully. If one then changes the receivers in band A so that they can no longer filter out the signals of transmissions in adjacent channel B, the service in band A will be degraded even though the transmissions that “cause” the interference haven’t changed.

[b] The burden on transmitters is exemplified by the international definition of interference in Article 1 of the ITU Radio Regulations, para 1.166: “The effect of unwanted energy due to one or a combination of emissions, radiations, or inductions upon reception in a radiocommunication system, manifested by any performance degradation, misinterpretation, or loss of information which could be extracted in the absence of such unwanted energy.” Note that only effects due to transmission are counted, not those due to poor receivers. Licensing appears to be conditioned on transmission, though receivers are mentioned in the chapter on licenses (Article 18).

[c] Cases where receivers are not controlled by transmitter licensees include television, GPS navigation devices, and weather satellite data receivers (EMWIN). These are usually broadcast systems; the operators of communication systems usually control both transmitters and receivers, since any comms device necessarily includes both. (In some broadcast systems the receivers are controlled by the licensee, e.g. satellite audio and video services like SiriusXM, Echostar and DirectTV.)

It’s not quite so black-and-white, though: in Article 15 the list of interference avoidance techniques includes the choice of receivers (para 15.6), and Article 3 requires that “all technically and economically justifiable measures have been taken . . . to reduce the susceptibility to interference of the … receiving equipment” (para 3.3). Still, the detailed rules and requirements deal exclusively with transmitting stations.

The dark side of whitespace databases

2011-08-09T09:36:00.000-07:00

Back in May 2009 I drafted a blog about the unintended side-effects of regulating unlicensed radios using databases. I was in the thick of TV whitespace proceeding (on the side of the proponents), and decided not to post it since it might have muddied the waters for my client.

Databases have become the Great White Hope of “dynamic spectrum access” over the last two-plus years. They are seen not only as a way to compensate for the weaknesses of “spectrum sensing” solutions, but as a way for regulators to change the rules quickly, and for unlicensed devices to work together more efficiently. For a quick background, see: FCC names nine white-space database providers, FierceWireless, Jan 2011; Michael Calabrese, “Ending Spectrum Scarcity: Building on the TV Bands Database to Access Unused Public Airwaves,” New America Foundation, Wireless Future Working Paper #25 (June 2009).

Looking back at my note, I think it's still valid. Rather than rewrite it, I’ve decided simply to repost it here as originally drafted (omitting a couple of introductory paragraphs).

The benefits

I expect the database approach to be a big hit with regulators because it is a powerful technique that seems to solve pressing problems. The fairy dust of “cloud computing” and “smart radios” also helps. However, in practice it may prove to be rather less of a panacea than expected.

The attraction of the database approach boils down to control. If the regulator finds that a rule doesn’t work well – for example, the exclusion zone around some protected service is the wrong size – it can be changed literally overnight by changing how the database responds to requests for permission to operate. Depending on the information supplied to and/or provided by the database, a regulator could reduce a maximum allowed power level or deny permission for a device to operate. In the past, rule changes could take years to implement, because once devices were deployed, their behavior was fixed; rule changes could only apply to new devices, and they would only gradually replace the old ones.

Consequently, regulators don’t have to start with most conservative assumptions about operating parameters needed to protect incumbent services like TV. They can start with more reasonable values, and change them later if a problem appears.

The limitations

(i) Commercial uncertainty

This all sounds like good news for innovators, too, since the most restrictive assumptions might lead to rules so tough that usable devices could not be built. On the other hand, however, a regulator can easily change its mind about rules, which could cripple an industry overnight if the result is that devices are no longer usable.

(ii) Unequal burdens

Televisions can be adversely affected by bursts of electromagnetic radiation when home appliances switch on. However, these old devices can’t be controlled – but the smart new ones can. The new generation of smart devices will paradoxically bear a heavier regulatory burden than the old, dumb ones. Before the days of fine-grained control promised by database lookup, all participants had to accept that protection against interference isn’t absolute; there has always been spurious interference from electrical appliances switching on and off, but it’s been ignored because there was no way of tracking and managing it. Striving for perfection in the protection afforded by databases will be a disincentive for device innovation.

(iii) Fragile complexity

A suitable database would also allow a regulator to prescribe device behavior for an unlimited number of edge cases. Since it’s simply a “small matter of programming”, there’s no limit to the sophistication of the possible rules. In practice, of course, there is a heavy price to pay for each increment in sophistication; and the price increases exponentially. As rules get more complicated, the possibility for unexpected rule conflicts increases, let alone the number of software bugs. The old rule of thumb in software design applies: “Just because you can, doesn’t mean you should.” (For more on applying the lessons of managing complex ecosystems to internet policy, see my "The Resilience Principles: A Framework for New ICT Governance," 9 J. on Telecomm. and High Tech. L. 137 PDF.)

Changes to database parameters will come with clarion calls for “improving spectrum efficiency”. However, efficient systems are fragile ones. The set of parameters that generates maximum output from an engineering or economic system usually places it at the edge of chaos; collapse is only a small step away. Changing the rules too often or in too many ways will lead to unexpected and probably undesired outcomes.

(iv) Temptation to meddle

Database solutions start a regulator down the slippery slope of detailed management of operations in a band. That’s OK for a licensee that can extract rents from their efforts, but goes against the hands-off philosophy being developed by regulators like the FCC in the US, and Ofcom in the UK. For a regulator looking for a light touch approach to unlicensed operation, overly sophisticated database techniques will be a tar baby.

Changing the parameters in a database amounts to a rule change. In some jurisdictions this increased autonomy for the agency could amount to changing rules without due process.

(v) Privacy

Another selling point is that a database regime gives regulators the power to prohibit the operation of devices – categories, certain models, or even individual ones – on the basis of identifiers that have to be provided before obtaining operating permission from the database. This is useful if some device(s) turn out to operate in unacceptable ways. However, this benefit has to be set against the risks of collecting detailed data. Information on where and how device classes are operating would allow a database operator to aggregate valuable competitive information, and database queries that require devices to provide unique serial numbers and their location allow fine grain tracking that has privacy implications.

Remedies

When a database is intended to support on-the-fly adjustments to operating rules (not the case with the proposed white space rules in the US, but it could happen in the UK), it will be important to establish principles to guide behavior before the database comes into operation.

Since adding complexity can lead to unexpected interactions between database rules, a riff on Einstein’s famous maxim applies: database rules should be as complicated as necessary, but no more so.

The regulator should give innovators the certainty they need to make upfront technology development investments by establishing floors and ceilings on parameter values.

When choosing parameters, robustness of operations should be taken into consideration in addition to efficiency of use of radio operating resources.

Operating parameters should be changed with care and after due process. The procedure for deciding whether and how to change parameters should be established in advance. Changes to database parameters should not be driven by anecdote except in egregious cases.

There should be a high hurdle to adding new rules, and the hurdle should get higher as more rules are added. Every new “feature” of a database regime, such as a clever way to protect against some newly discovered, rare interference case, carries a hidden cost of increased system fragility.

No Common Authority: Why spectrum sharing across the Fed/non-Fed boundary is a bad idea

2011-08-04T13:03:00.000-07:00

The ISART conference this year was about sharing in the radar bands, in line with the Administration’s efforts to encourage frequency sharing between Federal and non-Federal (e.g. commercial and civilian) users (NTIA Fast Track Evaluation PDF, FCC proceeding ET docket 10-123).

While it’s true that the NTIA has studied the feasibility of reallocating Federal Government spectrum, or relocating Federal Government systems, the current political focus is on “spectrum sharing” (cf. my post Spectrum Sharing: Not really sharing, and not just spectrum) – and Federal/non-Federal sharing is the hardest possible problem.

Federal/non-Federal sharing is hard for many reasons, notably the chasm between the goals and incentives between the two groups, and thus a profound lack of trust. I’m going to focus here, though, on a seemingly technical but profound problem: the lack of a common authority that can resolve conflicts.

Follow-up post

Partition, not sharing: An alternative approach to the Fed/non-Fed spectrum divide

Sharing a resource between parties entails an agreement about what the asset is, how it should be apportioned, and what happens when something goes wrong. Agreements can be made without a venue where disputes can be resolved, but they’re risky and infrequent; imagine how many contracts would be signed if there were no rule of law, and no courts. As the OECD Policy Framework for Investment notes: “The ability to make and enforce contracts and resolve disputes is fundamental if markets are to function properly.”

The most reliable and straightforward way to ensure dispute resolution is a common authority that both parties recognize and are subject to. The catch is that no such authority exists for disputes between Federal and non-Federal radio operators.

Federal spectrum is nominally managed by the NTIA, an agency of the Department of Commerce, though in fact the real decisions are jointly made by the Federal departments and agencies in the IRAC. In the end, though, everyone answers to the President of the United States (aka POTUS). On the other hand, non-Federal spectrum is managed by the FCC, an independent agency whose Commissioners are appointed by the President, but who not answerable to him. As this chart shows, there is no common authority between Federal and non-Federal radio operations. (It’s schematic in the extreme; green boxes are devices/users, blue boxes are authorities. For example, charting public safety operations and light-licensed assignments are left as an exercise for the reader.)

A partial view of devices, operators and authorities in US radio operations. Click for full-size image.

Fostering trust and enforcing contracts is easiest if parties are in the same organization. The closer parties are to each other (e.g. as measured by nearest common ancestor metrics, cf. graph theory and computer science), the easier sharing will be.

Apportioning operating rights and resolving disputes (that’s what I take “sharing” to mean) is easiest when parties have a close common authority, and they’re operating the same service. For example, Verizon bears the burden of making sure all its cellphone customers efficiently share its licensed channels. Sharing between Verizon and AT&T is also pretty straightforward, because they run similar services and are in the same industry. Similarly, microwave and satellite ground stations can share pretty well when they’re all registered with the same frequency coordinator, Comsearch in this example. On the Federal side, it’s straightforward for the Navy to coordinate all its radio operations, and for all the Armed Services to coexist; in the end they all report to the Secretary of the Department of Defense (DoD).

It gets harder as the closest common authority moves up the hierarchy. On the non-Federal side, disputes between (say) Verizon and a microwave operator have to be resolved by the FCC, and on the Federal side, the IRAC has to umpire arguments between (say) the Army and the FBI, or the FBI and NOAA.

Note that even though everyone on the non-Federal side is subject to FCC jurisdiction, disputes over unlicensed operation are more difficult to resolve, since there is no authority that controls all those devices; there is nobody for Verizon, say, to negotiate with if it would like to address transmissions by unlicensed devices.

The really tough case occurs when there is no common authority at all, e.g. between a non-Federal and a Federal service. There is no institution that both the FCC and POTUS are answerable to.

The most difficult situation of all arises when one combines unlicensed with Federal/non-Federal sharing, such as an unlicensed hotspot operating in frequencies assigned to Naval radar. Ironically, this is just the case that seems to be emerging poster child for sharing: allowing light-licensed wireless internet service providers (WISPs) to operate in the 3550-3650 MHz band identified in NTIA’s Fast Track evaluation (PDF); see e.g. filings by WISPA and Motorola.

Conclusion

It seems to me that efforts to encourage frequency sharing between Federal (e.g. military) and non-Federal (e.g. commercial) users are premature at best, and misguided at worst. Federal/non-Federal sharing is the hardest possible case, because of widely divergent goals and incentives, lack of trust, and the absence of a venue to resolve disputes. Government efforts are better focused on more tractable cases: sharing within administrative domains, and then between parties with a close common authority.

Commercial interests are clamoring today for more radio allocations, but Federal users are not far behind. Rather than trying to force sharing between them across a chasm of difference, the Administration should minimize the obstacles to sharing within each domain. A solid basis already exists for commercial/commercial sharing thanks to the FCC’s Secondary Market order (see the discussion in paras 38-41 of the Notice of Inquiry in ET Docket 10-237 PDF), and the Commission should ignore calls for a transfer of Federal allocations to non-Federal use until companies have shown they’ve exhausted these possibilities. The top priority should be to prepare the way for intensive sharing between Federal agencies. Adequate allocations and hands-off management have given these players a relatively easy go until recently. Crunch time is coming.

Spectrum Sharing: Not really sharing, and not just spectrum

2011-08-03T08:54:00.000-07:00

There was endless talk about spectrum sharing at ISART in Boulder last week. I’ve become increasingly confused about what those words mean, since wireless has been about more than one radio system is operating at the same time and place pretty much since the beginning.
For example, whitespace devices are said to share the UHF band with television, but the operating rules have been drawn up to ensure that whitespace devices never interfere with TV, i.e. never operate in the same place, channel and time. What’s “sharing” about that? The purpose of radio allocation from the start has been to avoid harmful interference between different radio operations, which has always been done by ensuring that two systems don’t operate in the same place, channel and time – such as two TV stations not interfering with each other.

It seems that the “new sharing” has three characteristics: (1) more boundaries (in geography, frequency and particularly time) than ever before; (2) the juxtaposition of different kinds of services that differ more from each other than they used to; and (3) sharing without central control. It’s a difference in degree, not in kind.

It’s not about sharing, since the goal is to avoid interference, i.e. to avoid sharing. It’s not about spectrum, i.e. radio frequencies, since non-interference is achieved not only by partitioning frequencies but also by dividing space, time, transmit power and the right to operate.

1. More, closer boundaries

The Prime Directive for spectrum management has been to minimize harmful interference. (I believe this is misguided; minimizing interference is important, but is simply a means to a more important end: Maximizing the value of radio operation. See How I Learned to Stop Worrying and Love Interference) This can be achieved by arranging for large buffer zones in space and frequency between radio operations, and removing time boundaries by giving a licensee 24x7 control of an allocation.

The increasing demand for radio operating rights – we now all want wireless, for every possible use, everywhere – means that the buffer zones have had to shrink. This leads to more risk of interference as the signals of different services begin to overlap; implementing sharp cut-offs in space and frequency are impossible in theory, and expensive in practice.

The time boundary comes into play when a secondary user is allowed to operate when the primary is silent. My impression is that this is what most people have in mind when they talk about sharing, and particularly when they use the term “dynamic” sharing – although the dynamism can, and does, also refer to hopping around in frequency and space.

In both these cases, “sharing” actually means “not sharing”: that is, the point of the rules and the technology is to keep competing systems from operating at the same time and place.

The addition of time boundaries makes things “dynamic”, another favorite word these days (Google Ngram): phrases such as dynamic spectrum access or DSA (cf. the name of the DySPAN conference), dynamic allocation of spectrum. Dynamic access – multiple devices coming and going – is nothing new; it’s the essence of cellular networks. Dynamic spectrum allocation – dynamic assignment to be more precise, since in the regulatory jargon allocation is determining what kind of services are allowed, and assignment determines which particular parties may operate – is new, since traditionally license assignments are effectively perpetual. (Unlicensed is a transitional case, since any type-compliant device may operate; the assignment is not made at all.)

Dynamic assignment of operating authorizations allows independent transmissions to be interleaved, and is the characteristic of the new spectrum sharing. Note, though, that the key to this flavor of DSA is sharing not spectrum (i.e. frequencies), but assignments.

Perhaps the only kind of sharing that counts as sharing, in my book, is when distinct systems radiate energy at the same time and frequency at the same time; the classic “underlay” case is UWB, where a very wide-band but very low power signal coexists with a number of relatively narrow-band but high power signals. This can also be done by having different systems using the same frequency, place and power level use different signal modulations that can be teased apart by smart receivers (i.e. orthogonal spreading codes).

2. Less like-with-like

Close-packed systems (in space and frequency) have been deployed for decades, but they have been like-with-like operation, e.g. a number of cellular operations in adjacent channels. This is relatively easy to manage, since the interference is symmetrical. The harder, and growing, problem arises when the near neighbors have very different technologies, spatial layouts and business models. The goal of sharing is still to avoid sharing (i.e. to avoid spill-overs from to the other), but it’s harder to manage.

3. Decentralized coordination

The debate about sharing is confounded by the introduction of systems where many devices coexist without coordinating with each other, i.e. unlicensed systems. These systems “share” because they may well trample on each other’s signals, but even here the goal – implemented through technology and standards, rather than regulatory rules – is to avoid sharing, e.g. by having systems listen for radio silence before transmitting, and using orthogonal modulation codes when they do transmit at the same time.

This confuses the debate because for some advocates, dynamic sharing is largely synonymous with more unlicensed operation, e.g. allowing unlicensed devices to be the secondary operators in the spatial, frequency or time gaps left by primaries. However, while the secondaries could be unlicensed, they obviously don’t need to be. In fact, coexistence between a primary and uncoordinated, unlicensed secondaries is the hardest problem to solve, both technically and institutionally; the challenges include adding horizontal coexistence (independent secondaries among themselves) to vertical coexistence (primary to secondary), and identifying bad actors among the unlicensed secondaries when coexistence breaks down. The problems, especially the engineering ones, are not insoluble; but why start with the hardest part?

Conclusion

I think my confusion about the meaning of "sharing" derives from two meanings of the verb "to share": (1) to divide and distribute in shares, to apportion; (2) to use, experience or occupy with others, to have in common (Merriam-Webster). Definition #1 is what spectrum allocation has been about since time immemorial, and is still the meaning that many ascribe to the term, while the very different Meaning #2 has come into the radio regulation debate in the baggage of Open Spectrum Access advocates, and is not universally accepted.

The LightSquared Mess Shouldn’t Count Against Coase

2011-06-26T20:04:00.000-07:00

It seems there’s a new meme floating around DC: I’ve been asked from both sides of the spectrum rights polemic whether the Lightsquared/GPS situation proves that Coasian make-spectrum-property advocates are crazy because the rights seem to be pretty well defined in this case, and yet the argument drags on at the FCC rather than being resolved through market deals. I suspect the source is Harold Feld’s blog My Insanely Long Field Guide to Lightsquared v. The GPS Guys where he says:

For a spectrum wonk such as myself, it simply does not get better than this. I also get one more real world example where I say to all the “property is the answer to everything” guys: “Ha! You think property is so hot? The rights are clearly defined here. Where’s your precious Coasian solution now, smart guys?”

The “Coasian” position does have its problems (see below), but this isn’t an example of one of them. I think Harold’s premise is incorrect: the rights are NOT well-defined. While LightSquared’s transmission rights are clear, GPS’s right to protection – or equivalently, LightSquared’s obligation to protect GPS receivers from its transmissions – is entirely unclear. There’s no objective, predictable definition of the protection that’s required, just a vague generalities, built into statute (see e.g. Mike Marcus’s Harmful Interference: The Definitional Challenge).

LightSquared’s transmission permissions are in some sense meaningless, since “avoiding harmful interference” will always trump whatever transmit right they have, and there’s no way to know in advance what will constitute harmful interference. I believe that’s a fundamental problem with almost all radio rights definitions to date, and why I’ve proposed the Three Ps.

The “Coasian” position’s real important problems are on view elsewhere:

(1) While negotiation between cellular operators to shift cell boundaries show that transactions can succeed in special cases, there is no evidence yet that transaction costs for disputes between different kinds of service will be low, and thus that negotiations will succeed in the general case. Even if one can ensure that rights are well defined, it may prove politically impossible to reduce the number of negotiating parties to manageable levels since radio licenses are a cheap way for the government to distribute largesse to interest groups. This is most obvious in the case of unlicensed operation, but many licensed services such as public safety and rural communications also result in a myriad of licensees.

(2) The FCC’s ability and proclivity to jump in and change operating rules (i.e. licensees rights) in the middle of the game makes regulatory lobbying more efficient than market negotiation. This may be unavoidable given law and precedent. There is no way for today’s Commission to bind tomorrow’s Commission to a path of action; legislation is the only way to do that, and even statute is subject to change.

(3) A significant chunk of radio services aren’t amenable to market forces since they’re operated by government agencies that can’t put a monetary value on their operations, and/or can’t take money in exchange for adjusted rights. Nobody is willing to quantify the cost of a slightly increased risk that an emergency responder won’t be able complete a call, or that a radar system won’t see a missile, even if those systems have a non-zero failure rate to begin with. And even if the Defense Department were willing to do a deal with a cellular company to enable cellular service somewhere, it can’t take the Cellco’s money; the dollars would flow to the Treasury, so there’s absolutely no incentive for the DoD (let alone the people who work for it) to come to some arrangement.

Protection Limits are not "Interference Temperature Redux"

2011-06-22T09:28:00.000-07:00

My post Receiver Protection Limits may have left the impression that reception protection limits are similar to the dreaded and ill-fated interference temperature notion introduced in 2002 by the FCC’s Spectrum Policy Task Force.

Receiver protections are part of the "Three Ps" approach (Probabilistic reception Protections and transmission Permissions - see e.g. the earlier post How I Learned to Stop Worrying and Love Interference, or the full paper on SSRN). While both the Three P and Interference Temperatur approaches share a desire to “shift the current method for assessing interference which is based on transmitter operations, to an approach that is based on the actual radiofrequency (RF) environment,” to quote from the first paragraph of the Interference Temperature NOI and NPRM (ET Docket No. 03-237), the Three Ps approach differs from Interference Temperature in four important ways:

1. The Three Ps focus on solving out-of-band, cross-channel interference, whereas Interference Temperature is concerned with in-band, co-channel operation

2. The Three Ps are used to define new operating rights, whereas Interference Temperature tried to open up opportunities for additional operations in frequencies allocated to existing licensees

3. The Three Ps do not grant second party rights, whereas Interference Temperature permits second party operation.

4. Three Ps rights are probabilistic, whereas Interference Temperature definitions are deterministic.

Receiver protection limits: Two Analogies

2011-06-22T09:24:00.000-07:00

I argued in Receiver protection limits that there are better ways to manage poor receivers causing cross-channel interference problems than specifying receiver standards. Here are two analogies to sharpen one’s intuition for the most appropriate way to handle such situations.

Cities increase the salinity of rivers running through them, affecting downstream agriculture. However, the choices that farmers make determine the degree of harm; some crops are much more salt-tolerant than others. In order to ensure that farms bear their part of the burden, regulators have a choice: they can either regulate which crops may be grown downstream, or they can specify a ceiling on the salinity of the water leaving the city limits, leaving it up to farmers to decide whether to plant salt-tolerant crops, perform desalination, or move their business elsewhere. Limits on salinity protection are a less interventionist solution, and don’t require regulators to have a deep understanding of the interaction between salinity, crops and local geography.

Sound pollution is another analogy to radio operation. Let’s imagine that the state has an interest in the noise levels inside houses near a freeway. It can either provide detailed regulations prescribing building set-backs and comprehensive specifications on how houses should be sound-proofed, or it could ensure that the noise level at the freeway-residential boundary won’t exceed a certain limit, leaving it up to home-owners to decide where and how to build. Again, noise ceilings are a simple and generic regulatory approach that does not limit the freedom of citizens to live as they choose, and that does not require the regulator to keep pace with ever-evolving technologies to sound-proof buildings.

Receiver protection limits: a better way to manage interference than receiver standards

2011-06-22T09:21:00.000-07:00

Radio interference cannot simply be blamed on a transmitter; a service can also break down because a receiver should be able to, but does not, reject a signal transmitted on an adjacent channel.

More on this topic in subsequent posts:

Receiver protection limits: Two Analogies
Protection Limits are not "Interference Temperature Redux"
The LightSquared Mess Shouldn’t Count Against Coase
Licensing radio receivers as a way to facilitate negotiation about interference

The LightSquared vs. GPS bun fight is a good example of this “two to tango” situation. GPS receivers – some more so than others – are designed to receive energy way outside the allocated GPS bands which means that operation in the adjacent band due to a new service like LightSquared can cause satellite location services to fail. Without the LightSquared transmissions, there wouldn’t be a problem; but likewise, if GPS receivers were designed with the appropriate filters, they could reject the adjacent LightSquared transmissions while continuing to receive the satellite location signal and function normally. [1]

While the responsibility for interference is, in theory, shared between transmitters and receivers, radio regulation has traditionally placed the onus on a new transmitter to fix any problems that may arise. [2] As I will argue, receiver standards are an impractical response; limits on reception protection, formulated in terms of the RF environment rather than equipment performance, are preferable.

Receiver Standards - an impractical response

Transmitters are thus at the mercy of cheap and nasty receivers. This has led to repeated calls for receiver standards, i.e. government mandated performance specifications. However, for the same reasons outlined above for water pollution and noise control, I feel strongly that receiver standards are the wrong way to go: they require the regulator to understand receiver performance in detail, they lock in a particular service scenario since they refer to particular receiver architectures, and they unnecessarily limit the freedom of receiver operators to respond to interference in ways that the regulator did not anticipate. [3]

Specifying receiver performance parameters to prevent cross-channel interference is complicated, since there are so many of them, and different types of receivers are characterized in different ways. The complexity of standards for receiver performance in on display in the NTIA Report 03-404, Receiver Spectrum Standards: Phase 1 – Summary of Research into Existing Standards. This document summarizes US federal agency, US industry association, and international standards. The parameters used to specify receiver standards in the NTIA Manual vary from service to service, and include adjacent channel rejection (different values for analog and digital), EMC tolerance, frequency stability, image rejection, intermodulation rejection, receiver interference suppression circuitry, selectivity, and spurious rejection (see Table 1). The Department of Agriculture’s specifications for VHF High-Band receivers is a 12 row by 6 column table of parameter values (see Table 5).
The only case I’m aware of where the FCC has provided receiver specifications, 47 CFR 15.118 regarding cable TV receivers provides very detailed requirements and measurement methods for adjacent channel interference, image channel interference, direct pickup interference, tuner overload and cable input conducted emissions.

Even if receiver standards were future-proof and practical in terms of regulatory engineering, experience shows that they are impractical in political or regulatory terms. For example, the notice of inquiry into “Interference Immunity Performance Specifications for Radio Receivers” launched in March of 2003 (ET Docket No. 03-65) was formally terminated by the Commission in 2007 without any action being taken after extensive opposition from industry.

A better way: Receiver Protection Limits

Fortunately, I believe there is an alternative approach: reception protection limits, part of the “Three P” approach (Probabilistic reception Protections and transmission Permissions - see e.g. the earlier post How I Learned to Stop Worrying and Love Interference, or the full paper on SSRN). A license would state the degree of receiver protection that it affords; then it’s up to the licensee to decide whether they want to pay extra for better receivers, or opt for cheaper ones with poorer interference rejection. A license might state that energy from other allocations would not exceed of a field strength X inside the licensed frequency range (aka “in-band”), and would not exceed field strength Y outside that range (“out-of-band”), for some large percentage of times and locations, e.g. 90% or locations, 90% of the time. Any energy up to this amount would be deemed not to be grounds for a harmful interference complaint. [4]

The number of parameters to be specified in the protection limit approach is small compared to the multifarious possibilities for receiver standards; and more importantly, the parameters are independent of the technology used in the receiver. The method for managing receiver performance, i.e. the variables to be used, is therefore the same for every band, though the parameter values may differ from band to band. With receiver standards, experience shows that each instance uses a different set of variables.

Summary

To sum up, specifying receiver performance standards is complicated, technology-bound and unlikely to survive the political gauntlet. It’s simpler and more future-proof to declare the ceiling of energy in the adjacent band that a licensee’s receivers (will) have to deal with, and then let it make the commercial calculation. Reception is protected, but only up to a point. If receivers are lousy but the regulator wants to force them to get better over time (cf. GPS) it would start with a lower out-of-band protection number (i.e. adjacent signals will be forced to be weaker) and then push it up in steps over (say) five year increments.

NOTES

[1] Some background in the blogs (for some reason, all the GPS/LightSquared briefers seem to be insanely long…): Pro GPS 1 (Andrew Seybold), Pro LightSquared (Mike Marcus), Pro GPS 2 (GPS World), Pro LightSquared 2 (Harold Feld)

[2] Even when operating within the terms of its license, a transmitter may not cause “harmful interference” (a term that is nowhere satisfactorily defined cf. Mitch Lazarus, Mike Marcus) to an existing service. 47 CFR § 2.102 (f) provides a blanket requirement that licensees shall “use frequencies so separated from the limits of a band allocated to that service as not to cause harmful interference to allocated services in immediately adjoining frequency bands.” The presumption is that newly entering transmitters cause interference to incumbent receivers. 47 USC § 302a authorizes the Commission to make rules to govern the interference potential of any transmitter (a)(1), but only receivers in the home (a)(2). Avoiding harmful interference is the (misguided, in my opinion) guiding idea: 47 USC § 303 para (f) authorizes it to “[m]ake such regulations not inconsistent with law as it may deem necessary to prevent interference between stations”; para (y)(2)(C) gives it “authority to allocate electromagnetic spectrum so as to provide flexibility of use, if . . . (2) the Commission finds, after notice and an opportunity for public comment, that . . . (C) such use would not result in harmful interference among users.”

[3] For example, the choice of whether to deal with out-of-band interference in a radio’s front end vs. baseband (e.g. improving the LNB vs. increasing the ENOB required in the ADC) is a dynamic, case-by-case trade-off; receiver standards would lock in the performance required of the front-end, and not give the design engineer the option of improving the ADC if that were cheaper.

[4] Sample Parameter Values

As an example of possible parameter values, the regulator might choose the in-band ceiling X = 57 dBuV/(m.MHz) in-band peak and average, which is the 47 CFR §15.209 per-transmitter out-of-band emission limit for intentional radiators, plus 3 dB added for the case there happen to be two close-in transmitters. Alternatively, it could be based on the ceiling on the resulting field strength outside cellular license area, e.g. 47 dBuV/m in the 2 GHz bands (§ 27.55) plus a 3 dB fudge factor for multiple signals: let’s say 50 dBuV/(m.MHz).

The out-of-band limit would be higher, e.g. Y = 108 dBuV/(m.MHz) peak and average, which is the field strength 3 meters from a 13 dBm/MHz transmitter like a CDMA cellphone: 13 dBm = 20dB (100 mW) spread over 5 MHz, less 7 dB = 5 MHz/1 MHz).

These definitions assumed that peak and average power (in time) were essentially the same, which is a valid assumption for continuous broadband operations like digital cellular. A specification might give different limits on peak and average power to accommodate intermittent transmissions, since an occasional transmission of very high power could have a low average field strength but high interference potential.

Note also that these definitions are given in terms of flux density (e.g. field strength per hertz) rather than total power in the band – again, a reasonable assumption for broadband services. Since out-of-band interference is usually the result of total delivered power, regardless of whether it is smoothed out or peaked in frequency, it may be sufficient to specify an aggregate out-of-band field strength limit, e.g. Y’ = 115 dBuV/m over the 5 MHz nearest the band edge.

Too strategic to be true?

2011-04-19T10:30:00.000-07:00

The cellular industry has been very vocal in calling on the FCC to allocate more spectrum licenses to satisfy the forecast demand for mobile data services. For two examples more or less at random, see this CTIA white paper, and the 4G Americas white paper “Sustaining the Mobile Miracle” (PDF).

On reflection, though, it strikes me as rather curious behavior for cut-throat competitors. More spectrum licenses won’t satisfy the insatiable demand for wireless data capacity any more than building highways reduces traffic congestion, and while it might make strategic sense, in the short term – and isn’t that all that really matters for listed companies, when all the rhetoric is said and done? – it means that the cellcos are giving up a wonderful opportunity to make money.

If the supply of spectrum licenses were fixed, and not increased by reallocation of other services to mobile wireless, then Economics 101 dictates that the price for wireless data would rise. (This is ignored in the forecasts; see e.g. my post Cisco’s Fascinating Flaky Forecast.) Operators wouldn’t incur the capital costs of lighting up new frequencies, and so their profits would rise – a lot!

On the other hand, if more cellular licenses were made available, the carriers would not only have to buy them at auction, but they would have to buy and install the infrastructure to use them. The price they could charge for wireless data service wouldn’t change much, and so their profits would go down, or at best stay flat.

All that said, though: these companies are much, much smarter business people than I am. I must be missing something. But what?

Perhaps this is all just a big CYA operation. When the inevitable demand crunch happens (with or without new cellular licenses, demand is set to outstrip supply), the operators will be able to blame the government: “Dear customer, it’s not our fault, we’ve been asking the government to help us give you the services you want, but they didn’t come through. We’re sorry, but all we can do to make sure that those who really need wireless services get them is to increase prices.”

“Quiet” doesn’t mean “unused”: The Downside of Under-defined Radio Rights

2011-03-01T12:36:00.000-08:00

The FCC has promised to find and reallocate 500 MHz of radio frequencies to satisfy the burgeoning demand for high bandwidth mobile services such as video on cell phones. The idea, the hope, is that there are lots of unused bands to be repurposed. “Unused” is a tricky notion, though. I’ll take it to mean “radio quiet”: a radio energy detector doesn’t observe much if anything at certain frequencies, and the assumption is that a new service could transmit here.

Of course, nothing is as simple as that. Let’s assume that the services that actually operate in these quiet bands – and there are always incumbents, since every frequency has one if not many nominal users – can be found a new home, and that they’ll relocate. The harder problem is that a quiet band may not in fact be usable because of the equipment in neighboring bands. The LightSquared/GPS argument is a conveniently current example. The proposal to allow LightSquared to deploy lots of ground-based transmitters in a band where to date only satellite transmissions were allowed has caused shock and outrage among GPS users who claim that their receivers cannot distinguish between the LightSquared signal in the adjacent band and the satellite location signals in the GPS channel.

Since the FCC’s rules and precedents provide almost unlimited protection against "harmful interference" (a notoriously vague term) caused by new services, an incumbent is pretty much assured that it will be held harmless against any change. The situation is exacerbated because FCC licenses only specify transmission parameters and say nothing about the radio interference environment that receivers should be able to cope with. Radio receivers are thus designed and built as if their radio environment will never change; if a band has been quiet, none of the receivers in the adjacent frequencies can cope with more intensive use, since building in that protection costs money. (For complementary perspectives on this problem, and suggested remedies, see two short papers presented at a recent conference in Washington, DC: Kwerel and Williams, De Vries and Sieh.)

Thus, just because a band is quiet doesn’t mean that it’s unoccupied; it’s probably effectively occupied by the protection afforded to the cheap receivers next door that haven’t been required to, and therefore don’t, tolerate any substantial operation in the quiet channel. It’s as if the incumbent were a householder whose property used to be passed by track along which only ox wagons passed. She didn’t have to take any precaution against her dogs being run over by a wagon, such as building a fence, and this unlimited protection still holds even when the track is turned into an arterial road, holding passing vehicles completely responsible if a dog is run over.

Money could, but might not, solve the problem. Let’s say Tom Transmitter wants to deploy a new service
in the formerly quiet band, and that this would cost the incumbent neighbor, Rae Receiver, $300 million, either in lost revenue from diminished service and/or because of precautions such as new receiver filters that are needed to reject Tom’s adjacent band signals. If the benefit to Tom is big enough, if for example he could generate $500 million in profit, Tom could compensate Rae and still come out ahead. But how is the $200 million of potential gain ($500 million - $300 million) to be divided? This depends on Rae’s rights. If she has the right to prevent any operation by Tom (i.e. she can take out an injunction against him), she can demand essentially all his profits as a condition of operationlet’s say $499 million of his $500 million, whereas if she’s entitled to damages, she can only demand $300 million for actual losses. These are very different outcomes. Under an injunction, Tom’s incremental net profit is $1 million ($500 million - $499 million) and Rae’s is $199 million ($499 million - $300 million), whereas under damages, Tom’s net profit is $200 million and Rae’s is zero.

However, since the FCC doesn’t specify whether licenses are protected by damages or injunctions, Tom and Rae can’t begin to deal, since the legal basis of the negotiation is unclear. Tom will hope that he can get the whole $200 million incremental gain, and Rae will hope for it, too – a huge difference in expectations that will almost inevitably prevent the parties from coming to an agreement.

(There are further obstacles to reaching a settlement that I won’t go into here, such as uncertainty over what action by Tom actually constitutes damage to Rae due to ambiguity in the way FCC rules are currently formulated, and the freeloader/hold-out problems with negotiations involving many parties.)

What's to be done?

1. Any inventory of “unused” radio capacity should not only itemize radio quiet bands, but also the nature of the service and receivers next door, so that the cost of relocating, protecting or degrading the incumbent service can be estimated.

2. Any new licenses that are issued should specify whether they’re protected by injunctions or damages; this will facilitate negotiation.

3. Any new license should specify the receiver protection parameters the operator can rely on, and by implication what will not be protected.

4. Regulators should start retrofitting existing licenses to this new approach by specifying the remedy (#2) and laying out a timeline over which receiver protections (#3) will be dialed down from the current open-ended “no harmful interference” condition to more realistic and objective received energy levels.

The two page position paper I referenced above gives a quick introduction to these measures; for all the gory details, see the 15 page long version on SSRN.

Juggling Pipes: orchestrating scarce radio resources to serve multifarious applications

2011-02-21T14:53:00.000-08:00

I concluded in Cisco’s Fascinating Flaky Forecast that the impending supply/demand mismatch in wireless data services presents opportunities for “innovations that improve effective throughput and the user experience”. This post explains one example: a software layer that that matches up various applications on a device to the most appropriate connectivity option available, mixing and matching apps to pipes to make the cheapest, fastest, or most energy efficient connection. (In academic terms, it’s a version of Joe Mitola’s Cognitive Radio vision.)

Peter Haynes recently prompted me to ask some experts what they thought the most exciting wireless technology developments were likely to be in the next decade. Mostly the answer was More of The Same; a lot of work still has to be done to realize Mitola’s vision. The most striking response was from Milind Buddhikot at Bell Labs, who suggested that the wireless network as we know it today will disappear into a datacenter by 2020, which I take to mean that network elements will be virtualized.

I don’t know about the data center, but from a device perspective it reminded me of something that’s been clear for some time: as a device’s connectivity options keep growing, from a single wired network jack to include one or more cellular data connections, Wi-Fi, Bluetooth, UWB, ZigBee etc., as the diversity of applications and their needs keeps growing, from an email client to many apps with different needs including asynchronous downloads, voice and video streams, and data uploads, and as choosing among becomes more complicated, such as trade-offs between connectivity price, speed, quality of the connection, and energy usage, there is a growing need for a layer that sits between all these components and orchestrates all these connections. Can you say “multi-sided market”?

The operating system is the obvious place to do such trade-offs. It sits between applications and peripherals, and already provides apps with abstractions of network connectivity. As far as I know, no OS provider has stepped up with a road map “smart connectivity.” It’s decidedly not just “smart radio” as we’ve heard about with “white spaces”; the white space radio is just one of the many resources that need to be coordinated.

For example, one Wi-Fi card should be virtualized as multiple pipes, one for every app that wants to use it. Conversely, a Wi-Fi card and a 3G modem could be bonded into a single pipe should an application need additional burst capacity. And the OS should be able to swap out the physical connection associated with a logical pipe without the app having to know about it, e.g. when one walks out of a Wi-Fi hotspot and needs to switch to wide-area connectivity; the mobile phone companies are already doing this with Wi-Fi, though I don’t know how well it’s working.

That said, the natural winner in this area isn’t clear. Microsoft should be the front-runner given its installed base on laptops, its deep relationships with silicon vendors, and its experience virtualizing hardware for the benefit of applications – but it doesn’t seem interested in this kind of innovation.

Google has an existential need to make connectivity to its servers as good as it could possibly be, and the success of Android in smartphones gives it a platform for shipping client code, and credibility in writing an OS. However, it is still early in developing expertise in managing an ecosystem of hardware vendors and app developers.

The network operators don’t much end-user software expertise, but they won’t allow themselves to be commoditized without a fight, as they would be if a user’s software could choose moment-to-moment between AT&T and Verizon’s connectivity offers. The telcos have experience building and deploying connectivity management layers through orgs like 3GPP. Something like this could be built on IMS, but it’s currently a network rather than device architecture. And the network operators are unlikely to deploy software that allows the user to roam to another provider’s data pipes.

The chipset and handset vendors are in a weaker position since they compete amongst themselves so much for access to telcos. Qualcomm seems to get it, as evidenced by their Gobi vision, which is several years old now: “With Gobi, the notebook computer becomes the unifying agent between the different high speed wireless networking technologies deployed around the world and that means freedom from having to locate hotspots, more choice in carrier networks, and, ultimately, freedom to Gobi where you want without fear of losing connectivity – your lifeline to your world.” As far as I can tell, though, it doesn’t go much beyond hardware and an API for supporting multiple 3G/4G service providers on one laptop. Handset vendors like

Vendors like Samsung or HTC could make a go of it, but since network operators are very unlikely to pick a single hardware vendor, they will only be able to get an ecosystem up to scale if they collaborate in developing a standard. It’s more likely that they will line up behind the software giants when Google and/or Microsoft come forward with their solutions.

It is also possible that Cisco (or more likely, a start-up it acquires) will drive this functionality from the network layer, competing with or complementing app/pipe multiplexing software on individual devices. As Preston Marshall has outlined for cognitive radio,* future networks will adapt to user needs and organize themselves to respond to traffic flow and quality of service needs, using policy engines and cross-layer adaptation to manage multiple network structures. There is a perpetual tussle for control between the edge of the network and the center; smart communications modules will be just another installment.

* See Table 4 in Preston F Marshall, “Extending the Reach of Cognitive Radio,” Proceedings of the IEEE, vol. 97 no. 4 p. 612, April 2009

Cisco’s Fascinating Flaky Forecast

2011-02-12T10:26:00.000-08:00

Ed Thomas prompted me to have a look at Cisco’s recently published Visual Networking Index: Global Mobile Data Traffic Forecast Update, 2010–2015.

The numbers are staggering: global mobile data traffic grew 2.6-fold in 2010, nearly tripling for the third year in a row; mobile video traffic will exceed 50% for the first time in 2011; and Cisco predicts that global mobile data traffic will increase 26-fold between 2010 and 2015. Big numbers forecast by someone who’ll make money if they come true are always suspect, though. While the historical data are largely indisputable – and amazing – I think the forecasts are bogus, though in interesting ways.

Flags went up at the projection of 92% CAGR in mobile traffic growth over the next five years. From the scant details on assumptions provided in the report, I suspect the overall growth is driven (more than driven, in fact) by the growth in the number of users, not by increases in per-user usage. For example, Cisco predicts that the number of mobile-only Internet users will grow 25-fold between 2010 and 2015 to reach 788 million, over half of them in “Asia Pacific” (defined to exclude Japan).

Working back from their forecast data volumes and assumptions on user growth, however, suggests that usage per user (I prefer to think in terms of Megabits/second rather than ExaBytes/month) doesn’t increase over the study period, an in fact declines.

The growth in traffic thus hinges on the global user base growing to almost 800 million mobile-only users in five years, from 14 million today. That’s staggering, and to me implausible.

If nothing else, though, this demonstrates that using Cisco’s meganumbers don’t necessarily imply an impending bandwidth crunch doesn’t hold water. It doesn’t mean there isn’t going to be one, just that growth numbers don’t imply/require it, because they’re in large part driven by hundreds of millions of new users in China.

A more fundamental flaw is that the analysis is entirely demand driven. This was probably fine when Cisco was predicting wireline use, since there is so much dark fiber that supply is essentially unlimited. However, one cannot ignore the scarcity of radio licenses. We’re near the Shannon limit of the number of bits/second that can be extracted from a Hertz of bandwidth, and massive new frequency allocations will not show up overnight. An alternative is to reduce cell size and serve more users per cell by using smart antennas; however, such a build-out will take time. I don’t know how much extra traffic one can fit into the existing infrastructure and frequencies, but Cisco should at least have made an argument that this doesn’t matter, or that it can ramp up as fast as the demand.

While there may be spare capacity in China, there’s clearly a supply question in markets that are already halfway up the growth curve, though, like the US. Cisco ignores this. In North America they’re forecasting that the number of mobile-only internet users will go from 2.6 million to 55.6 million (!). It’s reasonable to assume that these most of these new users are in places that are already consuming a lot of capacity, and that one will need more radio bandwidth to deliver more data throughput.

Cisco is forecasting that throughput will go from 0.05 ExaB/mo to 1.0 ExaB/mo for North American users. That’s a factor of 20. It’s hard to see how you get there from here without massive reengineering of the infrastructure.

One could get 2x by doubling available licenses from 400 MHz to 800 MHz; the FCC is talking about finding 500 MHz of new licenses for mobile data, but this is a pipe dream; if not in principle, then in the next five years given how slowly the gears grind in DC.

The extra throughput isn’t coming from offloading traffic from the wireless onto the wired network; Cisco considered this, and is forecasting 39% for offload that by 2015. Let’s say they’re conservative, and it’s 50%: that’s just another 2x.

Spectral efficiency, the bits/second that can be extracted from a Hertz of bandwidth, isn’t going to increase much. Engineers have made great strides in the last decade, we’re approaching the theoretical limits. Maybe another 50%, from 4 bps/Hz to 6 bps/Hz? Even an implausible doubling to 8 bps/Hz is just another 2x.

So by using heroically optimistic assumptions one can get an 8x increase in capacity – nowhere near that 20x Cisco is forecasting.

And last but not least, the forecast method ignores Econ 101: if demand increases with limited supply, prices will go up, and this will suppress demand. Not only does the study ignores supply, it also ignores supply/demand interactions.

Still, let’s stipulate that the demand forecast is accurate, and that grant me that supply is going to be constrained. The consequence is that there will be millions of screaming customers over the next few years when they discover that the promise of unlimited mobile connectivity cannot be delivered. The pressure on government will be huge, and the opportunities for innovations that improve effective throughput and the user experience in a world of scarcity (relative to expectations) will be immense. A crisis is coming; and with it the opportunity to make fundamental fixes to how wireless licenses are managed, and how applications are delivered.

Ways of Knowing

2011-02-03T14:03:00.000-08:00

Reading St Augustine’s Confessions reminded me of the Buddhist tradition's three ways of knowing, or "wisdoms": experiential/mystical, cerebral/rational, and learning/textual. (The Pāli terms are bhavana-mayā paññā, cintā-mayā paññā and suta-mayā paññā, respectively.)What strikes me about Augustine is his depth in all three methods; most people seem comfortable in one or at most two of them.

People may debate at cross purposes because they use different approaches to understand the world. Someone who thinks about the world experientially will have difficulty finding common ground with someone grounded in logic, and both may belittle someone who defers to tradition or social norms.

When I shared this idea with Dor Deasy, she pointed out that John Wesley thought faith should be approached from four perspectives: Experience, Reason, Scripture and Tradition, which map to the three above if one combines Scripture and Tradition. According to Wikipedia, the Wesleyan Quadrilateral can be seen as a matrix for interpreting the Bible in mutually complementary ways: “[T]he living core of the Christian faith was revealed in Scripture, illumined by tradition, vivified in personal experience, and confirmed by reason.”

Different personality types approach faith in different ways, though. Peter Richardson’s Four Spiritualities: Expressions of Self, Expression of Spirit uses the Meyers-Briggs personality inventory to characterize an individual’s bent. It may come down to brain physiology: I would not be surprised to learn that some people's brains are built in a way that predispose them to mystical experiences, while others are optimized for logic, or absorbing social norms.

Forging bits

2011-01-02T14:39:00.000-08:00

William Gibson makes passing reference to the art and craft forging documents early on in Spook Country, telling about trips to second hand bookstores to buy just the right paper, and ageing credentials by carrying them around.

Nowadays, though, paper is optional. Checks can be deposited by snapping pictures of front and back and sending to the bank, and airlines scan pictures of boarding passes from your phone at the gate.

Paper credentials decentralize verification. When it's difficult to "call HQ" to check identity - which it used to be until very recently - the attestation had to stand on its own feet, carrying the full burden of authenticating not only its bearer but also itself. Nowadays a database look-up is instantaneous, and the database can not only produce the photo of the person making the identity claim, but can also track whether multiple claims are being asserted simultaneously in different places.

The locus of forgery thus moves from the edge to the middle: you don't hack the passport, you hack the passport database. With a suitably large investment in securing the center, it becomes harder for street freelancers to generate credentials as they go, "at retail". However, there is now a single point of failure, and a successful hack of the central database can generate an unlimited number of false documents. As always when moving from bricks to clicks, the upfront cost is huge, but the marginal cost is negligible.

The discretion of, and trust required in, the agent at the edge diminishes. When paper documents had to be checked, officers developed a feel for a fake by handling tens of thousands of them over years, and their instincts could tell them something was off long before the official notice came around. Not all of them were equally good, though, and a rookie might miss a dud that an old hand would see a mile off. Now the quality of authentication depends on the security and agility of the central repository; if it can be broken, or is slow to respond to an exploit, a hack that works will work everywhere, immediately.

One might therefore expect that digital spooks and their paymasters are working not only on building bit-bombs to disable infrastructure, but constructing trapdoors to facilitate the forgery of digital credentials. "Identity theft" is probably not the half of it; identity creation (and destruction) is much more valuable.

Law without Categories?

2010-12-29T09:29:00.000-08:00

A recent New Scientist story about the descent of birds from dinosaurs (James O'Donoghue, Living dinosaurs: How birds took over the world, Section 2, Was archaeopteryx really a bird?, 08 December 2010; subscription required) contained this passage:

The real question is, where do you draw the line between dinosaurs and birds? Ask different palaeontologists and you will get subtly different answers. That is because the distinction is basically arbitrary, says Xing Xu of the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing, China, who discovered many of the Chinese fossils [of feathered dinosaurs].

This is a common theme in biology: the boundaries between species are arbitrary. And yet we continue to think in terms of species, since categorization is such a strong human reflex.

Jurisprudence and regulation in particular is built on categorization, defining categories that determine the response to a particular situation. At the heart of current network neutrality argument is the question of whether a company falls in "Title II" in which case a whole raft of telecommunication regulation regarding common carriage applies, or "Title I" in which case they are much more lightly regulated.

However, as the analogy to biology illustrates, most interesting categories have fuzzy boundaries, making for a delightful amount of work for lawyers and lobbyists, but not necessarily helpful outcomes.

Taxonomies are backward-looking; they attempt to fossilize a reality but are constantly open to revision. (This necessity for revision undermines the certainty which category-based rules purport to offer since categories are less robust than they appear, necessitating the case-by-case interpretation which proponents of rules contend is the weakness of the alternative approach, principles-based regulation.) They evidently work well enough, though; they're pervasive. A paper by David Bach & Jonathan Sallet about VOIP regulation (The challenges of classification: Emerging VOIP regulation in Europe and the United States, First Monday, Volume 10, Number 7, 4 July 2005) explains the situation very well:

From a practical point of view, classification stands out because classifying different services is what regulators principally do. In an ideal world, one could just draw up rules for VOIP that address the aforementioned critical issues, keeping in mind the technology’s novelty and the substantial differences that exist between conventional circuit–switched telephony and innovative packet–switched VOIP. In the real world, however, a first step in the regulation of new technologies is usually to try to fit them into existing service categories, in part because those are the tools that regulators work with and in part because classification can provide shortcuts through complex regulatory problems. Alternatively, regulators may be inclined to ask whether VOIP service is "like" or "substitutable" for current services — an approach that may obscure technological achievement. Either way, much is at stake in these decisions.

Fitting VOIP into existing regulatory categories is not simply an administrative or technical act. Since categories are associated with distinct sets of rights and responsibilities that have distributional and market strategic implications, a large number of stakeholders have mobilized to affect the outcome. . . .

Unpacking the political economic dynamics of evolving VOIP regulation highlights a second, more analytic reason to focus on classification. The debate over how to classify VOIP represents the leading edge of the question whether regulatory classification is useful in a world of converging technologies. . . .

In the eyes of most regulators and industry observers, correctly categorizing VOIP provides a shortcut through regulatory uncertainty. Yet precisely this is the problem with classification. As policymakers almost reflexively ask how a new technology fits into existing categories, the underlying political and social objectives of regulation can get lost.

A behavioral alternative comes to mind: the regulations that should apply do not derive from the category into which an action falls, but from its consequences; in Bach & Sallet's terms, one needs to look to the political and social outcomes, not the inputs.

Not even a metaphor

2010-12-09T09:11:00.000-08:00

Said Industry Minister Eric Besson, describing an upcoming auction of radio licenses in France, "These frequencies are of very, very high quality." What? How can a frequency, merely an attribute of electromagnetic radiation, be of high quality?

I’ve been inveighing against the misuse of spectrum metaphors for some time, but it took this quote to make me realize that the figure of speech at issue is really metonymy, not metaphor.

Metonymy is referring to something not by its name, but by something that is intimately associated with it (Wikipedia). Some examples:

The designers come up with the ideas, but the suits (worn by executives) make the big bonuses.

The pen (associated with thoughts written down) is mightier than the sword (associated with military action).

Freedom of the press (associated with the journalists and what they write) is an important value.

The White House (associated with the President and his staff) stood above the fray.

He bought the best acres (associated with the land measured in acres).

Both metaphor and metonymy substitute one term for another: metaphor by some specific similarity, and metonymy by some association. In spectrum language both are at work, for example in “Guard bands leave too many frequencies (or spectrum) lying fallow.”

Metonymy: Frequencies are associated with radio licenses

Metaphor: Radio licenses are like title to property

Metonymy: Property title is associated with the land to which it relates

Metaphor: Fallow land stands for any underused asset

Heresy as Diagnostic

2010-12-04T19:25:00.000-08:00

Heresies, or more exactly, the arguments that lead to one perspective being labeled as orthodoxy and the other as heresy, are pulsing pointers to a religion’s sore spots, those questions of doctrine or practice that have multiple plausible but incompatible answers. Heresy seems to be a useful tool for analyzing a set of beliefs. (Any book recommendations gratefully received.)

I was drawn to the question of heresy by reading Augustine’s Confessions, and Peter Brown’s masterful biography, Augustine of Hippo (1967, 2000). For instance, comparing Augustine and Pelagius, he writes

“The two men disagreed radically on an issue that is still relevant, and where the basic lines of division have remained the same: on the nature and sources of a fully good, creative action. How could this rare thing happen? For one person, a good action could man one that fulfilled successfully certain conditions of behavior, for another, one that marked the culmination of an inner evolution. The first view, was roughly that of Pelagius; the second, that of Augustine.”

My guess is that the choice between solutions that leads to a perspective being labeled heresy is necessary for a consistent set of beliefs, but that something is lost when the choice is made. I’m reminded of Isaiah Berlin’s approach to conflicts of values, summed up thus by John Gray in an interview with Alan Saunders on the Philosopher’s Zone (Australian Radio National, 6 June 2009)

“ . . . the idea that some fundamental concepts of human values are intractable, rationally intractable, in the sense that first of all they can't be resolved without some important loss, and secondly reason is very important in thinking about these conflicts, and then being clear about what they are, what they're between and what's at stake in them. [E]qually reasonable people can come to different judgments as to what ought to be done, so certain types of conflict of value are intractable. . . . So this idea of a kind of fundamental and intractable moral scarcity if you like in human life, such that there have been and there will always be intractable, the conflicts of values, and we can resolve them more or less intelligently in particular contexts that can be more or less skillful and intelligent and reasonable settlements of these conflicts, but they can never be overcome or left behind.”

Such differences may point to a conflict between incommensurable world views. For example, in an article about “relativity deniers”, (Einstein's sceptics: Who were the relativity deniers?, New Scientist 18 November 2010, subscription required) Milena Wazeck explains,

"Einstein's opponents were seriously concerned about the future of science. They did not simply disagree with the theory of general relativity; they opposed the new foundations of physics altogether. The increasingly mathematical approach of theoretical physics collided with the then widely held view that science is essentially simple mechanics, comprehensible to every educated layperson."

I would not be at all surprised if there is at least something like this at play in the argument over climate change; opponents have been all but branded as heretics, and there is religious fervor on both sides.

Better Radio Rights

2010-11-30T12:20:00.000-08:00

Demand for wireless services is growing relentlessly, but the ambiguous definition of rights and unpredictable enforcement has led to prolonged inter-service interference disputes that impede innovation and investment.

Silicon Flatirons organized a conference on this topic in DC a couple of weeks ago. The goal was to explore how radio operating rights could best be defined, assigned and enforced in order to obtain the maximum benefit from wireless operations. The event web site has links a fascinating set of position papers prepared by the panelists. There’s also a compendium that collects them all in one place (PDF).

Kaleb Sieh and I proposed (position paper, full paper on SSRN) an approach to radio operating rights based on three principles: (1) aim regulation at maximizing concurrent operation, not minimizing harmful interference; (2) delegate management of interference to operators; (3) define, assign and enforce entitlements in a way that facilitates transactions.

We argue that radio rights should be articulated using transmission permissions and reception protections, defined probabilistically (the Three Ps): transmission permissions should be based on resulting field strength over space and frequency, rather than radiated power at a transmitter; reception protections should state the maximum electromagnetic energy an operator can expect from other operations; both are specified probabilistically. This formulation of operating rights does not require a definition of harmful interference, and does not require receiver standards.

Since any initial entitlement point is unlikely to be optimal, the regulator should facilitate the adjustment of rights by: limiting the number of parties to a negotiation should be limited by minimizing the number of recipients, and enabling direct bargaining by effective delegation; recording a complete and current description of every entitlement in a public registry; stipulating the remedy (injunctions or damages) that attaches to an operating right when it is issued; the regulator refraining from rulemaking during adjudication; leaving parameter values unchanged after an entitlement has been defined, although values may be adjusted though bilateral negotiation between operators, and the regulator may add new parameters at license renewal.

Who gets the apple? Part II: A salty problem

2010-10-16T08:19:00.000-07:00

Here's another analogy; one that includes a nod to dispute resolution. For those who know and/or love Coasian economics, it's our old friend the pollution example, though tweaked to be radio interference in light disguise. It's also, incidentally, based on a true story I heard from someone who works for a large county's water district.

Imagine a city along a river, and a downstream farming community. Urban development results in more salt being added to the river; increased salinity can reduce crop yield. Salty water is therefore analogous to radio interference between transmitters (cities) and receivers (farms).

The harm to crops is a shared responsibility, though. For example, the city can reduce the amount of downstream salt by building a water treatment plant, and the farmers can accomodate more salty water by changing crops - spinach will be fine on water that's too salty for celery.

Let's imagine that a Federal Crops Commission (call it the FCC2) is responsible for managing this problem. It might instruct the city and farms to "coordinate" to find a solution to the problem, with a guideline that water may not be "too salty". As in the apple example, this is difficult to do without defining what counts as too salty, and who bears the responsibility for salinity.

If the FCC2 limits the salt the city can dump in the river like the FCC controls radio emissions, it would specify a ceiling of, say, 5 tons of salt per day - with a rider that the resulting water can't be "too salty". This is not very helpful to the farmers, however, since they care about the resulting salinity; seasonal variations in water volume or the salinity entering the city limits from upstream affect the resulting salinty. It doesn't help the city either, since it can't be sure how much water treatment capacity to build; 4 tons/day of salt might still turn out to be too much if the farmers downstream choose salt-intolerant crops and/or the river level is too low.

Matters are compounded when the city and the farming community fail to reach agreement, and go to the FCC2 to resolve a conflict. (They have nowhere else to go, since the courts defer to the FCC2 as an expert agency to decide what "too salty" means in a particular case.)

Neither side can predict what the outcome of the FCC2's deliberations will be, since it doesn't always decide the merits of individual cases in isolation. It has many proceedings before it at any given time; for example, the FCC2 might be pushing the farmers to get organic certification, and negotiating with the city about the rezoning of agricultural land for urban development. The solution the FCC2 negotiates between the city and the farmers might encompass all these other matters, not only making the result of the salinity dispute unpredictable, but failing to establish a precedent that others might use later.

A better approach would be for the FCC2 to regulate the resulting salinity in water leaving the city (to, say, 5 ppm), remove any mention of "too salty" from its regulations, and provide a way for contending parties to get a specific case resolved efficiently. It might give the farmers the right to stop the city water plant releasing water into the river if the salinity exceeds 5 ppm (leading to a negotiated solution, where the city might pay the farmers' coop $300,000 to raise the limit up to 10 ppm in dry months), or if there are too farmers to negotiate with individually it might choose a liability regime (leading to a court-imposed payment of say $30/acre if salinity exceeds 5 ppm and some farmers sue the city).

Who gets the apple?

2010-10-12T16:29:00.000-07:00

I’ve been looking for a metaphor to illustrate the weaknesses I see in the FCC’s “you two just go off and coordinate” approach to solving wireless interference problems among operators.

Let's think of the responsibility to bear the cost of harmful interference as an apple.* It’s as if the FCC says to Alice and Bob, “I've got an apple, and it belongs to one of you. I’m not going to decide which of you should have the apple; you decide among yourselves.”

Now, if Alice were the owner of the apple and valued it at 80 cents, then the answer would simply depend on how much Bob valued having the apple (and rational negotiation, of course). If having an apple was worth 90 cents to him, he’d get it for some price between 80 and 90 cents; if it was worth only 60 cents to him, Alice would keep it. Problem solved.

Trouble is, the FCC doesn’t tell them who actually owns the apple, and even if it did, it doesn’t tell them whether it’s a Granny Smith or a Gala. The odds of Alice and Bob coming to an agreement without going back to the FCC is slim.

The analogy: The FCC’s rules often don’t make clear who’s responsible, in the end, for solving a mutual interference problem (i.e. who owns the apple); and it’s impossible to know short of a rule making by the FCC what amounts to harm (i.e. what kind of apple it is).

-----------------------
* There's always interference between two nearby radio operators (near in geography or frequency). While the blame is usually laid on the transmitter operator, it can just as reasonably be placed on the receiver operator for not buying better equipment that could reject the interference.

Social network visualizations - an online symposium

2010-07-02T14:39:00.000-07:00

My work on the evolution of FCC lobbying coalitions has been accepted in the JoSS (Journal of Social Structure) Visualization Symposium 2010 (link to my entry). Jim Moody of Duke has done a wonderful job collecting a dozen visualizations of social networks. Each is worth exploring; in particular, see the thoughtful comments that the JoSS staff provided to each entry in order to stimulate debate.

How I Learned to Stop Worrying and Love Interference

2010-06-07T12:27:00.000-07:00

(With apologies to Stanley Kubrick.)

Radio policy is fixated on reducing or preventing harmful interference. Interference is seen as A Bad Thing, a sign of failure. This is a glass-half-empty view. While it is certainly a warning sign when a service that used to work suddenly fails, rules that try to prevent interference at all costs lead to over-conservative allocations that under-estimate the amount of coexistence that is possible between radio systems.

The primary goal should not be to minimize interference, but to maximize concurrent operation of multiple radio systems.

Minimizing interference and maximizing coexistence (i.e. concurrent operation) are two ends of the same rope. Imagine metering vehicles at a freeway on-ramp: if you allow just one vehicle at a time onto a section of freeway, people won’t have to worry about looking out for other drivers, but very few cars would be able to move around at one time. Conversely, allowing everybody to enter at will during rush hour will lead to gridlock. Fixating on the prevention of interference is like preventing all possible traffic problems by only allowing a few cars onto the freeway during rush hour.

Interference is nature’s way of saying that you’re not being wasteful. When there is no interference, even though there is a lot of demand, it’s time to start worrying. Rather than minimizing interference with the second-order requirement of maximizing concurrent operation, regulation should strive to maximize coexistence while providing ways for operators to allocate the burden of minimizing interference when it is harmful.

I am developing a proposal that outlines a way of doing this. Here are some of the salient points that are emerging as I draft my ISART paper:

The first principle is delegation. The political process is designed to respond carefully and deliberatively to change, and is necessarily slower than markets and technologies. Therefore, regulators should define radio operating rights in such a way that management of coexistence (or equivalently, interference) is delegated to operators. Disputes about interference are unavoidable and, in fact, a sign of productively pushing the envelope. Resolving them shouldn’t be the regulator’s function, though; parties should be given the means to resolve disputes among themselves by a clear allocation of operating rights. This works today for conflicts between operators running similar systems; most conflicts between cellular operators, say, are resolved bilaterally. It’s much harder when dissimilar operations come into conflict (see e.g. my report (PDF) on the Silicon Flatirons September 2009 summit on defining inter-channel operating rules); to solve that, we need better rights definitions.

The second principle is to think holistically in terms of transmission, reception and propagation; this is a shift away from today’s rules which simply define transmitter parameters. I think of this as the “Three P's”: probabilistic permissions and protections.

Since the radio propagation environment changes constantly, regulators and operators have to accept that operating parameters will be probabilistic; there is no certainty. The determinism of today’s rules that specify absolute transmit power is illusory; coexistence and interference only occur once the signal has propagated away from the transmitter, and most propagation mechanisms vary with time. Even though US radio regulators seem resistant to statistical approaches, some of the oldest radio rules are built on probability: the “protection contours” around television stations are defined in terms of (say) a signal level sufficiently strong to provide such a good picture at least 50% of the time, at the best 50% of receiving locations. [1]

Transmission permissions of licensee A should be defined in such a way that licensee B who wants to operate concurrently (e.g. on nearby frequencies, or close physical proximity) can determine the environment in which its receivers will have to operate. There are various ways to do this, e.g. the Australian “space-centric” approach [2] and Ofcom’s Spectrum Usage Rights [3]. These approaches implicitly or explicitly define the field strength resulting from A’s operation at all locations where receivers might be found, giving operator B the information it needs to design its system.

Receiver protections are declared explicitly during rule making, but defined indirectly in the assigned rights. When a new allocation is made, the regulator explicitly declares the field strength ceilings at receivers that it intends to result from transmissions. In aggregate, these amount to indirectly defined receiver protections. Operators of receivers are given some assurance that no future transmission permissions should exceed these limits. (Such an approach could have prevented the AWS-3 argument.) However, receivers are not directly protected, as might be the case if they are given a guaranteed “interference temperature”, nor is there a need to regulate receiver standards.

While this approach has been outlines in terms of licensed operation, it also applies to unlicensed. Individual devices are given permissions to transmit that are designed by regulator to achieve the desired aggregate permissions that would otherwise be imposed on a licensee. Comparisons of results in the field with these aggregate permissions will be used as a tripwire for changing the device rules. If it turns out that the transmission permissions are more conservative than required to achieve the needed receiver protections, they can be relaxed. Conversely, if the aggregate transmitted energy exceeds the probabilistic limits, e.g. because more devices are shipped than expected or they’re used more intensively, device permissions can be restricted going forward. This is an incentive for collective action by manufacturers to implement “politeness protocols” without regulator having to specify them.

Notes

[1] O’Connor, Robert A (1968) Understanding Television’s Grade A and Grade B Service Contours, IEEE Transactions on Broadcasting, Vol. 47, No. 3, September 2001, p. 309, http://dx.doi.org/10.1109/11.969381

[2] Whittaker, Michael (2002) Shortcut to harmonization with Australian spectrum licensing, IEEE Communications Magazine, Vol. 40, No. 1. (Jan 2002), pp. 148-155, http://dx.doi.org/10.1109/35.978062

[3] Ofcom (2007) Spectrum Usage Rights: A statement on controlling interference using Spectrum Usage Rights, 14 December 2007, http://www.ofcom.org.uk/consult/condocs/surfurtherinfo/statement/statement.pdf

Why I give service

2010-05-31T18:28:00.000-07:00

I have just returned from working in the kitchen during a course at the Northwest Vipassana Center. During one of the breaks I had a fascinating conversation with one of the center managers, who it turns out experiences service very differently from me. She asked that I record my thoughts, and this is what I came up with.

Serving is no fun – for me, at least. Serving a course is about stress, anxiety and fatigue, with a few happy moments when I wish the meditators well as I pass them by. There’s no joy in doing the work, as there is for some, and no joyful release at the end; only relief that it’s over. It’s pretty much like sitting a course, with the difference that I’m just banging my head against a wooden wall, not a brick one.

So why do I do it?

I do it because I think it’s good for me. Working in the kitchen amplifies my weaknesses, and makes it easier to see when and where I’m being unskillful. I come face-to-face with my frailties and failings, and hopefully end the course with another sliver of wisdom.

I do it because serving is a middle ground between sitting practice and living in the real world. Like developing any skill - think about playing a musical instrument - meditation requires hours of solitary practice every day, over decades. However, that practice is only the means to an end, which is to live better with, and for, others. Serving on a course helps me try out the skills I’m learning in a realistically stressful but safe environment. Things can’t spin too far out of control; I’m back on the cushion every few hours, with an opportunity to reboot and start again. And I’m surrounded by people of good will, with direct access to teachers if I need it.

And I do it to repay, in small part, the debt I owe to all those people whose service have made it possible for me to learn this technique, and sit courses. I was able to sit because someone else was in the kitchen; now it’s my turn.

Improving FCC filing metadata

2010-05-12T10:45:00.000-07:00

On 10 May 2010 I filed a comment on two FCC proceedings (10-43 and 10-44, if you must know) concerning ways to improve the way it does business. I argued that transparency and rule-making efficiency could be improved by improving the metadata on documents submitted to the Electronic Comments Filing System (ECFS).

I recommended that the FCC:

Associate a unique identifier with each filer
Require that the names of all petitioners are provided when submitting ECFS metadata
Improve RSS feed and search functionality
Require the posting of digital audio recordings of ex parte meetings
Provide a machine interface for both ECFS search and submission

Opt-in for Memory

2010-05-12T10:16:00.000-07:00

The Boucher-Stearns privacy measure tries to do many things (press release; May 3 staff discussion draft); too many, according to Daniel Castro at ITIF.

One of the issues it doesn’t tackle – and legislation may or may not be the solution – is the persistence of digital information once it has been collected.

In a NY Times context piece called Tell-All Generation Learns to Keep Things Offline, Laura Holson writes that members of the “tell-all generation” are becoming more picky about what they disclose. There’s growing mistrust of social networking sites, and young people keep a closer eye on their privacy settings than oldsters. Holson reports on a Yale junior who says he has learned not to trust any social network to keep his information private, since “If I go back and look, there are things four years ago I would not say today.”

I expect that this concern will grow beyond information collection to encompass retention. (That's already a big concern of law enforcement, of course.) Explicit posts (photos, status updates) will live forever, if for no other reason than sites like the Internet Archive. However, the linkages that people make between themselves and their friends, or themselves and items on the web, are less explicit – and probably more telling. These links are held by the social network services, and I expect that there will be growing pressure on them to forget these links after some time. Finally, there are the inferences that companies make from these links and other user behavior; their ownership is more ambiguous, since they’re the result of a third party’s observations, not the subject’s actions.

My bet is that norms will emerge (by market pressure and/or regulation) that force companies to forget what they know about us. For the three categories I noted above, it might work something like this:

Posts: Retained permanently by default. Explicit user action (i.e. an opt-out) required for it to be deleted
Linkages: Deleted automatically after a period, say five years. User has to elect to have information be retained (opt-in).
Inferences: Deleted after a period, say five years, if user opts out; otherwise kept. This one is tricky; I can also see good reasons to make deletion automatic with an opt-in for retention.

However these practices evolve, it’s become clear to me that neither the traditional “notice and choice” regime nor the emerging “approve use” approach are sufficient without a mechanism for forgetting.

New Ethics as a Second Language

2010-05-11T09:00:00.000-07:00

In lecture 27 of the Teaching Company course on Understanding the Brain, Jeanette Norden observes that we seem to learn morality using the same mechanisms we use for learning language.

Newborns can form all the sounds used in all the languages on the planet, but with exposure to their mother tongue they become fluent in a subset. It eventually becomes almost impossible to form some of unused sounds, and the idiosyncrasies of their language seem natural and universal.

This makes me wonder about the difficulties an immigrant might have in learning the peculiarities of a new culture. I’ve definitely been confounded from time to time by unexpected variations in “the right thing to do” – and there’s really very little difference between the culture I grew up in and the ones I moved to as an adult. “Culture shock” may not just be language and customs; it probably involves morality, too, since every system of ethics is a mixture of universals and particulars.

Of course, that’s not to say that one cannot become fluent in an alternative morality. It might just be harder than a native “moralizer”, particularly one who has never had to learn "ethics as a second language”, might assume.

And traditionalists around the world who claim that wall-to-wall American media “corrupt the morals of our youth” are probably right: I'd guess young people pick up the ethical biases of American culture by watching movies and TV even more easily than they pick up English.

Negotiating the Price of Privacy

2010-05-10T09:06:00.000-07:00

Kurt Opsahl at EFF’s time line of changes to Facebook’s privacy policies over the last 5 years tells me a story of a shifting power balance. (Thanks to Peter Cullen for the link.)

It’s a quick read, but in a nutshell: in 2005, the user controlled where information went. By December 2009, Facebook considered some information to be publicly available to everyone, and exempt from privacy settings.

I vaguely remember Esther Dyson describing privacy more than two decades ago as a good users would trade. That’s how I read the time line. It’s an implicit negotiation between Facebook and its users over the value of personal information (let’s call it Privacy, for short) vs. the value of the service Facebook provides (call it Service).

In the early days, the service had few users, and the network effect hadn’t kicked in. Facebook needed users more than they needed Facebook, and so Facebook had to respect privacy – it was worth more to users than the Facebook service was:

Service << Privacy

Since the value of a social network grows exponentially as the number of members increases, the value of the service S grew rapidly as membership increased. A user’s perception of the value of privacy didn’t change much; it probably grew a little, but not exponentially. Probably sometime around 2008, the value of the service started overtaking the value of privacy:

Service ≈ Privacy

Facebook’s hard-nosed approach to privacy (or lack of it) makes clear that it now has the upper hand in the negotiation. An individual user needs Facebook more than vice versa:

Service > Privacy

One take-away from this story is that the privacy settings users will accept are not a general social norm, but the result of an implicit negotiation between the customer and supplier. When a supplier becomes indispensable, it can raise its prices, in explicitly ($$) or implicitly (e.g. privacy conditions). Other services therefore should not assume that they can get away with Facebook’s approach. They can make virtue of necessity by offering better privacy protection – at least until the day when their service is so valuable that they, too, can change the terms.

Non-privacy goes non-linear

2010-04-29T14:52:00.000-07:00

I’ve never been able to “get” Privacy as a policy issue. Sure, I can see that there are plausible nightmare scenarios, but most people just don’t seem to care. What a company, or a government, knows about one just doesn’t rate as something to worry about. Perhaps the only angle that might get the pulse racing is identity theft; losing money matters. But no identity theft stories have inflamed the public’s imagination, or mine.

The recent spate of stories about privacy on social networking sites have led me to reconsider – a little. I still don’t think Joe Public cares, but the technical and policy questions of networked privacy intrigue me more than the flow of personal information from a citizen to an organization and its friends.

The trigger for the current round of privacy worries was the launch of Google Buzz. Good Morning Silicon Valley puts it in context with Google, Buzz and the Silicon Tower, and danah boyd’s keynote at SXSW 2010 reviews the lessons and implications.

Mathew Ingram’s post Your Mom’s Guide to Those Facebook Changes, and How to Block Them alerted me to the implications of Facebook’s “Instant Personalization” features.

Woody Leonhard’s article Hotmail's social networking busts your privacy showed that Google and Facebook aren’t the only ones who can scare users about what personal information is being broadcast about them.

I think there may be a profound mismatch between the technical architectures of social networking sites, and the mental model of users.

This mismatch is an example of the “hard intangibles” problem that I wrestled with inconclusively a few years ago: our minds can’t effectively process the complexity of the systems we’re confronted with.

Two examples: attenuation and scale.

We assume that information about us flows more sluggishly there further it goes. My friends know me quite well, their friends might know me a little, and the friends-of-friends-of-friends are effectively ignorant. In a data network, though, perfect fidelity is maintained no matter how many times information is copied. We therefore have poor intuition about the fidelity with which information can flow away from us across social networks.

It’s a truism that the mind cannot grasp non-linear growth; we’re always surprised by the explosion of compound interest, for example. On a social network, the number of people who are friends-of-friends-of-…-of-friends grows exponentially; but I would bet that most people think it grows only linearly, or perhaps even stays constant. Thus, we grossly underestimate the number of people to whom our activities visible.

My most recent personal experience was when I noticed a new (?) feature on Facebook two days ago. One of my friends had commented on the status update of one of their friends, who is not in my friend network. Not only did I see the friend-of-my-friend’s update; I saw all the comments that their friends (all strangers to me) had made. I’m pretty sure the friends-of-my-friend’s-friend had no idea that some stranger at three removes would be reading their comments.

If you find the “friends-of-my-friend’s-friend” construct hard to parse, then good: I made it on purpose. I suspect that such relationships are related to the “relational complexity” metrics defined by Graeme Halford and colleagues; Halford suggests that our brains max out at around four concurrent relationships.

I’m pretty confident that the Big Name Players all just want to do right by their users; the trouble is that the social networks they’re building for us are (of necessity?) more complicated than we can handle. It hit home when I tried to grok the short blog post Managing your contacts with Windows Live People. I think I figured it out, but (a) I’m not sure I did, and (b) I'd rather not have had to.

Knowing with the Body

2010-04-24T01:52:00.000-07:00

The neurological patient known as Emily cannot recognize the faces of her loved ones, or even herself in a mirror. [1] She doesn’t have conscious awareness that she knows these people, but her body does. When she is shown a series of photos of known and unknown people, she cannot tell them apart; however, the electrical conductance of her skin increases measurably when she’s looking at the face of someone she knows. [2] It’s not that she’s lost the ability to recognize people in general; she can still recognize her family, and herself, by their voices.

Damasio notes that skin-conductance responses are not noticed by the patient. [3] However, perhaps someone who is well-practiced in observing body sensations – for example, a very experienced meditator in the Burmese vipassana tradition [4] – would be able to discern such changes. I suspect so; in which case, a patient like Emily would be able to work around their recognition problem by noting when their skin sensations change. It’s known that they use workarounds; Emily, for example, “sits for hours observing people’s gaits and tries to guess who they are, often successfully.” [5]

Both Damasio’s theory of consciousness and Burmese vipassana place great importance on the interactions between body sensations and the mind. As I understand Damasio, he proposes that consciousness works like this:

1. The brain creates representations (he calls them maps or images) of things in the world (e.g. a face), and of the body itself (e.g. skin conductance, position of limbs, state of the viscera, activity of the muscles, hormone levels).

2. In response to these representations, the brain changes the state of the body. For example, when it sees a certain face, it might change skin conductance; when it discerns a snake it might secrete adrenaline to prepare for flight, tighten muscle tone, etc. Damasio calls these responses “emotions”, which he defines as “a patterned collection of chemical and neural responses that is produced by the brain when it detects the presence of an emotionally competent stimulus — an object or situation, for example.” [6]

3. In a sufficiently capable brain (which is probably most of them) there is a higher order representation that correlates changes in the body with the object that triggered these changes. This second-order map is a feeling, in Damasio’s terminology: “Feelings are the mental representation of the physiological changes that characterize emotions.” [6], [7] Feelings generate (or constitute – I’m not sure which…) what he calls the “core self” or “core consciousness”.

Since I find pictures to be helpful, I’ve created a short slide animation on SlideShare that shows my understanding of this process; click on this thumbnail to go there:

In Emily’s case, steps 1 and 2 function perfectly well, but the correlation between a face and changes in the body fails in step 3. The higher-order correlation still works for voices and body changes, however, since she can recognize people by their speech.

More acute awareness of body sensation might not just help clinical patients like Emily. In the famous Iowa Gambling Task, Damasio and Antoine Bechara showed that showed that test subjects were responding physiologically (again, changes in skin conductance) to risky situations long before they were consciously aware of them. A 2009 blog post by Jonah Lehrer includes a good summary of the Iowa Gambling Task and its results. Lehrer reports new research indicating that people who are more sensitive to “fleshy emotions” are better at learning from positive and negative experiences.

NOTES

[1] This post is based on material in Antonio Damasio’s book The Feeling of What Happens: Body and Emotion in the Making of Consciousness (Harcourt 1999). Emily’s case is described on p. 162 ff. I also blogged about this topic in 2005 after reading “Feeling” for the first time.

[2] Damasio op. cit. [1], p. 300

[3] Ibid.

[4] For example, vipassana as taught by S N Goenka. Other mindfulness meditation traditions also attend to body sensations (see e.g. Phillip Moffitt, “Awakening in the Body”, Shambhala Sun, September 2007) but the Burmese tradition places particular emphasis on it.

[5] Damasio op. cit. [1], p. 163

[6] Damasio, A. (2001) "Fundamental feelings", Nature 413 (6858), 781. doi:10.1038/35101669. Note that Damasio’s definition of “emotion” is narrower than usual usage, which refers to affective states of consciousness like joy, sorrow, fear, or hate. Damasio limits himself to the physiological changes which are more typically considered to be an accompaniment to, or component of, these mental agitations.

[7] Feelings so defined seem to correspond to what S. N. Goenka, a well-known teacher in the Burmese vipassana tradition, calls sensations, his preferred translation of the Pāli term vedanā, also often translated as “feelings”. There is some debate about whether vedanā refers just to sensations-in-the body, as Goenka contends, or to any and all pleasant, painful or neutral feelings such as joy, sorrow, etc.

Bill delegates caller ID regulations to FCC

2010-04-16T02:24:00.000-07:00

SiliconValley.com reports that the US House has approved a measure that would outlaw deceptive Caller ID spoofing.

Since I'm currently enamored of a principles-based approach to regulating rapidly changing technology businesses -- that is, policy makers should specify the goals to be achieved, and delegate the means to agents nearer the action -- I'm on the look-out for working examples.

This seems to be one: the bill leaves it up to the FCC to figure out the details of regulation and enforcement.

The FCC itself could delegate further if is so chose, for example by waiting to see if telephone companies come up with effective ways of regulating this problem themselves before trying to devising and imposing its own detailed rules.

Trying to explain the Resilience Principles

2010-03-26T15:36:00.000-07:00

I was honored to participate in a panel in DC on "An FCC for the Internet Age: Reform and Standard-Setting" organized by Silicon Flatirons, ITIF and Public Knowledge on March 5th, 2010. My introductory comments tried to summary the "resilience principles" in five minutes: the video is available on the Public Knowledge event page, starting at time code 02:04:45. The panel starts at around 01:57:00.

The earlier, fifteen minute pitch I gave on a panel on "The Governance Challenges of Cooperation in the Internet Ecosystem" at the Silicon Flatirons annual conference in Boulder on February 1st, 2010 can be found here at time code 01:36:00. My slides are up on Slideshare.net, and a paper is in preparation for JTHTL.

This work is an outgrowth of my TPRC 2008 paper “Internet Governance as Forestry” (SSRN).

Obviating mandatory receiver standards

2010-03-06T13:28:00.000-08:00

Two remarks I heard at a meeting of a DC spectrum advisory committee helped me understand that endless debates about radio receiver standards are the result of old fashioned wireless rights definitions. The new generation of rights definitions could render the entire receiver standards topic moot.

First, a mobile phone executive explained to me that his company was forced to develop and install filters in the receiver cabinets used by broadcasters for electronic newsgathering because it had a “statutory obligation to protect” these services, even though they operated in different frequency ranges.

Second, during the meeting the hoary topic of receiver standards was raised again; it’s long-rehearsed problem that shows no sign of being solved. It’s a perennial topic because wireless interference depends as much on the quality of the receiver as the characteristics of the transmitted signal. A transmission that would be ignored by a well-designed receiver could cause severe degradation in a poor (read: cheap) receiver. Transmitters are thus at the mercy of the worst receiver they need to protect.

A statutory obligation to protect effectively gives the protectee a blank check; for example, the protectee can change to a lousy receiver, and force the transmitting licensee to pay for changes (in either their transmitters or the protectee’s receivers) to prevent interference. This is an open-ended transfer of costs from the protectee to the protector.

The protectors thus dream of limiting their downside by having the regulator impose receiver standards on the protectee. If the receiver’s performance can be no worse than some lower limit, there is a limit on the degree of protection the transmitter has to provide.

The problem with mandatory receiver standards is that it gets the regulator into the game of specifying equipment. This is a bad idea, since any choice of parameters (let alone parameter values) enshrines a set of assumptions about receiver design, locks in specific solutions, and obviates innovation that might solve the problem in new ways. Manufacturers have always successfully blocked the introduction of mandatory standards on the basis that they constrain innovation and commercial choice.

An open-ended statutory obligation to protect therefore necessarily leads to futile calls for receiver standards.

One could moot receiver standards by changing how wireless rights are defined. Rather than bearing an open-ended obligation to protect, the transmitter should have an obligation to operate within specific limits on energy delivered into frequencies other than their own. These transmission limits could be chosen to ensure that adjacent receivers are no worse off than they were under an “open-ended obligation to protect” regime. (The “victim” licensee will, though, lose the option value of being able to change their system specification at will.)

The main benefit is certainty: the recipient of a license will know at the time of issue what kind of protection they’ll have to provide. The cellular company mentioned above didn’t find out until after the auction how much work they would have to do to protect broadcasters since nobody (including the FCC) understood how lousy the broadcasters’ receivers were.

The regulatory mechanisms for doing this are well known, and have been implemented; they include the “space-centric” licensing approach used in Australia (PDF), and Spectrum Usage Rights (SURs) in the UK.

Moving to new rights regimes is a challenging; Ofcom’s progress has been slow. One of the main difficulties is that licensees for new allocations prefer to do things the old, known, way. One of the supposed drawbacks of SURs is that the benefits of certainty seem to accrue a licensee’s neighbor, rather than the new licensee themselves. However, removing the unlimited downside in an open-ended obligation to protect adjacent operations should prove attractive. The whining will now come from the neighbors who will lose their blank check; careful definition of the licensee’s cross-channel interference limits to maintain the status quo should take the sting out of the transition.

Engineers, Commissars and Regulators: Layered self-regulation of network neutrality

2010-02-26T19:13:00.000-08:00

My post Ostrom and Network Neutrality suggested that a nested set of self- or co- regulatory enterprises (Ostrom 1990:90) could be useful when designing regulatory approaches to network neutrality, but I didn’t give any concrete suggestions. Here’s a first step: create separate arenas for discussing engineering vs. business.

One’s immediate instinct when devising a shared regulatory regime (see the list of examples at the end) might be to involve all the key players; at least, that’s what I pointed to in When Gorillas Make Nice. However, I suspect that successful self-regulatory initiatives have to start with a relatively narrow membership and scope: typically, a single industry, rather than a whole value chain. That’s the only way to have a decent shot at creating and enforcing basic norms. Legitimacy will require broadening the list of stakeholder, but too many cooks at the beginning will lead to kitchen gridlock.

Let’s stipulate for now that the key problem is defining what “acceptable network management practices” amount to. Most participants in the network neutrality debate agree that ISPs should be able to manage their networks for security and efficiency, even if there is disagreement about whether specific practices are just good housekeeping or evil rent-seeking.

The engineering culture and operating constraints of different networks are quite distinct: phone companies vs. cable guys; more or less symmetrical last mile pipes; terminating fiber in the home vs. at cabinet; and not least, available capacity in wireline vs. wireless networks. Reconciling these differences and creating common best practices within the network access industry will be hard; that’s the lowest layer of self-regulation. The “Engineers” should be tasked with determining the basic mechanisms of service provision, monitoring compliance with norms, and enforcing penalties against members who break the rules.

The core participants are the telcos (e.g. Verizon, AT&T) and cable companies (e.g. Comcast, Time Warner Cable), in both their wireline and wireless incarnations. Only within a circumscribed group like this is there is any hope of detailed agreement about best practices, let alone the monitoring and enforcement that is essential for a well-functioning self-regulatory organization. Many important network parameters are considered secret sauce; while engineers inside the industry circle can probably devise ways monitor each other’s compliance without giving the MBAs fits, there’s no chance that they’ll be allowed to let Google or Disney look inside their network operating centers.

The next layer of the onion adds the companies who use these networks to deliver their products: web service providers like Google, and content creators like Disney. Let’s call this group the “Commissars”. This is where questions of political economy are addressed. The Commissars shape the framework within which the network engineers decide technical best practices. It’s the business negotiation group, the place where everybody fights over dividing up the rents; it needs to find political solutions that reconcile the very different interests at stake:

The ISPs want to prevent regulation, and be able to monetize their infrastructure by putting their hand in Google’s wallet, and squeezing content creators.
Google wants to keep their wallet firmly shut, and funnel small content creators’ surplus to Mountain View, not the ISPs.
Large content creators want to get everybody else to protect their IPR for them.
New content aggregators (e.g. Miro) want a shot at competing in the video business with the network facility owners.

This is not an engineering argument, and a Technical Advisory Group (TAG) along the lines described by Verizon and Google (FCC filing) would not be a suitable vehicle for addressing such questions. The Commissars are responsible for answering questions of collective choice regarding the trade-offs in network management rules, and adjudicating disputes that cannot be resolved by the Engineers among themselves.

The Engineers can work in parallel to the Commissars, and don’t need to wait for the political economists to fight out questions about rents; in any case, it will be helpful for the Commissars to have concrete network management proposals to argue about. There will be a loop, with the conclusions of one group influencing the other. The Commissars inform the Engineers about the constraints on what would constitute acceptable network management, and the Engineers inform the Commissars about what is practical.

Finally, government actors – call them the “Regulators” – set the rules of the game and provide a backstop if the Engineers and Commissars fail to come up with a socially acceptable solution, or fail to discipline bad behavior. Since the internet and the web are critical infrastructure, governments speaking for citizens are entitled to frame the overall goals that these industries should serve, even though they are not well qualified to define the means for achieving them. Final adjudication of unresolved disputes rests with the Regulators.

References

Ofcom, Initial assessments of when to adopt self- or co-regulation, December 10, 2008,
http://www.ofcom.org.uk/consult/condocs/coregulation/condoc.pdf

Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, 1990

Philip J. Weiser, Exploring Self Regulatory Strategies for Network Management: A Flatirons Summit on Information Policy, August 25, 2008,
http://www.silicon-flatirons.org/documents/publications/summits/WeiserNetworkManagement.pdf

Examples of self- and co-regulatory bodies

The Internet Watch Foundation (IWF) in the UK works to standardize procedures for the reporting and taking-down of abusive images of children. It was established in 1996 by the internet industry to allow the public and IT professionals to report criminal online content in a secure and confidential way. (Ofcom 2008:9, and IWF)

The UK “Classification Framework” for content on mobile phones is provided by the Independent Mobile Classification Body (IMCB) with the aim of restricting young people’s access to inappropriate content. It is the responsibility of content providers to self-classify their own content as “18” where appropriate; access to such content will be restricted by the mobile operators until customers have verified their age as 18 or over with their operator. (Ofcom 2008:9, and IMCB)

The Dutch organization NICAM (Nederlands Instituut voor de Classificatie van Audiovisuele Media) administers a scheme for audiovisual media classification. It includes representatives of representatives of public and commercial broadcasters, film distributors and cinema operators, distributors, videotheques and retailers. (Ofcom 2008:9, and NICAM)

Amateur radio service and frequency coordinators provide examples of self-regulation in spectrum policy. The American Radio Relay League (ARRL) has an understanding with the FCC that it manages the relevant enforcement activities related to the use of ham radio. Only in the most egregious cases will ARRL report misbehavior to the FCC Enforcement Bureau. (Weiser 2008:23)

The Better Business Bureau’s National Advertising Division (NAD) enforces US rules governing false advertising, using threats of referrals to the FTC to encourage compliance with its rules. (Weiser 2008:24, and NAD)

US movie ratings are provided by a voluntary system operated by the MPAA and the National Association of Theater Owners.

Ostrom and Network Neutrality

2010-02-12T13:51:00.000-08:00

My previous post scratched the surface of a self-regulatory solution to network neutrality concerns. While this isn’t exactly a common pool resource (CPR) problem, I find Elinor Ostrom’s eight principles for managing CPRs are helpful here (Governing the Commons: The evolution of institutions for collective action, 1990).

Jonathan Sallet boils them down to norms, monitoring and enforcement, and that’s a good aide memoire. It’s useful, though, to look at all of them (Ostrom 1990:90, Table 3.1):

1. Clearly defined boundaries: Individuals of households who have rights to withdraw resource units from the CPR must be clearly defined, as must the boundaries of the CPR itself.

2. Congruence between appropriation and provision rules and local conditions: Appropriation rules restricting time, place, technology, and/or quantity of resource units are related to local conditions and to provision rules requiring labor, material, and/or money.

3. Collective-choice arrangements: Most individuals affected by the operational rules can participate in modifying the operational rules.

4. Monitoring: Monitors, who actively audit CPR conditions and appropriator behavior, are accountable to the appropriators or are the appropriators.

5. Graduated sanctions: Appropriators who violate operational rules are likely to be assessed graduated sanctions (depending on the seriousness and context of the offense) by other appropriators, by officials accountable to these appropriators, or by both.

6. Conflict-resolution mechanisms: Appropriators and their officials have rapid access to low-cost local arenas to resolve conflicts among appropriators or between appropriators and officials.

7. Minimal recognition of rights to organize: The rights of appropriators to devise their own institutions are not challenged by external governmental authorities.

8. (For CPRs that are parts of larger systems) Nested enterprises: Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises.

Many but not all of these considerations are addressed in the filing and my comments: The headline of section B that “self-governance has been the hallmark of the growth and success of the Internet” reflects #2. My point about involving consumers speaks to #3. The TAGs mooted in the letter address #4 and #6, but not #5. The purpose of the letter is to achieve #7.

In addition to the lack of sanctions, two other key issues are not addressed. Principle #1 addresses a key requisite for a successful co-regulatory approach: that industry is able to establish clear objectives. Given the vagueness of the principles in the filing, it’s still an open question whether the parties can draw a bright line around the problem.

I believe #8 can help: create a nested set of (self- or co-) regulatory enterprises. While I don’t yet have concrete suggestions, I’m emboldened by the fact that nested hierarchy is also a hallmark of complex adaptive systems, which I contend are a usable model for the internet governance problem. Ostrom’s three levels of analysis and processes offer a framework for nesting (1990:53):

Constitutional choice: Formulation, Governance, Adjudication, Modification
Collective choice: Policy-making, Management, Adjudication
Operational choice: Appropriation, Provision, Monitoring, Enforcement

I think the TAGs are at the collective choice level. It would be productive to investigate the institutions one might construct at the other two levels. The FCC could usefully be involved at the constitutional level; even if one doesn't dive into a full-scale negotiated rule-making or "Reg-Neg", government involvement would improve legitimacy (cf. Principle #7). At the other end of the scale, operational choices include mechanisms not just for monitoring (and some tricky questions about disclosure of "commercially confidential" information) but also enforcement. The latter could be as simple as the threat of reporting bad behavior to the appropriate agency, as the Better Business Bureau’s National Advertising Division does (see Weiser 2008:21 PDF).

When Gorillas Make Nice

2010-02-12T13:36:00.000-08:00

Verizon and Google’s recent joint FCC filing about the values and governance of the internet largely echoes the conclusions of a Silicon Flatirons summit in August 2008 (PDF): that self-governing institutions are the best way to manage day-to-day questions of network neutrality, with the government acting as a backstop when market forces and self-regulation fail.

The filing seems to come in two parts: a statement of principles, and a sketch of how self-governance might work. I’ll largely ignore the first part, since clearly Google and Verizon found little to agree on. The three key principles are motherhood (consumer transparency and control), Google’s non-negotiable (openness) and Verizon’s (encouraging investment), respectively; it’s hard to argue with any of this, except to observe that the hard work lies in achieving them simultaneously.

The most useful resource on self-regulation in communications I’ve seen is Ofcom’s 2008 statement on “Identifying appropriate regulatory solutions: principles for analysing self- and co-regulation” (PDF). It concluded that self-regulation is most likely to work when “industry collectively has an interest in solving the issue; industry is able to establish clear objectives for a potential scheme; and the likely industry solution matches the legitimate needs of citizens and consumers.”

If their effort is to succeed, the companies will have to build an institution that represents all interests. Let's stipulate that the three main stakeholder groups are content players, network operators and consumers; Google and Verizon fall in the first two groups. On the network side, they’ll need to add the cable industry (always much more leery of network neutrality than the long-regulated telcos), and on the content side, the studios. The trickiest part will be finding a “consumer voice” with some legitimacy; everybody, not least these companies, claims to have the consumer’s best interest at heart.

The filing is predictably vague about the basis on which government would become involved, and is silent about how its proposed institution would enforce its own norms. That’s a mistake. It’s in the companies’ best interest to declare which sword they want hanging over their heads. If they don’t, there won’t be sufficient incentive to Do the Right Thing in the short term (the CEO will ask, “If I’m not breaking a law, why should I go the extra mile?”), which means that eventually a mountain of punctilious rules will be imposed on them. (It’s my understanding that this is what happened over the last decade with accessibility to the internet for those with disabilities: tech companies promised a decade ago they’d solve the problem, didn’t do all that much, and now Rep. Markey is writing detailed rules.)

It’s not clear to me whether the filing is proposing self- or co-regulation, defined by Ofcom (2008) as follows:

Self-regulation: Industry collectively administers a solution to address citizen or consumer issues, or other regulatory objectives, without formal oversight from government or regulator. There are no explicit ex ante legal backstops in relation to rules agreed by the scheme (although general obligations may still apply to providers in this area).

Co-regulation: Schemes that involve elements of self- and statutory regulation, with public authorities and industry collectively administering a solution to an identified issue. The split of responsibilities may vary, but typically government or regulators have legal backstop powers to secure desired objectives.

I think co-regulation is indicated here. Without a backstop there will not be sufficient incentive for good behavior. Politically, too, the term “self-regulation” has become anathema in Washington DC because the financial melt-down is deemed to have been due to a failure in the same. (Not that it matters, but I think this assessment is incorrect on two counts: self-regulation is only part of a much larger problem in the financial crisis; and even if it weren’t, the lessons learned are not easily transposable to communications policy. Still, it’s probably best to use another term, like shared regulation, supervised delegation or bounded autonomy.)

The internet is not an ecosystem, but…

2010-02-10T15:41:00.000-08:00

The “internet ecosystem” metaphor is ubiquitous; I’ve used it myself, though with some trepidation. I think I can now reconcile why it’s both wrong and useful.

It’s wrong, strictly speaking, since many aspects of the ecosystem-internet mapping are questionable. As I blogged in 2007 about the “business ecosystem” terminology, the validity of the metaphor is undermined by quite a large number of mapping mismatches:

Number: a food web consists of billions of interactions among animals and plants; a business web comprises a relatively small number of companies

Metrics: Biomass a typical rough measure of an ecosystem; does that map to total revenue, profitability, return on investment, or something else?

Topology: An ecosystem is a lossy, one-way energy flow; as each organism is eaten by the next, energy is lost. Business relationships are reciprocal, and generate value.

Time scales: Species change slowly, but companies can change their role in a system overnight through merger, acquisition or divestiture.

Choice: Interactions between firms can be changed by contract, whereas that between species is not negotiable except perhaps over very long time scales by evolution of defensive strategies.

Foresight: Humans are pre-eminent among animals in their ability to anticipate the behavior of other actors, explore counter-factuals, think through What If scenarios, etc. The response of a system containing humans to some change is therefore much more complex than that of a human-free ecosystem. “Dumb” agents in an adaptive system respond to the change; humans respond to how they think other humans will respond to their response to those people’s responses etc.

Goals: Biological systems don’t have goals, but human ones do. There are no regulatory systems external to ecosystems in a state of nature (if such things still exist on this planet), but there are many, such as rule of law and anti-trust, in human markets. Natural processes don’t care about equity or justice, but societies do, and impose them on business systems. If ecosystems were a good model for business networks, there would be no need for anti-trust regulation.

The connotations of the metaphor are also misleading. Ecosystems are often used to connote stability and vibrant self-regulation; in fact, they often suffer catastrophic collapses. Companies are exhorted to invest in their ecosystem with the goal of becoming a keystone species. It’s not clear why they should do so, from the ecosystem perspective: keystone species don’t typically represent a lot of biomass. Their “bottleneck position”, however, is attractive from the perspective of a company that wants to extract rents through market power.

However, the ecosystem concept has gained traction because there is a deeper truth: both the internet and ecosystems are both examples of complex adaptive systems. (A complex adaptive system may be defined as a collection of interacting, adaptive agents; other examples include the immune system, the human body, stock markets, and economies. Note that adaptive systems are often nested.)

Thus, the internet is to an ecosystem as a whale is to an elephant. It could be useful to think in terms of elephants if one has to manage oceans but doesn’t know much about whales, since both are large, social mammals. However, one can just as well explain whales in terms of elephants – and the differences, e.g. living on land vs. in water – can be decisive in some cases.

With this realization, the utility and limitations of using an ecosystem metaphor when thinking about the internet, as I did in my Internet Governance as Forestry paper, have become much clearer to me. Lessons from managed ecosystems can illuminate the dynamics and pitfalls of managing the internet, and principles (such as the Resilience Principles I outlined in my recent talk at Silicon Flatirons; my presentation starts around time code 01:36:00 of the video) derived from one can be applied to the other.

Resilience and Realpolitik

2010-02-08T13:34:00.000-08:00

Resilience is a fashionable meme - rightly so, since it offers an alternative to the "find the efficient optimum" approach to solving problems in political economy. (I would say so, of course; see e.g. my presentation at Silicon Flatirons recently, and my paper on forestry as a metaphor for internet governance.)

As reported by The Economist (A needier era: The politics of global disruption, and how they may change, Jan 28th 2010), a report for the Brookings Institution on international politics in an age of want suggests that Governments should think more in terms of reducing risk and increasing resilience to shocks than about boosting sovereign power. This is analogous to advocating reducing risk and increasing resilience vs. boosting wealth creation in the economy. The reason given is that the new threats are networks (of states and non-state actors) and unintended consequences (of global flows of finance, technology and so on).

I've seen (and propagated) the same memes in the context technology policy: the determining factors are inter-locking networks of agents, and unintended consequences that shift more quickly than legislation.

It's ironic, given my claim that the resilience approach is a counter to neoclassical economics, that the article closes with a Milton Friedman quote...

Spectrum as Roads

2009-12-28T10:37:00.000-08:00

This is about as explicit as the spectrum-as-land metaphor gets:

"Spectrum is the equivalent of our highways," says Christopher Guttman-McCabe, vice president of regulatory affairs for CTIA-The Wireless Association, an industry trade group. "That's how we move our traffic. And the volume of that traffic is increasing so dramatically that we need more lanes. We need more highways." (Joelle Tessler, "Wireless companies want a bigger slice of airwaves", Associated Press, posted to SiliconValley.com 12/28/2009)

And its also as self-serving as they come. What the cellular companies need is data capacity. There are many ways to get that that don't require new radio licenses, notably increasing the density of cell towers and improving antenna technology. But those are more expensive than new licenses, hence the claim that they need "the land".

A music/governance metaphor

2009-12-27T19:12:00.000-08:00

I’m still struggling to find a usable taxonomy for “new methods of governance” for the internet. A conversation with Grisha Krivchenia, a music teacher, prompted this attempt at analogy. Since my knowledge of music and its history is sketchy, any corrective comments would be gratefully received.

Let’s start with a particular musical tradition: harpsichord pieces in the High Baroque. Bach wrote the Goldberg Variations, for example, with a particular instrument and even performer (Goldberg) in mind. The performer has many options, however, regarding tempo and mood. When the same score is played on a different instrument, e.g. the piano, an additional set of choices and opportunities arise.

Same score, different instrument(s)

A score written for one instrument can be played by another one with no change; for example, one can play a flute piece on the oboe. However, figurations that were easy for the intended instrument may be hard for the new one. Some instrument changes require transpositions of notes to a new key, for example playing the flute piece on a clarinet (pitched in C and B-flat, respectively). Even if the notes are the same, the music will be different.

As an example of the music/governance metaphor in action, consider libel. The same laws of defamation apply to web pages just as much as to a paper pamphlet; however, some additional interpretation is required from the judge when applying statute and common law developed for paper to the internet.

A slightly more extensive change comes about when music scored for one ensemble (e.g. strings) is re-arranged for another (woodwinds). Both the individual and blended characters of the instruments differ, and the character of the piece can change quite markedly. A possible analogy is the application 911 requirements for phone access to emergency services to Voice over IP devices. The desired policy result and the requirements in law are the same, but the implementation may have to be different. For example, “911” is actually an area code rather than a phone number, and its implementation in VoIP was debated. Further, 911 calls are delivered to a Public Safety Answering Point (PSAP) determined by the location of the caller – which may not be easy to determine for an internet device.

Once the piano exists, it enables new forms of music. First, performers can radically rethink a piece: Glenn Gould’s Goldbergs, to cite a late example. Second, composers wrote pieces for the piano in ways that were inconceivable in the age of the harpsichord: Liszt and Chopin. An analogy in regulation might be the way in which the Kodak camera prompted the overhaul (or arguably invention) of privacy law. [1] Another one might be the way in which the internet if forcing a rethinking of common carriage rules as they apply to telecommunications carriers. [2]

New compositions, same instruments

However, new approaches to composition can come about without new instruments – the shift to atonal music (i.e. lacking a central key) associated with Berg, Schoenberg and Webern supposedly arose from the “crisis of tonality” in the late nineteenth and early twentieth century . An analogy in communications policy might be the emergence of exclusive-use radio licenses allocated by auctions in wireless regulation: they were prompted by insights from economics (e.g. Coase and the privatization movement more generally) rather than by changes in technology.

New performances

Music can also change purely as a result of changes in performance practice. An unattributed assertion in Wikipedia states that “changes in performance practice made by prominent musicians often reverberated in the playing of many other musicians.” Other candidates for this phenomenon is the use of bel canto in early 20th century opera, the use of a clear declamatory vocal style in the French operatic tradition, and the dramatic increases in the minimum technical accuracy required of performers of classical music. One can see this effect in governance too, particularly where common-law is used; interpretations and precedent are cumulative. An ongoing example is software patents: legal scholar Mark Lemley stated at a Silicon Flatirons conference in March 2009 that over the last three years, courts have fixed most of the problems that have been grist for the software patent debate. I presume there are also fashions in jurisprudence, just as there are in music – but here again my lack of knowledge fails me…

A change in venue also makes a difference. The Wikipedia article on the history of the orchestra suggests that the 18th century change from civic music making where the composer had some degree of time or control, to smaller court music making and one-off performance, placed a premium on music that was easy to learn, often with little or no rehearsal. The results were changes in musical style from the counterpoint of the baroque period to the classical style, and emphasis on new techniques such as notated dynamics and phrasing. I believe that the shift in the stakeholder landscape in telecoms from an insider’s club of a few, large firms and regulators to a global plethora of companies and regulators of all sizes is in the process of changing governance, but we don’t have the luxury of 200 years to discern the key developments.

Tentative conclusions

The analogy of music to governance is as follows:

Composer – policy maker (legislator or regulator with quasi-legislative powers, like the FCC)
Score – law, rule or regulation
Instrument – technology and social context
Performer – judge (or quasi-judicial actor, e.g. the FCC)
Audience – interest groups, stakeholders, citizens, etc.

Discerning the metaphor mapping for governance itself is harder – though no harder than I find understanding what “governance” means, period… Perhaps they’re both just the “meta”: music is the aggregate of all the actors and activities related to making music, and the same for governance. The useful insight for me is that all the elements – composer, score, etc. – are necessary to make music, and likewise for governance. Any focus just on policy makers, or just on regulations, or just on the courts etc. will understate the problem.

In terms of new kinds of music/governance, we see

Changes of instruments (technology) that require only minor changes in the score (law)
Changes that prompt composers (policy makers) to invent new genres (rules), either as a result of new technologies or the internal development of the genre itself
Changes brought about by shifts in performance (judicial) practice

The performers (judges) plays an important creative role; they can change the import of a score (law) by their interpretation in the context of a new instrument (technology). It may be that judges are most influential when the policy makers have not yet caught up with changes in technology – they are making music on new instruments using the old scores.

This short taxonomy focuses on the upstream part of the performance value chain. New kinds of music arise most visibly from new compositions and/or new instruments, but performance and audience play roles in disseminating and validating them. Likewise, new forms of governance need to be enacted by courts and accepted by stakeholders before taking hold; new technology and new law are only part of the picture.

Update 12/28/2009: See the comments for some great thoughts from Jon Sallet about the role of improvisation in music and governance. His conclusion: "In a world of change and uncertainty, discretion is an important tool; discretion that is applied by professionals (like trained musicians), within guidelines (like the old rule against using augmented fourths) but that calls upon the expertise of the composer and the performer both to work, as it were, in harmony."

Footnotes

[1] Robert E. Mensel, ""Kodakers Lying in Wait": Amateur Photography and the Right of Privacy in New York 1885-1915", American Quarterly, Vol. 43, No. 1 (Mar., 1991), pp. 24-45, PDF available.

[2] James V DeLong, “Avoiding a Tech Train Wreck”, The American, May/June 2008

A skeptic’s approach to regulation

2009-12-26T15:15:00.000-08:00

I don’t know.

You don’t know either, even if you’re a lawyer or scholar who’s written confident diagnoses of, and persuasive curative prescriptions for, various policy problems.

If you’re a regulator, you know you don’t know.

Decision makers have always operated in a world of complexity, contradiction and confusion: you never have all the information you’d like to make a decision, and the data you do have are often inconsistent. It is not clear what is happening, and it is not clear what to do about it. What’s most striking about the last century is that policy makers seem to have been persuaded by economists that they have more control, and more insight, than they used to.

We have less control over the world than we’d like. We are either confronted by unwanted situations we cannot prevent, or desired situations are precluded. We would like to prevent unwanted situations, but can’t; or we would like favorable circumstances to continue, but they don’t.

There is a small part of the world where the will has effective control; for the rest, one has to deal with necessity, i.e. circumstances that arise whether you will or no. Science and technology since the Enlightenment has dramatically widened our scope of control; economics has piggy-backed on the success of classical physics to make large claims about its ability to explain and manage society. However, this has had the unfortunate consequence that we no longer feel comfortable accepting necessity. If a situation is avoidable – say, postponing the moment of death through a medical intervention – then it becomes tempting to think that when it comes, someone or something can be held responsible.

As Genevieve Lloyd tells it (and I understand it) in Providence Lost (2009), our culture opted to follow Descartes in his framing of free will: we should do the best we can, and leave the rest to divine Providence, which provides a comforting bound to our responsibilities. In the absence of providence, however, we have no guidance on how to deal with what lies beyond our control. As Lloyd puts it, “the fate of the Cartesian will has been to outlive the model of providence that made it emotionally viable.” She argues that Spinoza’s alternative account of free will, built on the acceptance of necessity, is better suited to our time; there is freedom in how we shape our lives in the face of necessity, and a providential deity is not required.

Our Cartesian heritage can be seen in the response to the financial collapse of recent years: someone or something had to be responsible. If only X had done Y rather than Z… but an equally plausible account is that crises and collapse are inevitable; it was only a matter of time.

I submit that the best response to an uncertain and ever-changing world is to accept it and aim at resilience rather than efficiency. Any diagnosis and prescription should always be provisional; it should be made in the knowledge that it will have to be changed. Using efficiency as the measure of a solution, as neoclassical economics might, is the mark of the neo-Cartesian mind: it assumes that we have enough knowledge of the entire system to find an optimum solution, and that we have enough control to effectuate it. In fact, an optimum probably doesn’t exist; if it does exist, it’s probably unstable; and even if a stable solution exists, we have so little control over the system that we can’t implement it.

The best conceptual framework I’ve found for analyzing problems in this way is the complex systems view, and the most helpful instantiation is the approach to managing ecosystems encapsulated in C. S. Holling’s “adaptive cycle” thinking. (See e.g. Ten Conclusions from the Resilience Project). The adaptive cycle consists of four stages: (1) exploitation of new opportunities following a disturbance; (2) conservation, the slow accumulation of capital and system richness; (3) release of accumulation through a crisis event – cf. Shumpeter’s creative destruction; and (4) reorganization, in which the groundwork for the next round is laid.

Two techniques seem to be particularly helpful in applying this approach to governance: simulation and common law. Simulation and modeling exploit the computing power we now have to explore the kinds of outcomes that may be possible given a starting point and alternative strategies; it gives one a feel for how resilient or fragile different proposed solutions may be. Simulation may also help understand outcomes; for example, Ofcom uses modeling of radio signal propagation rather than measurement to determine whether licensees in it Spectrum Usage Rights regime are guilty of harmful interference with other licensees. (See e.g. William Webb (2009), “Licensing Spectrum: A discussion of the different approaches to setting spectrum licensing terms”.)

A common law approach helps at the other end of the process: Jonathan Sallet has argued persuasively that common-law reasoning is advantageous because it is a good way of creating innovative public policies, and is a sensible method of adapting government oversight to changing technological and economic conditions.

But I could be wrong…

Update 12/28/2009: See the fascinating comments from Rich Thanki, below. He takes two salient lessons from complexity theory: avoid monoculture, and develop rules of thumb. He also provides more of the usual quick Keynes quote about "slaves of some defunct economist."

Hard consequences of the soft revolution

2009-12-24T11:43:00.000-08:00

What characteristics (if any) of 21st century communications justify a change in methods of governance?

Any change in policy has unintended consequences; some of them will be adverse. One has to think carefully before advocating radical change: the benefits of change or the costs of doing nothing should be substantial. One way of beginning a cost/benefit analysis is to understand the underlying forces.

Many arguments have been given for new internet regulation. Cowhey and Aronson (Transforming Global Information and Communication Markets 2009:17) cite three factors that will force change: the modular mixing and matching of technology building blocks; the need to span traditional policy and jurisdictional divides (aka Convergence); and the need to rely more on non-governmental institutions to coordinate and implement global policy. In my paper “Internet Governance as Forestry”, I cite three characteristics of the internet that require new responses: modularity, decentralized self-organization, and rapid change.

Let’s consider, then, the following candidates for radical, unprecedented and transformational change in the internet economy taken from these two lists: modularity, convergence, the “third sector”, decentralization, and rate of change.

Modularity

I doubt modularity will persist as a characteristic of the internet business. While it is clearly a hallmark of our current stage, it has a long history: the standardization of interchangeable parts is dated back to Eli Whitney’s process for manufacturing muskets for the US government in 1798, but there is evidence for standardization of arrowheads and uniform manufacturing techniques in the bronze age, and some anthropologists claim there was standardization of stone age tools. However, modular technology does not lead inescapably to a modular industry structure. Standard parts have not rendered pre-internet industries immune to anti-trust problems, and it is likely they will do so now. The role of modularity in the relationships between companies waxes and wanes, depending on rather than driving industry consolidation and market power.

Convergence

The good old convergence argument is a true enough, but tired. The mixing of broadcasting, telecom and intellectual property regulation brought about by common digital formats will undoubtedly require a huge amount of creative reform of regulation, but I no longer think that the result will be the abolition of regulatory categories based on the commercial and technological status quo.

I would very much like to see such an abolition; I proposed a re-organizing the FCC by policy imperatives rather than industry categories in my FCC Reform paper, but I don’t think it’s going to be practical. The human rage to classify [1] will reassert itself. Classifying by policy concern probably won’t work, sad to say, because of how regulation tends to work: take a new problem, fit it into an existing category, and apply the rules of that category. Even if this mechanism yields weird results in times of transition, it’s usually efficient and is likely to persist, even as categories change. We don’t yet have the new categories, but they may well emerge based more on how industry self-organizes than by logic. Judging by today’s behemoths, they might perhaps be networks, cloud services, devices and content (i.e. AT&T, Google/Microsoft, Apple/Dell and Hollywood) replacing broadcasting, telecom, cable and intellectual property (ABC/CBS/NBC, the old AT&T, Comcast and Hollywood).

Decentralization

The internet is no doubt much more decentralized than its forebears, e.g. the telephone network; it is by definition an affiliation of many networks, and a lot of processing is done “at the edges” rather than “in the middle”. There is a linkage between a decentralized architecture and modularity. Modularity allows decentralization, and is amplified by it. If or when either regresses to the mean, the other will tend to do so as well. Since I don’t believe that a high and increasing amount of modularity is an persistent attribute of the 21st century communications industry, I don’t believe that high and increasing decentralization is either. However, the current degree of modularity and decentralization in has probably put us into a qualitatively different regime; there has phase change, so to speak. The polity has just begun to work through the implications, and this will take a decade or more.

The “third sector”: Non-Governmental Institutions (NGOs), non-profits and civil society

Cowhey and Aronson’s interest in NGOs is based in trade, and the organizations they have in mind (ICANN, W3C, IETF) meet the four-part definition offered by Lester Salamon, a political scientist and scholar of US non-profits at Johns Hopkins: they are organizations, i.e., they have an institutional presence and structure; they are private, i.e., they are institutionally separate from the state; they are fundamentally in control of their own affairs; and membership/support is voluntary. Salamon argues that the prominence of NGOs represent an “associational revolution”. I cannot judge whether this phenomenon is transient or not; however, the large organizations clearly provide an alternative venue for governance. For example, Cowhey and Aronson argue that the IETF’s central role in internet standards came about because the US Government decided to delegate authority to it.

If one relaxes the requirement for formal institutional structure, the rise of private, voluntary engagement in politics facilitated by Web 2.0 represent an impetus and perhaps even a venue for new governance. Currently fashionable examples include http://transparencycorps.org/, http://opengov.ideascale.com/ and http://watchdog.net/; tools that facilitate engagement include http://www.opencongress.org/, http://www.opensecrets.org/lobbyists/ and http://www.govtrack.us/. The citizen’s ability to know about the activities of their legislators and petition has never been greater; tools for organizing into ad hoc coalitions (most famously the role of http://www.meetup.com/ in the 2004 and 2008 US campaigns) lead to a ferment of groups that may grow into more recognizable institutions. Policy makers will have to invent new ways to track and mollify these groups, at the very least; the Obama Administration appears to be using them to support policy making.

While the decentralized architecture of the internet and the rise of NGOs are different phenomena with different causes, Web 2.0 technologies are beginning to draw them together.

Rate of change

As to whether the rapidity of change is transformative and permanent, I think the answer is No and Yes. The rate of technical and commercial innovation on internet over the last two decades has been stunning. It has been abetted by modularity, and even more so by the ability of software to morph without having to retool a factory. (Retooling a code base is a non-trivial exercise, though.) However, the internet is growing up and it’s reasonable to expect that the industry and technology will settle into a phase of relative maturity. [2]

On the other hand, while the rate of change may not continue to accelerate, or even continue at its current pace, the political system has to adjust to the stresses that the increase to date has already imposed. William Scheuerman, for example, argues that the “social acceleration of time” has created a profound imbalance between the branches of government in liberal democratic systems like the US. [3] Even if the rate of techno-commercial innovation slows down, the rate at which global markets generate and propagate news will be a challenge for political systems whose time cycles are set in constitutions that change only very slowly, and human physiology which changes hardly at all. [4]

Back to Hard Intangibles

A change in context that forces a change in governance doesn’t need to be irreversible for the consequences to be profound. Since history is cumulative, a “phase change” in policy making is a change that never really reverts to its prior form, since the context changes with it. However, some changes are more portentous than others. I’ve argued above that the modularity, convergence and decentralization of the internet are temporary, and part of the regular cycle flow in industry structure. Changes in tempo and the rise of the third sector seem to me to be more momentous. I think both are rooted in the growing intangibility of our societies, which has been accelerated by ICT: complex software running on powerful processors linked by very fast networks.

I think there is a link back to my 2006/2007 obsession with “hard intangibles” (DeepFreeze9 thread). The ability to compose more components than the mind can manage makes programming/debugging very hard, particularly when those components are so easily mutable: it’s easier to change a line of code than to retool an assembly line. The “soft products” of these technologies, themselves complex, composable and mutable become the inputs for culture and thus policy making: it’s easier to change web artifacts and social networks than to manage a movement using letters and sailing ships.

Footnotes

[1] I first heard the term used by Rohan Bastin, Associate Professor of Anthropology at Deakin University, in a Philosopher’s Zone interview about Claude Levi-Strauss. “The human rage to classify” is also a chapter title in F. Allan Hanson, The Trouble With Culture : How Computers Are Calming The Culture Wars, SUNY Press 2007

[2] This prediction contradicts Ray Kurzweil’s contention that technological change accelerates at an exponential rate, and will continue to do so: his “Law of Accelerating Returns” [link, critique]

[3] William E. Scheuerman, Liberal Democracy and the Social Acceleration of Time (2004). Scheuerman defines social acceleration of time as “a long term yet relatively recent historical process consisting of three central elements: technological acceleration (e.g. the heightening of the rate of technological innovation), the acceleration of social change (referring to accelerated patterns of basic change in the workplace, e.g.), and the acceleration of everyday life (e.g., via new means of high-speed communication or transportation).” I’m indebted to Barb Cherry for introducing me to Scheuerman’s ideas; see e.g. her “Institutional Governance for Essential Industries Under Complexity: Providing Resilience Within the Rule of Law” CommLaw Conspectus 17.1

[4] Human thinking won’t speed up much, if at all – though tools can make it look as if it does. See for example the Edwin Hutchins’ wonderful Cognition in the Wild (1996). Hutchins contends that we need to think in terms of “socially distributed cognition” in a system that comprises people and the tools that were made for them by other people.

Objects of governance: From things to behaviors

2009-12-21T15:24:00.000-08:00

In spite of our penchant for abstraction, we think best in concrete terms. That means we prefer to think about things rather than processes, including when it comes to communications regulation. The growing intangibility of our world is making this harder to do, however.

The legal scholar William Boyd introduced me the concept of an “object of governance”, i.e. the explicit focus or nominal topic of regulatory activity. [1] Boyd is concerned with deforestation as an object of climate governance [2]; a quick web search throws up examples like organized crime, “The East”, the Sahel, and risk. Objects of communications regulation include personally identifiable information (PII), spectrum, phone service, and the internet.

While most of these objects are intangible, they are at least to some extent thing-like; they’re nouns. It becomes more tricky when regulation addresses behavior – that is, verbs. I’ll work through a few examples in communications regulation where the object of governance started off as a thing/noun, and is becoming a behavior/verb:

Privacy: From PII to Use

The current approach to protecting privacy on the web is rooted in the notion of data security: information exists somewhere, and needs to be protected. However, an alternative conception based on appropriate use rather than access restrictions is emerging. [3] [4] The idea is that the tradition Notice & Choice regime is complemented by use-and-obligations model where organizations disclose the purposes to which they intend to put information, and undertake to limit themselves to those uses.

Wireless regulation: From spectrum to radio operation

Radio regulation has been framed in terms of government management of a “spectrum asset” for many decades. Even though in practice the regulations concerned themselves with the operating parameters of transmitters, the idea that some underlying asset existed has been a useful fiction, particularly as the detailed technology and service choices have been increasingly privatized through auctions of general-use licenses.

However, a new generation of radio technologies has been used to call this approach into question. “Open Spectrum” advocates have argued that dynamic wireless technologies obviate many underlying assumptions of current regulation, and prefer “commons” access over exclusive licenses. [5] Some in the RF engineering community recommend that regulation take into account dynamic adaptation at all layers in the network stack, not just at the radio layer. [6] I have argued that a static, spectrum-as-asset approach is not a given; a more dynamic radio-as-trademark interference metaphor is perfectly workable. [7]

Universal Service: From telephony to internet access

The Universal Service Fund in the US, and its equivalents in other countries, was conceived of as guaranteeing phone service to those who would not otherwise be able to afford it, particularly in rural communities. There is no a great deal of debate about extending the universal service concept to the internet. However, since internet access can come in an unlimited variety of flavors, it is unclear what the goal of the program should be. Phone service is the same everywhere; but what broadband speed is “good enough”? The regulatory debate is moving away from how to fund phone service to how to define baseline access.

Common carriage: From a neutral network to network management

The most recent of these debates concerns the 21st century equivalent of common carriage for the internet. The rallying cry of Network Neutrality had satisfyingly thing-like connotations: there was a network, and it had to have the attribute of neutrality (noun/adjective). Over time is has largely been agreed that network operators should have some discretion in managing the behavior of their network. The question has now become a behavioral one: what is degree of network management (verb) is appropriate?

Implications

A shift in the objects of governance from things to behaviors suggests a shift in regulation from ex ante to ex post action, that is, from making detailed rules up-front to stating general principles and enforcing breach after the fact. In Law’s Order [8], economist David M. Friedman compares speed limits (ex ante) with reckless driving (ex post), and observes that ex post punishments are most useful when the behavior is determined by private knowledge that the regulator cannot observe.
"Ex ante punishments can be imposed only on behavior that a traffic cop can observe; so far, at least, that does not include what is going on inside my head. Ex post punishments can be imposed for outcomes that can be observed due to behavior that cannot—when what is going on inside my head results in my running a red light and colliding with another automobile."
When an object of governance is thing-like, and changes in the attributes of those things are easily observed – a data breach occurs, some packets don’t cross the network – then ex ante rules are attractive. When governance concerns behavior, particularly behavior that is difficult to observe – the uses to which data is put by a company, whether a particular network management technique discriminates against a competitor – then the regulator has to fall back on ex post enforcement. The difficulties with ex post are well-known, though: from providing sufficient clarity up-front about what would constitute a breach, to the political difficulty of exacting very occasional but very large penalties from powerful players.

Footnotes

[1] Note that this is not the traditional meaning of the term, which used “object” as synonymous with “objective”, e.g. Edmund Burke: “To govern according to the sense and agreement of the interests of the people is a great and glorious object of governance. This object cannot be obtained but through the medium of popular election, and popular election is a mighty evil.”

[2] Boyd, William, “Ways of Seeing in Environmental Law: How Deforestation Became an Object of
Climate Governance”, to be published in Ecology Law Quarterly

[3] Daniel J. Weitzner, Harold Abelson, Tim Berners-Lee, Joan Feigenbaum, James Hendler, Gerald J. Sussman (2007) “Information Accountability”, Computer Science and Artificial Intelligence Laboratory Technical Report, MIT-CSAIL-TR-2007-034, June 13, 2007

[4] Business Forum for Consumer Privacy, “A New Approach to Protecting Privacy in the Evolving Digital Economy: A Concept for Discussion”, March 2009

[5] Kevin Werbach (2003), "Radio Revolution: The Coming of Age of Unlicensed Wireless," New America Foundation and Public Knowledge, no date on document, dated 15 Dec 2003 on NAF site

[6] Preston Marshall (2009) “Quantifying Aspects of Cognitive Radio and Dynamic Spectrum Access Performance” (see slides 15, 16)

[7] J Pierre de Vries, (2008) "De-situating spectrum: Rethinking radio policy using non-spatial metaphors" New Frontiers in Dynamic Spectrum Access Networks, 2008 (DySPAN 2008). http://ssrn.com/abstract=1241342

[8] David M. Friedman, Law's Order: What Economics Has to Do with Law and Why It Matters, Princeton University Press: 2001. See Chapter 7 for a discussion of ex ante/ex post.

Norms, mechanisms and policy imperatives

2009-12-18T15:07:00.000-08:00

As I stumble towards a paper about changes in governance required by changing technology (part of the Silicon Flatirons New Models of Governance project) I’ve found Peter Cowhey and Jonathan Aronson’s magisterial new book on the political economy of global communications [1] very useful.

In the Summary and Conclusions, co-written with Don Abelson, they introduce four “principles” for market governance in the light of current conditions, and ten “norms” needed to implement the principles (see Appendix 1 below). They define market governance as “the mixture of formal and informal rules and the expectations about how markets should logically operate.”

When I look at their norms, I see a set of choices for the set-points of a small number of governance mechanisms:

Subsidy (Norm 2)
Competition policy (Norms 3, 5)
Regulatory “touch” (Norms 1, 4, 6)
Property rights (Norms 8, 9, 10)

As one might expect from a work devoted to the governance of markets, the norms (and the governance mechanisms they imply) do not address some non-economic public policy imperatives. [2] For contrast, consider the “permanent policy imperatives” I introduced in my paper “Internet Governance as Forestry” [3]

Public Safety. Protecting citizens is a primary responsibility of government.
Consumer Protection. Policy makers take action when lawmakers conclude that commercial activity needs to be circumscribed in the public interest.
Culture and Values. In order to protect and express a culture’s values, policy makers seek to limit some kinds of speech and promote others.
Government Revenue. Money needs to be raised and redistributed by federal, state and local treasuries; this includes taxes, fees, levies, subsidies, and tax breaks.
Economic Vitality. A healthy market produces goods and services that citizens value.

Now, these two lists are different in kind; Cowhey & Aronson’s norms and implied mechanisms are means, and my policy imperatives are ends. However, the mismatches are revealing.

My list of policy imperatives does not include subsidy, which is implied by Cowhey & Aronson’s Norm 2, “Invest in virtual common capabilities”. In the light of their work, I now realize that this is an omission; distributing government largesse is a permanent policy imperative.

The mechanisms of competition policy and property rights are means to the end of economic vitality, my fifth policy imperative. The mechanism of regulatory “touch” is a means that I address in my paper under the heading of Principles (see Appendix 2, below); as it happens, I concur with their recommendations for light touch regulation.

The difference in emphasis is perhaps most noticeable in the absence of norms/mechanism that speak to the “soft” policy imperatives. While Cowhey & Aronson’s Norm 7 addresses media content, and thus recognizes some value in “culture and values”, my third policy imperative, it is not implementable in the way the others are; it merely recommends a balance between encouraging trade and protecting cultural values. The “public safety” imperative is completely absent. While one may argue that Imperative 2, “consumer protection”, is to be achieved through competition policy (Norms 3 and 5), Cowhey & Aronson do not explicit mention of consumers.

Footnotes

[1] Cowhey, Peter F. and Jonathan D. Aronson, Transforming Global Information and Communication Markets: The Political Economy of Innovation, MIT Press (February 15, 2009). Softcopy available at http://globalinfoandtelecom.org/book/ (look for the “Download free under Creative Commons license” link)

[2] It is telling that Cowhey and Aronson seem to equate the public interest with consumer welfare, an economic construct. For example, on p. 17 they write: “The main challenge for governance is creating appropriate new spaces for market competition that allow the most important potential for innovation to play out in a manner that enhances consumer welfare (the public interest).”

[3] De Vries, Pierre, “Internet Governance as Forestry: Deriving Policy Principles from Managed Complex Adaptive Systems”, TPRC 2008. Available at SSRN: http://ssrn.com/abstract=1229482

Appendix 1: Four guiding principles and ten norms to help implement them
(Cowhey & Aronson (2009) Table S.1, p. 265

Principles

Enable transactions among modular ICT building blocks.
Facilitate interconnection of modular capabilities.
Facilitate supply chain efficiency, reduce transaction costs.
Reform domestically to help reorganize global governance.

Norms

Delegate authority flexibly.
Invest in virtual common capabilities; be competitively neutral.
Use competition policy to reinforce competitive supply chains.
Intervene lightly to promote broadband networks.
Narrow and reset network competition policy. All networks must accept all traffic from other networks. Narrow scope of rules to assure network neutrality. Separate peering and interconnection for provision of VANs.
Government should allow experiments with new applications.
Create rules for globalization of multimedia audiovisual content services that encourage international trade and foster localism, pluralism, and diversity.
Tip practices toward new markets for digital rights.
Promote commercial exchanges that enhance property rights for personal data and mechanisms to do so.
Users own their information and may freely transfer it.

Appendix 2: Four ecosystem management principles

De Vries (2008), Table 3, p. 26

Flexibility: Determine ends, not means.
Delegation: Most problems should be solved by the market and civil society.
Big Picture: Take a broad view of the problem and solution space.
Diversity: Multiple solutions are possible and desirable.

Polling x Lobbying = ?

2009-12-17T10:37:00.000-08:00

Polling and lobbying are powerful factors of government that aren’t usually covered in Civics 101. Both are huge industries, and both shape the way political decisions are made. The current wave of web technology is going to create a hybrid form that will reshape politics.

According to 2002 Census data, the marketing research & public opinion polling industry as a whole had revenues of $10.9 billion; special interests paid Washington lobbyists $3.2 billion in 2008 according to the Center for Responsive Politics. Lobbying is as old as politics, but polling is relatively new (19th century), as is its premise: the importance of mass public opinion in government and diplomacy (18th century). Lobbyists are key players in Washington DC, and there’s a revolving door that moves former federal employees into jobs as lobbyists, and that pulls former hired guns into government careers or political appointments. Polling expertise is a key attribute in top political advisors, and something that politicians – and administrations – do incessantly.

The social media technologies of Web 2.0 will create a lobbying/polling hybrid and create a new political power center to rival traditional lobbying and polling. Efforts by government to solicit citizen opinion, like the Ideascale site soliciting input on the National Broadband Plan, or the Open for Questions site run by the White House, are a way for citizens to engage in little-L lobbying. These channels invite manipulation that will amount to big-L lobbying. In the same way that astroturfing co-opted grassroots lobbying, political operatives will co-opt the forms of web 2.0 citizen participation. Those who are adept at viral marketing will propel political memes into real-time polling tools in way that amounts to lobbying.

The amplification of the randomly popular that is pervasive on social rating sites like digg will infuse politics, intensifying the temptations of “poll, then decide”. We’ll also likely see something akin to the hollowing out of the media industry mid-list that The Economist charted in “A world of hits”: In movies and books, both blockbusters and the long tail are doing well; the losers are titles (and retailers) in the not-quite-so-good middle ground. Similarly, blockbuster issues will be laid on for the mass public that doesn’t care about politics (shibboleths like taxes and abortion), and niche lobbying on topics like radio spectrum, prison reform, and privacy will become even more fine-grained. Citizen publics will be important in both: as armies of computer-generated extras in the first case, and as engaged semi-experts in the second. Worthy mid-ground issues like trade, education, and energy policy will get steadily shorter shrift.

One implication is that niche topics like hunger policy shouldn’t strive to move up the charts into the middle ground – they’ll just wither there. Rather, niche players should embrace their residence in the long tail and make the most of Web 2.0 phenomena, like Polling x Lobbying, that give them direct access to the appropriate sliver of the policy making elite.

Constructing spectrum – lessons from the history of economics

2009-12-14T16:03:00.000-08:00

“Spectrum” is powerful construct; most people assume such a thing exists, and this assumption has regulatory consequences. But how did it come into being? The stories that some social scientists tell about the construction of “the economy” by economics provide some insight.

Timothy Mitchell, for example, argues that the economy was created by economists:

The economy is a recent product of socio-technical practice, including the practice of academic economics. Previously, the term “economy” referred to ways of managing resources and exercising power. In the mid-twentieth century, it became an object of power and knowledge. Rival metrological projects brought the economy into being. [1]

In his chapter in Do Economists Make Markets? On the Performativity of Economics, Michel Callon puts it this way: “To claim that economics is performative is to argue that it does things, rather than simply describing (with greater or lesser degrees of accuracy) an external reality that is not affected by economics.” MacKenzie argues in his chapter of the same book that the Black-Scholes-Merton options pricing model not only helped traders price something that already existed; it also shaped it, since most traders ended up using the model, prices converged to what the model predicted. [2]

In the same way, economists who treat spectrum as an asset (see my post "Property rights without assets") are not simply describing an external reality; they are bringing something into being. One of the key tools in this process is metrology: for example, the gathering of GDP data brings into being “the economy” which is reified through numbers like the GDP. In the same way, the program to make an inventory of spectrum buttresses the spectrum-is-real perspective. (More on spectrum inventories in a future post.)

Implications

World views have consequences, and thus stakeholders. Those who have a stake in the existing spectrum-based regime gain from this view; questioning the validity of “spectrum” undermines the security of their rights and privileges. This applies not only to capitalists who own spectrum licenses, but also to progressives who base their claims to government supervision of radios on the public ownership of the supposed “spectrum asset”. On the other hand, if one thinks of radio regulation simply in terms of the operating rights associated with radios, then a much more dynamic regime can be imagined – one that would benefit both political and commercial entrepreneurs. A non-spectrum world view might also be attractive to current “spectrum owners” who are discontented with their rights. [3]

The political and engineering systems that have co-evolved with the spectrum concept have specific characteristics: largely static allocations of rights to operate defined in terms of fixed frequency ranges. More dynamic approaches don’t fit nicely. For example, Preston Marshall wants to guarantee the right to operate, but not exclusivity over one channel; he proposes to guarantee a licensee (along with others) an aggregate of access to sufficient frequencies to meet a certain amount of service. [4]

This is an approach that focuses on behavior, rather than the exclusive ownership of an asset. As I argued in "Property rights without assets", this is perfectly compatible with a property rights regime, since property rights don’t have to be based on an underlying asset.

The bottom line is that a spectrum-as-asset approach leads one to ignore elements, which leads to inferior rights design. Specifically, receivers have been ignored. If one thinks one’s job is to "carve up spectrum", then you don't have to worry about receivers. But when radio is considered as a system, the receivers determine interference just as much as transmitters, so one has to take them into account explicitly. By analogy: if you're deciding a land trespass case, you don't worry about whether farmer is grazing Holsteins or Friesians. But if you're deciding a trademark dispute, everything depends on what happens in the mind of the consumer (analogous to the receiver). [5]

Footnotes

[1] Mitchell, Timothy (2008) “Rethinking Economy”, Geoforum, Volume 39, Issue 3, May 2008, pp. 1116-1121.

[2] MacKenzie, Donald A, Fabian Muniesa, Lucia Siu (2007), Do economists make markets? On the Performativity of Economics, Princeton University Press, 2007

[3] There is debate about the origin and extent of government property rights in spectrum; see for example the Introduction of William L. Fishman, “Property Rights, Reliance, and Retroactivity Under the Communications Act of 1934”, Federal Communications Law Journal, Vol. 50, No. 1. Fishman concludes: “It would probably be better, therefore, to say that the government regulates electromagnetic radiation in certain defined frequencies, rather than to say it regulates spectrum.”

[4] See e.g. Section 5.4 in the report “Radio Regulation Summit: Defining Inter-channel Operating Rules”.

[5] For more on the virtues of the radio-as-trademark metaphor, see my blog post “De-situating Spectrum: Non-spatial metaphors for wireless communication”, and paper “De-Situating Spectrum: Rethinking Radio Policy Using Non-Spatial Metaphors”

Property rights without assets

2009-12-10T13:06:00.000-08:00

I’m still nagging away at the implications of the fallacy that spectrum exists. Sorry.

I’ve been struck recently that many if not most definitions of property rights seem to turn on a relationship to an asset. For example, Gary Libecap in Contracting for Property Rights defines them as "the social institutions that define or delimit the range of privileges granted to individuals to specific assets" (1990:1); or Yoram Barzel in The Economic Analysis of Property Rights: "Property rights of individuals over assets consist of the rights, or the powers, to consume, obtain income from, and alienate these assets" (1997:2). Such definitions set out to define rights which assure the owner of an asset that they can derive value from that asset.

However, one can have rights to create value that do not require the existence of an underlying asset – unless, of course, one takes the position that the existence of a property right necessarily implies an asset. [1]

Therefore, let me distinguish between any property right, which is an asset in itself, and a property right to exploit an asset, which entails two assets: the right itself, and the underlying asset. All assets can lead to property rights – perhaps tautologically, in that something might not be counted as an asset if it does not have rights associated with it – but not all property rights require assets.

Examples

It always helps to make things concrete. One property right without an underlying asset is a New York taxi cab medallion: it's a right to operate, but there isn't an underlying asset. The right is tied to a particular place (New York), but that place isn't the asset.

Another common asset-less right is a franchise, that is, an agreement to sell a company's products exclusively in a particular area or to operate a business that carries that company's name.

Perhaps my favorite is a trademark, that is, a word, symbol, or phrase, used to identify a particular manufacturer or seller's products and distinguish them from the products of another. One might use the word “Wired” to brand a magazine, but the word isn’t the asset; when I last counted about a year ago, there were about 27 distinct trademarks using the word "wired" in the US.

Notice that permission for an agent to behave in a particular way is the essence of all these rights – and of rights that require assets, too. Therefore, I’d contend that behavior is the key to property rights, and assets are optional.

There are of course many property rights to assets, from owning a pencil to the right to extract oil in a particular region. Note that the underlying assets don't have to be tangible: an algorithm over which one has a patent is a perfectly viable intangible asset (perhaps made so exactly by the property right).

Implications

This distinction between property rights that do and do not require underlying assets matters: if one assumes an underlying asset where there is none, one is liable to over-assign rights.

For example, if trademark regulation assumed that the word being used was the asset, then it might give the owner of the trademark the right to all possible (commercial) uses of the word. There would be only one “Wired” trademark in the US, let’s say owned by Condé Nast; the companies who wanted to use the word to sell cologne, art supplies, energy drinks, stationery, electronic door chimes or automobile wheels would be out of luck. This would be a loss because an entrepreneur could apply the letters w-i-r-e-d to some new product that couldn’t be confused with a magazine without seeking (and probably failing to get) Condé Nast’s permission.

Similar reasoning applies to radio regulation. The existence of radio licenses doesn’t mean that there is an underlying asset, “spectrum”. [2]

If one regards a radio channel as an asset, then (Anglo-American) regulators have shown a proclivity to grant an expansive array of rights. Following the norm of technology and service neutrality, they have defined operating rights so broadly that pretty much preclude all operations that radiate energy in that channel, regardless of its harm to the licensee, in order to allow the licensee to operate in any conceivable way. [3] Such a broad definition forecloses new entry by potentially useful but non-interfering services

A broad definition also forecloses future arrangements of radio operating rights that are not tied to channel-based world view. Bands and channels, as regulatory constructs, are in large part a consequence of the two-stage super-heterodyne radio design that first filters a broad range of frequencies at the "RF stage", and then after down-conversion picks out a narrow range at the "IF stage". This is an old-fashioned approach that is increasingly becoming obsolete [4] - but it is enshrined in regulation.

References

Barzel, Yoram, Economic Analysis of Property Rights, Cambridge University Press 1989, second edition 1997
Libecap, Gary D., Contracting for Property Rights, Cambridge University Press 1990

Footnotes

[1] A view of property rights that does not require the existence of underlying assets is not identical to the "bundle of rights" approach taught in law school property classes; there it's taken as a given that there's an underlying asset - paradigmatically, real estate - and the bundle explains how it can be simultaneously “owned" by multiple parties.

[2] The emergence of the spectrum concept suggests that this is, indeed, the conclusion that has been drawn. Perhaps the reasoning that a property right must entail an asset is one of the reasons why “spectrum” has become such an entrenched concept.

[3] I’m ignoring allowed inter-channel interference; for a discussion of that case, see my report on the meeting held at Silicon Flatirons, “Defining Inter-Channel Operating Rules”

[4] See e.g. Soni & Newman 2009, "Direct conversion receiver designs enable multi-standard/multi-band operation", RF Designline

Alfred Kahn, SURs, and new approaches to radio regulation

2009-12-07T16:03:00.000-08:00

I have at last finished writing up a Silicon Flatirons meeting on rules for inter-channel radio interference (web page, PDF). Reflecting on the event, I was struck again by the contrast Phil Weiser noted last year [1] between the success of airline deregulation, and the halting progress in doing the same for “spectrum”.

The meeting showed there was broad support for taking receivers into account more explicitly when drafting rules, for example by regulating resulting signal levels rather than the customary approach of specifying rules for individual transmitters. This approach focuses on the results of transmission – which includes interference, the bone of contention in most radio regulation debates – rather than the transmission itself.

Ofcom, the UK communications regulator, took an interference-based approach to licensing by creating Spectrum Usage Rights, also known as SURs [2]. However, SURs were roundly rejected by the cellular operators. Ofcom chose not to impose SURs on the mobile industry, on the premise that the goal of SURs was to improve the certainty of the license holders for their benefit, not the regulator’s benefit.

While there are many other reasons for the cellcos to reject SURs (the problems SURs address, like uncertainty about likely uses and technologies, or disparate uses in adjacent channels, are largely absent in cellular bands), it is clear that Ofcom deferred to the interests of incumbents – potentially at the cost of consumers or new entrants. One of the conclusions of a 2007 report for the European Commission on radio interference regulatory models [3] came to mind:

“Technology and service-neutral licensing (as would be supported by interference-based licensing techniques) offers significant benefit for end-users but not necessarily for spectrum owners and network providers.”

In an essay in honor of Alfred Kahn’s 90th birthday, Phil Weiser observed that airline regulation (where Kahn, the "Father of Airline Deregulation," made his name) and spectrum regulation share some basic characteristics: both regimes emerged from an effort to protect established interests; both limited output by restricting the use of the resource in question; and in both cases, early academic criticism calling for regulatory reform went unheeded. In making the case for Kahn as a political entrepreneur, Weiser argues that he “pursued the objective of eroding the airline industry’s commitment to the legacy regulatory regime by both undermining the manner in which it protected established incumbents and bolstering the strength of those interests that would benefit from deregulation.”

The radio incumbents Weiser had in mind were the broadcasters and not the cellular companies – but it’s not too much of a stretch to attribute at least some of the resistance to new methods of radio regulation to the New Incumbents.

REFERENCES

[1] Phil Weiser (2009), “Alfred Kahn as a Case Study of a Political Entrepreneur: An Essay in Honor of His 90th Birthday.” Journal Network Economics, 2009. Abstract at SSRN. The paper was first delivered at a conference at Silicon Flatirons in Boulder on September 5, 2008.

[2] See e.g. William Webb (2009), “Licensing Spectrum: A discussion of the different approaches to setting spectrum licensing terms” (PDF); and Ofcom (2008), “Spectrum Usage Rights: A Guide Describing SURs” (PDF)

[3] Eurostrategies and LS telcom, “Study on radio interference regulatory models in the European Community” 29 November 2007 (PDF)

Another problem with the ecosystem metaphor

2009-11-02T10:33:00.000-08:00

Lay commentators - including yours truly - have a weakness for using the biological communities as a metaphor for social interactions, particularly business competition. No matter the litany of quibbles (see my own at Eco mumbo jumbo); it's just so tempting to talk about a business ecosystem!

A recent News Focus piece in Science (Erik Stokstad, On the Origin of Ecological Structure, Science 2 October 2009, Vol. 326. no. 5949, pp. 33 - 35; there's also an interview with Stokstad in the podcast of 2 October, 2009) raises a more fundamental problem with the utility of the metaphor: ecologists themselves are still struggling to understand what dictates the kinds and proportions of organisms in communities ranging from meadows to montane forests.

Stokstad writes that "there is still no consensus on the relative importance of the various forces [that influence community formation, like competition, predation, and disturbance]. Darwin and many later ecologists emphasized competition among species, but proponents of a controversial theory of biodiversity that assumes competition has no impact argue that immigration and other random demographic events can account for much of the apparent makeup of communities. As a result, ecologists have a long way to go to come up with formulas that predict how communities might arise and change."

If ecologists can't explain community dynamics in biology, it's dangerous to make inferences by analogy about the influence of (say) competition and disturbance in commercial systems. Which is a pity, since that's just what I've tried to do myself...

Unlicensed Wireless :: Open Source

2009-10-27T08:52:00.000-07:00

Unlicensed is to the cellular business as open source was to the software industry.

Wireless companies are clamoring for "more spectrum", that is, new radio operating licenses that will allow them to satisfy the exploding demand for data capacity (think video streaming to an iPhone). New licenses are a cheaper way to increase capacity than the alternative: building additional cell towers.

The cellular companies have traditionally been opposed to unlicensed radio allocations (like the one that allows Wi-Fi networks) because they saw them as a substitute for licensed: the more unlicensed, the less licensed allocations; and less licensed meant higher prices at auction.

However, at least some of them seem to be realizing that unlicensed can help them off-load traffic from their licensed networks. A video stream that arrives on a phone in a coffee shop via a wired connection to the shop and a Wi-Fi link to the device has not crossed the licensed cellular network, freeing up cellular capacity.

Unlicensed is therefore a complement to licensed - just as open source can complement rather than substitute for proprietary software. Even Microsoft now offers some software under open source-like licenses.

Just as with software, one should expect kicking and screaming in the cellular industry, and variation in the degree of acceptance of unlicensed depending on business model and other assets. IBM is a big supporter of open source because it makes money on services rather than software licenses; T-Mobile is more open to unlicensed than Verizon because it has fewer licenses.

Spectrum Databases (Not)

2009-10-01T09:49:00.000-07:00

The idea of using a database to regulate radio operation – or, to “control access to spectrum”, to use the S-word nomenclature – has been gaining ground.

For example, both Michael Calabrese (The End of Spectrum ‘Scarcity’, New America Foundation Wireless Future Program Working Paper No. 25, June 2009) and Kevin Werbach (Castle in the Air: A Domain Name System for Spectrum, TPRC September 2009) have argued that the database(s) contemplated to manage device operation in the TV white spaces could be the foundation for a method to increase the amount of radio operation.

Thinking through how such a database might be used shows the advantage of approaching radio regulation as coordinating operations, rather than using conventional approach of “dividing up spectrum”.

The regulatory challenge is therefore not "spectrum databases" but "radio operation databases".

In a first approximation – and perhaps even as the ultimate solution, if one uses the “spectrum” approach – a database would be a listing of “vacant” channels; a device would query the database for “available” channels, and operate in one. When one starts from the basis that spectrum is an asset like land to be divided up and distributed, vacancy is a self-evident concept; it derives from the attributes of the underlying asset, and not by reference to the intended use.

However, context is everything in radio operation. Whether harmful interference will result from the operation of an added radio system depends not only on its transmissions, but also the transmit and receive characteristics of the incumbent system.

Consider, for example, three channels: A, B, and C. Let’s say incumbent system #1 operates using channel A. Channels B and C are nominally vacant. Can an incoming system #2 operate in those channels? If both system #1 and #2 use traditional cellular technology (i.e. FDM, e.g. 3G), the answer is yes. But if #1 uses 3G and #2 uses TDM technology like WiMAX, then the answer is No: there needs to be a guard band between them, and system #2 can only use Channel C. Channel B needs to be left “vacant”. (This is a live issue: see e.g. Ars Technica on the argument between T-Mobile and M2Z over the rules for the AWS-3 band band.)

A mental model informed by spectrum-as-land is therefore not an ideal guide to understanding what needs to be in the database. (More generally, one needs to refine the metaphor to better guide regulation, as Weiser and Hatfield did last year by introducing the concept of "zoning the spectrum" in Spectrum Policy Reform and the Next Frontier of Property Rights, 15 Geo. Mason L. Rev. 549.)

An approach grounded in coordinating operations, on the other hand, leads to the understanding that what needs to be in the database is not just a frequency range and geographic region, but all the relevant parameters of an incumbent operation. The short list would add receiver performance (ability to reject interference) and duty cycle (near-constant transmission like cellular systems, vs. very intermittent but intense uses like firefighting) to the usual suspects of transmitter location, emitted power, and transmit mask.

The task is not to find a “vacant channel”, but to determine if an incoming operator will cause harmful interference. This requires, in addition to the operating parameters of the incumbent and incoming systems, information about the spatial distribution of incumbent and incoming radios, and a propagation model to connect the two.

Ofcom is the regulator that has thought most deeply about ways to better characterize the interference characteristics of radio systems; see e.g. Ofcom’s Guide to Spectrum Usage Rights (SURs) and William Webb’s recent paper Licensing Spectrum: A discussion of the different approaches to setting spectrum licensing terms.

A well-founded framework for generalizing the white space database – where interference management between incumbents and new entrants hard-coded into the FCC rules for white space device operation – could benefit from new radio operating metaphors (grind axe: see my De-situating spectrum: Rethinking radio policy using non-spatial metaphors, DySPAN 2008) and the application of a SUR-like approach.

One will also have to think carefully about the minimal set of parameters needed to facilitate interference avoidance, since it's easy but economically inefficient to come up with a very long list of attributes that describe radio operations. The Silicon Flatirons Center recently examined this issue in a summit on defining out-of-band operating rules.

No more S-word

2009-09-29T09:40:00.000-07:00

I’ve undertaken a language challenge: can one talk about radio policy without using the word “spectrum” (except in quotes)?

My guess is that I don’t need to use the S-word at all; that it can be replaced with simple terms rather than clumsy paraphrases; and that the clarity of a text, and the understanding of the reader, will be greatly improved if one avoids it.

I first raised this possibility in the DySPAN 2008 paper De-Situating Spectrum: Rethinking Radio Policy Using Non-Spatial Metaphors where I recommended a “restatement of wireless policy in terms of system operation rather than spectrum.”

The word “spectrum” has many meanings, depending on context. In policy documents it’s usually short-hand for the topic dealt with by radio regulators, aka their “the object of governance”. There are a range of intended meanings, including a radio license; a range of frequencies; or all the parameters (frequency, geography, transmit power mask, allowed use, single or paired bands, etc.) associated with a radio license.

Engineers use “spectrum” to refer to a range of frequencies, or sometimes to electromagnetic phenomena. Less frequently – curiously, since this is closest to the dictionary definition – they use it to refer to the distribution of electromagnetic energy that results from radio operation.

In short, the following substitution covers most cases:

FOR spectrum SAY radio license OR radio operation OR frequencies

The S-word is also used in various combinations; here are some translations:

FOR acquire spectrum SAY acquire permissions

FOR sharing spectrum SAY coordinating operation

FOR use of spectrum SAY operation of radios

FOR spectrum rights SAY rights to operate

FOR spectrum allocation (noun) SAY license type

FOR spectrum allocation (verb) SAY deciding use

FOR spectrum assignment (verb) SAY authorizing a radio operator

FOR Dynamic Spectrum Access ("DSA") SAY dynamic radio operation

FOR stockpiling spectrum SAY stockpiling licenses

FOR demand for spectrum SAY demand for licenses

FOR manage spectrum SAY manage radio operation

FOR improve the efficiency of spectrum use SAY increase concurrent radio operation

FOR a chunk of spectrum SAY operations concentrated in a band

The value of more precise terminology becomes obvious when one looks through this list. One can distinguish between two distinct referents of “spectrum”: the parameters of operation of radios, and the rights to operate. It’s a distinction between assets and operations. One can easily put radio licenses (“spectrum”) on a balance sheet, but not the institutional and technological ways of coordinating radio operation (“spectrum”).

But why bother? An obvious retort is that this is just nitpicking: “Everybody knows what they mean by the word in a given context.” I argue that it’s important because the connotations of words matter, and change how we see the world.

Constant, thoughtless use of the S-word without teasing apart its meanings creates a thing: “the spectrum”. We come to accept as real the illusion that we’re dealing with a concrete thing (like bushels of corn) rather than the behavior of devices and their owners. If one takes away the radios, “spectrum” as an object of governance ceases to exist, although “spectrum” in the sense of “electromagnetic phenomena” persists. This illusion leads to the fallacy that “spectrum” can be counted like bushels of corn, whereas it is in fact a regulated socio-technical arrangement. It leads to the fallacy that “spectrum” can be counted, and that it is permanently divisible and inalienable.

I’m not denying that interference can occur between radio systems, nor that property rights can facilitate the coordination of radio operation. Rather, I’m suggesting that “spectrum” leads too easily to important conclusions that need to be considered more deeply, such as that wireless licenses are necessarily exclusive rights to operate in fixed frequency ranges. A focus on the behavior of radio systems, which changes constantly as technology and institutions evolve, rather than some spectrum-as-thing can produce a more robust and efficient way to coordinate radio operation.

A cybersecurity taxonomy

2009-09-24T08:01:00.000-07:00

I recently chaired a panel on “Cybersecurity and Digital Identity” at a USC roundtable preparing for the APEC Ministerial in 2011.

The links between cybersecurity, digital identity and trade were not immediately obvious to me, and since security isn’t an area where I can even pretend to have expertise, it forced me to think through the topic from the ground up.

I ended up reframing the topic as “the protection of assets in the digital age.” Not “digital assets”, although some assets are undoubtedly digital. Some concrete assets have digital dimensions: for example, a compromised SCADA system can deprive a city of its water supply. This is a new risk because the use of standardized/open solutions and the growing internet connections between SCADA systems and office networks has made them more vulnerable to attack. And while a person’s reputation isn’t digital as such, information technology has changed how reputations are constructed, disseminated, and need to be protected.

The next step is to categorize the assets that need to be protected, and for that one can consider various attributes. One useful categorization is the motive for threatening assets; I submit that Sex, Money, and Power are the three important motivations (in all things!).

Sex is about status – high status improves reproductive success; into this category would fall hackers who build exploits to show their prowess, and people who want to build a digital persona.

Money refers to economic motivations, whether protecting intellectual property rights in content through encryption, or building botnets for fraud or blackmail.

Power is perhaps least talked about until recently: it’s the pursuit of national interest through IT, e.g. “cyberwar”. The assets in question include critical national infrastructure, and sensitive intelligence.

As an alternative nomenclature to sex, money, and power, one might think of Fame, Fortune, and Foreign Affairs.

The motivations of sex, money, and power can be mapped against another categorization, that of the asset context. In increasing order of scale, the contexts are the personal, the corporate, and the national (aka social, commercial, and political). (However, note that global corporations actually operate at both a national and transnational scale.)

With these two categorizations, one can then plot topics on a handy grid (apologies about formatting; I haven't grokked how to import tables into blogger):

			Context
		Personal	Corporate	National
Assets		Reputation Money, goods Personal safety	Reputation, brand Intellectual property Tangible assets Employee & customer safety Business continuity	Critical infrastructure Intelligence State assets, incl. military Political power structures National wealth
	Sex (aka fame)	Privacy Harassment Defamation	Brand hijacking, web defacement	Embarrassment
Threats (by motive for attack)	Money (aka fortune)	Fraud Identity theft Botnet recruiting	Theft of goods Appropriation of know-how Diverted compute capacity Extortion	Advantage national champions Create non-tariff barriers
	Power (aka foreign affairs)	Suppression of speech, access	Appropriation of IPR Reduce ability to compete	Intelligence gathering Degrading infrastructure & assets Demoralizing populations

A few notes:

Threats to assets come in various flavors, notably appropriation, destruction, and constraint of use

Trade occurs both within and between columns, that is, between individual people and between individual companies, as well as between people and companies. Likewise, at a different resolution scale, between nations.

It helps to distinguish between the What vs. the How. Security doesn’t appear explicitly in the table, and neither does digital identity; both are means to end (“how) of protecting assets (“what”). Other means (they do overlap) include encryption, digital rights management, and norms, rules and treaties.

Resource management, spectrum, and cyborg fish

2009-09-18T14:38:00.000-07:00

One runs out of spectrum the way you run out of patience. While there is a supply of it, metaphorically speaking, in practice it’s a matter of the behavior of agents (radios) in a given context – just as patience is a matter of how one person, with certain proclivities, reacts to the actions of another. The outcome depends as much on the attributes of the agents as the details of the context, and both are constantly changing.

“Spectrum” is a useful fiction that is used to talk about the management of radio operation. As I argued in De-situating Spectrum: Non-spatial metaphors for wireless communication, this usage is grounded in the spectrum-as-space metaphor, and in particular the spectrum-as-land variant. However, there are other viable metaphors for radio operation, such as thinking about coordinating radio regulation as similar to trademark: an infringing trademark that confuses a customer is like one radio that causes interference in another. A radio license is like a trademark; it enables one to prevent harmful interference by others.

However, the spectrum-as-land metaphor is pervasive. It appears in debates about how best to “manage spectrum”. Opponents of unlicensed , for example, claim that it leads to a tragedy of the commons. However, there is no tragedy of the commons with Wi-Fi, because there is no commons; that is, there is no underlying, essential resource that bears any resemblance to land in Hardin’s classic analysis, or as far as I can see, to what either lawyers or economist mean by “a resource” (assuming you can pin them down long enough to extract a definition). One can measure land in acres, and sheep in heads. A moment’s thought will confirm that there is no agreed way to measure “spectrum”: it’s not MHz, not MHz/m^2, not MHz.m^2, not watt/MHz or watt.MHz or watt.MHz/m^2 or watt.MHz/bps.m^2 or … And this doesn’t even begin to address the role of receivers.

I suspect that the rise of the spectrum metaphor is related the entry of economics into radio regulation. In order for (neoclassical) economics to get purchase on a topic, it must have (or create) a thing to be traded; and since economics is the allocation of resources under scarcity, this thing must be a scarce resource. Just introducing radio licenses don’t seem to have been sufficient to operationalize radio operation in economic terms; there need to be something which is licensed, or to which access is licensed. “Spectrum” fits the bill; it’s reasonably stable, and given the spectrum-as-land metaphor, sufficiently thinglike.

An interesting analogy is the creation of individual transferable quotas (ITQs) detailed in Petter Holm’s “Which Way is Up on Callon?” in Do Economists Make Markets (2007), edited by Donald MacKenzie, Fabian Muniesa, and Lucia Su. According to Holm, the ITQ model is an invention of neoclassical economics whose adoption was facilitated by the rise of fisheries resource management. As he tells it, it’s a complicated story “about the construction and stabilization of a heterogeneous network, tying the fish in with fishermen, echo integrators, log books, legislation, computers, bureaucracies, mathematical formulas, and surveillance procedures. It is within such a network that that the fish-as-fit-for-management springs to life, as a true cyborg: part nature, part text, part computer, part symbol, part human, part political machine.” [1] [2]

The analogy to “spectrum” is pretty clear – the development of spectrum analyzers, the measurement program exemplified by the “spectrum occupancy maps” produced by SSC, NAF and others, and the stock analyst’s $/MHz.POP metric for analyzing auction results serve to construct “spectrum” as a manageable object of governance. [3]

I haven’t nailed down when spectrum entered the regulatory lexicon. It’s not used at all in the 1912 Radio Act, which mentions only “wave lengths”. The 1934 Act only refers to frequencies. However, Coase’s “Federal Communications Commission” paper (1959) cites Mr. Justice Frankfurter’s 1934 opinion in National Broadcasting Co. v. United States (319 U.S. 190, cited in Coase 1959), where the spectrum-as-space seems well established: “the radio spectrum simply is not large enough to accommodate everybody.” Frankfurter also articulates the resource notion underpinning subsequent development of the spectrum concept: “The facilities of radio are limited and therefore precious; they cannot be left to wasteful use without detriment to the public interest.” Spectrum-as-a-spatial-resource is therefore a pre-WW II concept; it evidently didn’t flow from the Coasian economization of radio, but I believe they reinforced each other. [4]

One can rightly ask why, given that the spectrum-as-land notion was well established in the Thirties, and Coase advanced the property rights case in 1960, it took another 30 years for the “propertization of spectrum” to be realized. I suspect that RF spectrum analyzers and broadband radio technology converged to bring about the “auction moment” in the 90s. It took time for both the analytical (game theory, law & economics) and engineering (real-time RF spectrum analyzers [5], commodity spread spectrum technologies) concepts to mature. They’re now well entrenched – and my quest to undermine the primacy of spectrum as an object of governance may be quixotic.

Notes

[1] Holm’s analysis is grounded in Callon and Latour’s Actor-Network Theory, which has been used to make the argument that economists “perform” the economy, that is, that economic models not only describe economic reality, but in part constitute it.

[2] Amusingly, it took a couple of trenchant critics of Callon to show how internecine politics in academic economics shaped the FCC’s spectrum auctions (Mirowski, P. and Nik-Kah, E. (2007) “Markets Made Flesh: Callon, Performativity, and a Crisis in Science Studies, Augmented with Consideration of the FCC Auctions,” in D. MacKenzie, F. Muniesa and L. Siu (eds.) Do Economists Make Markets? On the Performativity of Economics.)

[3] William Boyd introduced me to the notion of “objects of governance”.

[4] There seems to be tension in the law & economics community about this. Coase is a “property is a bundle of rights” man, which is decried in (2001) “What Happened to Property in Law and Economics?” (Thomas W Merrill and Henry E Smith, Yale Law Journal, 2001, Vol. 111, p. 357). Merrill & Smith advocate in rem rights that are referred back to things, as opposed to in personam rights that are arbitrary contractual arrangements. Thomas Hazlett has picked up on Merrill & Smith, and very much would like to treat spectrum as a thing which can be licensed unproblematically – but Coase got his rebuttal in at the start: “... what is being allocated by the Federal Communications Commission ... is the right to use a piece of equipment to transmit signals in a particular way. Once the question is looked at in this way, it is unnecessary to think in terms of ownership of frequencies or the ether.”

[5] Real-time spectrum analyzers for acoustic studies were available by 1957. Sri Welaratna’s "30 years of FFT Analyzers", in Sound and Vibration (January 1997, 30th anniversary issue) gives a historical review of hardware spectrum-analyzer devices; again, these are focused on vibration studies.

Spectrum nominalism

2009-08-25T16:03:00.000-07:00

James Franklin’s discussion of nominalism vs. realism on the Philosopher’s Zone struck me as relevant to my obsession with “spectrum” as a concept.

“To be realist about some concept is to say that there is such a thing, and it's not just made up by us, whereas to be nominalist, is to say it's just a way of speaking of ours, from the Latin 'nomen' word, just an empty sign. So for example, in the case of forces I was arguing for realism about forces. When you felt them by pressing the fingers together, you would naturally conclude from that that there is such a thing as forces. On the other hand you'd never be tempted to do that with something like the average Londoner. Scientists tell you that the average Londoner has 2.3 children; you're not tempted to think that that's anything except a way of speaking, that there's some individual entity called the average Londoner, that has 2.3 children. So it would be natural to take a realist view of forces, but a nominalist view of the average Londoner. There's this question about all the entities talked about in science. A classic case is numbers, so that's a very difficult one. Are there such things out there as numbers or are they just a way of speaking about the divisibility of things into parts or something?”

I’m a nominalist about spectrum: I believe “spectrum” is just a way of speaking and does not have a referent in world. Most people, on the other hand, seem to be knee-jerk realists – “Of course there's such a thing as spectrum!” – though often when you start digging they become nominalists: “Of course, I don’t just mean frequency, there are lots of other factors…”

This fine philosophical distinction matters: If one takes the realist position, you behave as if there is a resource (spectrum) to be divided up and allocated, which leads ineluctably to radio licenses defined by hard frequency boundaries.

The nominalist perspective offers another way to thinking about the situation – for example, coordinating the operation of radio systems – that is just as valid. From a nominalist perspective the “coordination” view is just as (in)valid a perspective as “resource” view, and radio licenses don’t have to be defined primarily in terms of frequency and geography.

(See also my earlier post “Newton, Leibnitz and the (non?)existence of spectrum”. There the distinction was between the “absolutist” position that time and space are real objects in themselves, or the “relationalist” position that they are merely orderings upon actual objects that do not exist independently of the mind that is making the ordering.)

How many poor people are there?

2009-08-15T12:33:00.000-07:00

Scott Forbes linked me to a thought-provoking 2005 article titled "How not to count the poor"by Sanja Reddy and Thomas Pogge at Columbia University (PDF). The bottom line is that the simple question of how many poor people there are in the world is surprisingly hard to answer.

Reddy & Pogge argue that "[t]he World Bank’s approach to estimating the extent, distribution and trend of global income poverty is neither meaningful nor reliable. The Bank uses an arbitrary international poverty line that is not adequately anchored in any specification of the real requirements of human beings. Moreover, it employs a concept of purchasing power "equivalence" that is neither well defined nor appropriate for poverty assessment. . . In addition, the Bank extrapolates incorrectly from limited data and thereby creates an appearance of precision that masks the high probable error of its estimates." Furthermore: "There is some reason to think that the distortion is in the direction of understating the extent of income poverty."

(A rebuttal by Mark Ravallion at the Bank can be found here.)

Their alternative: construct poverty lines in each country using a "common achievement interpretation". Such poverty lines would use the local costs of achieving universal, commonly specified ends like being adequately nourished. (Ravallion argues this is pretty much what countries already to do create national poverty lines.)

Reddy & Pogge argue that such poverty lines "would have a common meaning across space and time, offering a consistent framework for identifying the poor. As a result, they would permit of meaningful and consistent inter-country comparison and aggregation."

The catch seems to be that such an approach requires one "to carry out on a world scale an equivalent of the poverty measurement exercises conducted regularly by national governments, in which poverty lines that possess an explicit achievement interpretation are developed." This is difficult politically, since a common core conception of poverty will have to be agreed, and financially, since local poverty commissions in each country would have to be funded to construct and update poverty lines over time.

The authors don't claim that their metric would lead to substantially different, or better, policies. Better then, perhaps, to spend money on poverty-focused development assistance rather than improving the metrics. However, the Bank should be more honest about the flakiness of its numbers by at least not reporting them "with six-digit precision".

Is it, or isn’t it?

2009-08-01T11:48:00.000-07:00

Humans are inveterate classifiers. We can’t help ourselves, it seems: we just have to put things in hard-edged categories. Computing might help to blur the edges in a useful way.

An update on the Pluto controversy in New Scientist is a case in point. Discoveries of exoplanets and the anticipation of Earth-size objects in the Kuiper belt make the argument increasingly irrelevant, but yet even professional astronomers seem caught up in arguing for one definition of planets or the other.

Sensory systems like ours are complicated webs of classifiers: whether objects are moving or still, whether movements are animal-like or not, whether something is a face, whether a sound is speech or music, whether someone is a member of our group or not, and endlessly on. Categorization is innate and unavoidable.

But once embedded in culture, it can quickly spiral out into fraught territory. Problems arise because classification has consequences, often monetary, often political. Is that bond AAA or AA? Is that car a clunker or not? Is so-and-so in a special group, or not? Is that judge biased?

The difficulty arises because there are so many parameters that could be used for any classification; people argue about which parameters should count. Does roundness a planet make, or size, or not orbiting around another one, or having swept its orbit clear of other rocks? Cognitive limitations (the four-or-less rule, see e.g. Halford et al. (2004), “How many variables can humans process?” Psychological Science, 16, 70-76) mean that we end up picking a few criteria from the many – too few. And then we require that each criterion must yield a yes/no result, which even for hard science classifications can be contentious: what does it mean for a planet candidate to have “a nearly round shape”?

Computing can help by allowing many more criteria into the mix, and allowing them to vary continuously. This is an application of Edward Tufte’s design strategy “to clarify, add detail,” which he introduces in Envisioning Information (1990, p. 37) with the example of The Isometric Map of Midtown Manhattan. Human nature means we may be a little uneasy with the result, but perhaps we can learn to live with it; most people are comfortable nowadays with weather forecasts that say there’s a 50% chance of rain tomorrow (although many may not actually understand what it means ...).

Hiding the criteria has its own dangers. As Bowker and Star argue in Sorting things out: classification and its consequences (1999), any classification encodes a world view, and even “simple” classification systems succeed in making themselves invisible.

Still, with a little more computing we could, in response to the question “Is it, or isn’t it?” answer in a rigorous way, “Ish.” Computers can handle composing dozens or hundreds of continuous criteria in ways our (conscious) brains cannot.

Boggle: measuring how sunshine alters the moon's orbit

2009-07-29T08:21:00.000-07:00

The July 11, 2009 issue of New Scientist ran a series of stories of why the moon still matters to astronomy, in celebration of the thirtieth anniversary of the Apollo 11 landing. One of them, by Stuart Clark, described how reflectors left on the lunar surface are helping test Einstein's theory gravity. Laser pulses bounced off these reflectors allow researchers to measure the distance between the earth and the moon to a precision of a few millimetres.

At such accuracy one has to factor in the effect of solar radiation pressure "which pushes the moon's entire orbit from its calculated path by about 4 millimetres."