I was listening to NPR in my car this evening and they were talking about the Supreme Court's ruling today in FCC v. NextWave Personal Communications Inc. From what I understand (I don't have the opening briefs and have not read them, so I am going off of what NPR said), in the middle of the tech boom, when the FCC was auctioning off licenses in an effort to equal the playing field for smaller companies the FCC allowed people to pay for spectrum licenses in installments, much like most of us buy a house. NextWave bought a few spectrum licenses from the FCC in this manner but was unable to secure the funding required, became late in its payments, and then filed for bankruptcy. The FCC then took back the licenses to resell. It was this taking back the license that brought the court case and today the Court said that it is not legal for the FCC to re-auction off spectrum after a company files for bankruptcy.

The rules that regulate bankruptcy say that in bankruptcy the property of the bankrupted cannot be taken away by their creditors, in this case the FCC. What bothers me about that is that the Court essentially ruled that spectrum is property that the government is selling off, not that the government is holding spectrum in trust of the American people ensuring its fair, equitable, and efficient use.

It seems to me that the Court ruled very much against Open Spectrum without even knowing it. By saying that spectrum is private property that they government regulates (much like it does for land) the Court is ruling against the idea that spectrum is a resource that belongs to everybody.

-Grant M. Henninger
[ dram.teamslack.net ]

> The rules that regulate bankruptcy say that in bankruptcy
> the property of the bankrupted cannot be taken away by
> their creditors, in this case the FCC.

"property" or "assets"?

This may seem a niggling point to you, but it won't be to a bankruptcy court. Nextwave need not have purchased "property" - they may simply have purchased the authorization to use a certain chunk of spectrum for a certain time. When they account for the money they spent, they *must* also reflect the asset they acquired, or Securities and Exchange would be on their case. That assumes they could even pass an audit, which is unlikely - this is first year accounting stuff.

The asset - in tihs case, the authority to use the spectrum - cannot be reclaimed and resold by the creditor - the FCC. But this does not make the spectrum itself "property".

It could be similar to a long-term contract for freight hauling on a railroad. As a creditor, the railroad cannot reclaim the sold space on the railroad rolling stock and sell it elsewhere - but neither does that imply that the holder of the contract owns part of the railroad.

Even the first page of the decision makes clear that Nextwave did not hold the spectrum - it held a license.

I think the crowd here is getting nervous because you
want us to assume that the purchase of a license automatically means that someone has "bought the spectrum". As the licensing of software has made clear, that simply is not the case, and your insistence on this point is going to damage your credibility.

You don't need to go there. If you can show that your ideas for technology in receivers/tranceivers results in a vast increase in realized spectral efficiency, then current FCC rules and market pressures will - albeit slowly - result in the adoption of that technology and better use of the spectrum.

A few notes:

1] Jeremey, fortunately, not everyone thinks we over stated the case. It has been quite favorably received by many. It has raised the hackles on the backs of few necks as well. However, I would rather under promise and over deliver. Our FAQ was intended more for a "lay" audience than a technical one. We plan more detailed essays for technical reader as we move forward.

2] Repeaters: Our view is that cooperating end points forward everyone's messages closer to their destination. This reduces radiation levels per user. Lower power output not only has a positive result on the health issues, but also allows batteries to last longer. End points just whisper at the lowest levels possible to reach the next repeater. This is key to achieving the paradoxical result that the capacity of the EM spectrum increases as users are added to the network.

3] We are not throwing up our hands. We are actively looking for "radio havens" where demonstrations can be run. There are already interesting developments at the MIT Media Lab and, I believe, the GNU radio team is doing very interesting work as well. We do have to be realistic about what can be done legally. Part 15 has many limitations that are constraining, so we are looking for a much broader opportunity.

4] Shortly we will be blogging a note on our view of what a desirable regulatory approach would be. Stay tuned.

Regards,

Jock

Jock, re point 2, Chris (in /m25) was making the point that while part 15 forbids "repeaters", it could be argued that what you're talking about is really a transceiver, not a repeater. That is, a repeater is a receiver directly connected to a transmitter, which just repeats the raw signal it receives. What you're proposing involves a receiver, a processor, and a transmitter, so while what it's transmitting depends on what it receives, it's not just stupidly re-transmitting the raw signal that it received.

I'm not a lawyer, let alone an expert in FCC regs, so I don't know if Chris's argument would hold up, but it sounds plausible to me.

> Our view is that cooperating end points forward
> everyone's messages closer to their destination.
...
> End points just whisper at the lowest levels
> possible to reach the next repeater.

So this really IS like a Cybiko (http://www.cybiko.com/ ) using CDMA-style power control - possibly with CDMA-style coding?

The very existence of the Cybiko tends to suggest that you have mis-interpreted something - or not made clear what you are trying to do. Pick one, and fix it, then explain.

Very impressive FAQ, Jock. I agree that SDRs will be much more common, if not the rule, in the future. High bandwidth wireless will solve a whole lot of problems, as you say.

And I certainly agree that much of the current licensing system works against innovation. Indeed, as one of the commenter's pointed out, these are licenses for a fixed term (thus the Supremes applied traditional bankruptcy law to that asset of NextWave). A huge regulatory revolution of the last 20 years was to change the law to turn these licenses for a fixed term into a de facto property right, not because it says that, but because the licenses became renewable with effectively a post card.

Public auctions were a far better system than the prior freebies (public gets some value), but again no term was set with teeth (e.g. you get it for 15 years and then we re-auction).

I too value the clarion call when you are fighting for fundamental change, but I would caution you a little about overstatement of both the technology and business hurdles to get to your goal. I don't think most of the infrastructure for this is already deployed as the FAQ asserts, or that it will essentially be a very cheap or free good, as the same section implies. Similarly, I believe a lot of work needs to be done on SDRs and related technology before they are ready for commercial deployment.

Finally, having seen a NASA report and video of UWB transmissions killing a handful of critical avionic systems on a commercial jetliner last year (albeit parked on a runway), I'm not so sure that we can simply open the floodgates. And you wouldn't want to anyway. When I was on the Hill we fought with DOD, and with the NAB, but we never, ever, tried to go at them when they were allied!

The strategy of the Microsofts and Intels has been to push the unlicensed spectrum door wider and wider (i.e. adding more and more spectrum). This has limitations, but it means you don't have to attack all the castles at once.

Keep up the great work!

Short replies:

1] For repeater I would have been more accurate if I had used the term "smart router" -- ie knows where it is and thus something of its environment. I regret the confusion.

2] As for FUD demos, yes they actually prove what is wrong with current bandwidth dependent solutions: They are very vulnerable to intentional harm. Dumb UWB is not what we are talking about. We respectfully suggest that distributed, cooperating, "smart" UWB, using GPS for environmental clues, and spectrum underlay for 'invisibility', offers a much safer, more robust and secure solution.

3] Cybiko is an interesting and informative step, but very incomplete, towards Open Spectrum. As is 802.11X. Itis important that they show new models. A key question is their ability, or inability, to offer scalable capacity that increases as the number of users increases.

4] The key is to realize, to borrow wildly from an old political saying: It is the architecture, Stupid! It is not what happens in the EM space. It is what happens in the encoders and decoders of the messages carried by the photons. To improve the quality of the signal/content translation process we need to look at the translators. It is here that regulation and innovation can be appropriately applied to great effect. Looking any p;lace else is like looking for the car keys under the street light when you know you dropped them back a ways in the dark and in the snow drift.

5] Is it appropriate to compare the nature of the signal carrier in the spectrum to concrete objects? No. But we do and so we get faulty analogies that suggest spectrum has "property like land" which is "scarce" and can be "privatized" and "priced" .

6} What does,actually, carry the signal through the spectrum? If it is photons, we know that they can not interfere with one another and thus all of the analogies in #5 can not be true. The implications for our current approach to wireless communications regulation is interesting.

I've asked others on the team, with greater technical chops, to make more detailed replies.

Newly blogged at <www.greaterdemocracy.org >

The Regulators and the Candlestick Makers

Not too surprisingly, apparently only a very few of the recent comments to the FCC on spectrum policy reform have taken the Open Spectrum approach: David Weinberger's comments [based upon his work with the Greater Democracy group] and those of the EFF are two examples that look forward and not backwards.

There are two related questions here:

   1. Are the Congress, the NTIA and the FCC regulating to protect the status quo?

   2. Does the emperor have no clothes?

--- snip ...........>

I like the freedom talk. Unfortunately this really sounds like a sneaky sales pitch from a "wireless communications"
i.e. cellphone company. Exclusive use of radio bands is necessary because interference is a reality. True, FM demodulators will reject in-band signals that don't overpower them, but you can't have 2 stations on the same frequency in the same area without interference. This reminds me of the push for low-power FM--sounds like a new freedom, but now I get syndicated religimus fanatics where I could once get decent college stations. Digital computer geeks do not understand that this is an analog world, and their digital universes are limited by analog reality. You twist some techno-speak then con the majority of people into believing your vapor-ware. Another thing: not all bands are the same. They have different propagation characteristics, which is why you can get Chicago on the AM from 1000 miles away. The exclusive use of a frequency (the "clear channels" which only a few stations have) makes this possible. Shortwave stations which can be heard worldwide use frequencies and times arranged by international agreement. You could replace all radios and TVs with computerized frequency-hopping receivers
then make this change, but you simply can not "open
spectrum" without making billions of radios and TVs obsolete. Furthermore, "freedom" is NOT being available on the cell anywhere, anytime. More sacred antennae on sacred mountains is not an improvement on our world. Don't believe the hype.

unClem: If, by "interference" you mean the problems that occur when photons collide with electrons in the send/receive hardware etc., then I agree that these photon/electron interactions cause poorly architected hardware with legacy technology systems some real problems. You are of course right that when two photons interact with an electron they can produce other photons with surprising behaviors.

We see these interactions every day. That is how signals interact inside antennas. (which are seas of electrons). That is how signals reflect and refract. You can get frequency doubling, you can get tropospheric tunneling, etc.

QED explains everything about radio, in theory, and it is remarkably precise. and predictive. It's one of the most well established parts of science.

However, it does appear from QED that the problem is not in the signal in EM space. So, if you meant by "interference" photon with photon interactions in EM space, I do not think that is actually possible. Are you suggesting that you want to re-write the work of Feynman, Schwinger and Tomonaga et al?

So the real question is why we write laws and regulations that ignore QED - that treat photons as if they exist in an Ether that can be divided up and "owned".

[Thanks to David Reed for his technical contributions to this note.]

Jock, part of what bothers me about the "Open Spectrum" movement, is your parrotting of QED and other physics without (apparently) understanding what it means. It makes it look like you're trying to snow people. Ted Nelson coined the term "cyber-crud" for this sort of argument in the computer field. It makes me distrust you even in the areas where I agree with you (of which there seem to be many)!

It's true that photons don't interact with other photons in free space. So what? No one is claiming that they do. You claim "photons can only interfere when they interact with electrons" like that's something really rare and irrelevant. But the ONLY time we have any contact with photons is when they interact with electrons! Photons interacting with electrons are _IT_. That's how we transmit radio signals, receive radio signals, even how we see.

If a few photons from a source we're interested in (the "signal") interact with some electrons (the "antenna") at the same time as large numbers of photons from a source we're not interested in (the "noise") also interact with the antenna electrons, practically speaking we may not be able to distinguish the "signal" from the "noise". This is what I (at least) mean by "interference". So stop pretending that we're talking about photons interfering with other photons in free space. When I talk about photons interfering, I mean at _my_ receiver, because I don't _care_ what happens anywhere else!

Based on your Open Spectrum writings, I'd be inclined to agree with uhClem, that you're spewing these technical half-truths to con society into giving you something you (or your backers) want, but can't get honestly. You're painting this utopia where all sources of RF are smart and polite, and saying that if we'd all just throw away all our TVs, radios, and other wireless electronic communication devices, everything will be wonderful. While I agree that smarter radios will do a better job, and regulations should be changed to allow that, I want to know why we should trash $billions of equipment so that _then_ we can get something better. If it's so much better, why can't it be demonstrated today?

And if what you're selling is so much better than the status quo, why can't you just show us? Instead of babbling about how photons in free space don't interfere, which while technically accurate, is irrelevant.

Edited by author 02-07-2003 06:19 PM
Open Spectrum certainly makes a lot of sense when you're talking about a world filled entirely with intelligent devices that are courteous, don't collide with one another, speak softly, and can hear the faintest whisper themselves. But the reality is we're living in a world filled with stupid, boorish devices: transmitters that shout, receivers that can't hear, etc. Even worse, these devices are supported by an infrastructure of corporations and lobbyists with very deep pockets.

How do Open Spectrum advocates intend to deal with this reality? I have yet to hear a good (i.e. specific) technical explanation as to how this massive dumb-to-smart phase-in of devices can be accomplished. According to the Open Spectrum FAQs, it cannot be phased in. It has to be done all at once. Good luck.

This doesn't provide much guidance for others who see the merits of Open Spectrum and want to do something to support it.

On the other hand, there are prominent Open Spectrum advocates like Dave Hughes and big tech companies like Intel and Microsoft who are pushing for a gradual phase-in. They're saying let's get any unlicensed spectrum we can get, do clever things with it, and then push for more.

Who are we supposed to believe? I'm inclined to believe the latter.

So here you have one faction of the Open Spectrum movement saying one thing, another faction saying something else, and NOBODY offering any technical specifics. Confusing to say the least.

Photons, shmotons! Yes, waves can go right through one another without interference. Waves of a same frequency being received by one radio do cause interference. Period.
We should be talking about a better internet, with access uncontrolled by AOL Time-Warner, or whoever. Fiber to the home could offer unimaginable bandwidth. Here, photons can go both directions at once. Wavelength-division multiplexing
and other ingenious scemes now make terabytes per second data rates state-of-the-art on fiber links. I'm talking thousands of gigabytes per second on a single fiber. That's what we consumers could have. Filling the "ether" with more electromagnetic energy is not desirable. The wave/particles may go right through each other, but they do interact with matter. As a matter of fact, the CTIA is no longer able to hold back the tide of research that shows harmful effects to the brain from cell-phone radiation. By the way, do you know who Nikola Tesla and Edwin H. Armstrong were? Do you understand how radio works?

Just to pick up on the easiest part of aok's message (m/38), the FAQ explicitly says (#28) that Open Spectrum would be phased in.

Look, folks. Open Spectrum is not a technology proposal. It's a policy proposal grounded in technology. There are several key tenets:

1. Communications by EM in free space are becoming more important. Devices move, and wires don't.

2. Today's EM communications architectures are wasteful, largely because they are based on pre-allocation of fixed channels and frozen technical architectures based on old technologies, reinforced by regulators that have been captured by incumbents.

3. Today's EM communications architectures are not rapidly scalable or adaptive to physical conditions as they occur in the natural environment.

4. Today's EM communications regulations are grounded in a theory that links "services" to "frequencies" in "areas". None of the elements of this theory match physical reality, nor do they match our understanding of information. E.g. by service is not a sensible way to allocate frequencies. Frequencies and areas show bad understandings of physics, because they don't accurately model the most useful coding of information into EM disturbances.

5. Presupposing that more communications involves more EM radiation which might be harmful to health misses a crucial point - the current radio systems expend energy many orders of magnitude more get bits delivered to the right place, on time. One can greatly increase the amount of effective communication capability using EM while drastically reducing the amount of EM energy incident on individual people and objects.

Open Spectrum's goal is to make room in policy for innovations that can demonstrate these things. Unlicensed spectrum is only a first step. If all that happens is that bad legacy system designs move into the unlicensed spectrum, it will be worthless.

dpreed, /m41 sounds a lot better, clearer, and more sensible to me than the Open Spectrum FAQ. Only one problem at the end, where you can't seem to decide what you want:

"Unlicensed spectrum is only a first step. If all that happens is that bad legacy system designs move into the unlicensed spectrum, it will be worthless."

Um, if it's unlicensed, how are you going to prevent "bad legacy system designs" from moving into it? That's the missing piece I see in Open Spectrum: how do you prevent "bad legacy system designs" from causing problems for new, more polite and intelligent system designs? So far, the only answer I've heard is that there won't be any problems because, well, there just won't.