The Iron Laws of Network Cost Scaling

The shape of the cost curves that show up as we build and run communications networks have properties that seem counterintuitive to many people, but that have been surprisingly consistent across lots of different technologies since at least the days of the telegraph, and probably further back than that.

Herewith, the Iron Laws of Network Cost Scaling:

1. Upgrade cost per increment of capacity decreases as capacity rises.

2. Network costs scale primarily with the number of troubleshooters required to run them, not with capacity.

3. Under market pressure, network pricing evolves from metered to flat-rate.

When you learn to apply all three of these together, you can make useful qualitative predictions across a surprisingly broad set of real-world cases.

The easiest way to see why upgrade cost per increment of capacity decreases as capacity rises is to think about the high capital cost associated with laying the first cable from A to B. You’re going to have to pay to dig a trench and lay conduit, or put the functional equivalent of telephone poles in a right-of-way. If we’re talking wireless, you need two antenna towers – OK, maybe one if the start end is on your net already. Trenches are expensive; rights-of-way and poles are expensive; towers are expensive.

But once you’ve got that physical conduit or poles or towers in place, pulling replacement wire or upgrading your radio repeaters is much less expensive. As your tech level rises, you (mostly) stop having to do that, even; you find cleverer ways to squeeze bandwidth out of fiber, copper, or air by using denser encodings, better noise cancellation – better algorithms. The action moves from hardware to software and upgrade costs drop.

As a very recent example of how the shift from hardware to software affects developing communications networks, the differences between the two major fourth-generation wireless data technologies, WiMAX and LTE, are so slight that the same hardware, running different software, can support either. This means that on any timescale longer than that required to push firmware upgrades to your repeaters, the differences between the two aren’t of consequence for planning.

The amortized cost of network capacity gets cheaper fast, partly due to the first Iron Law and partly due to the Moore’s Law cost curve of hardware. Skilled people on the spot don’t get cheaper. Therefore the dominant cost driver is salaries for people required to run the network. Furthermore, hardware/software maintainence costs tend to be low for the links (which are simple) and high for the switching nodes (which are complex).

The consequence is that cost scales not with network capacity but roughly with the number of routers and switches in the network, and is primarily salaries for people to watch and troubleshoot the routers and switches. This fact is well known to anyone who has ever had to actually run a data center or a network; it’s a reality that recurs very forcefully every time you have to pay the monthly bills.

There’s actually more we can say about this. In a roughly scale-free network (which communication networks with smart routing tend to become; it’s an effective way of maximizing robustness against random failures), the node count is coupled to the link count by about n log n. This means that as network reach or coverage (proportional to the number of leaf nodes, aka customers) rises linearly, the number of interior nodes (which counts routers and switches) actually rises sublinearly.

This is all in stark contrast with most peoples’ intuitions about network costs, which heavily overweight capital expenditures, heavily overweight bandwidth cost, and predict linear or superlinear rises in administrative costs as coverage increases (the “high-friction” model of network costs). But with a more correct model in hand, we can approach the third Iron Law: under market pressure, network pricing evolves from metered to flat-rate.

This is certainly the way network pricing has moved historically. Can we say anything generative about why this is so?

Yes, as a matter of fact, we can. We saw before that as customer count rises linearly, the major cost drivers in the network (router and switch count, and salaries for people to watch them) rise sublinearly. But to do per-transaction metering you have to store, manage, and process an amount of state that rises directly with customer usage – that is, linearly. This means that, especially on a maturing network, the cost to meter usage rises faster than the cost to serve new customers!

The qualifier “under market pressure” is important. Customers really don’t like being charged for both usage and maximum capacity. But they dislike being charged for usage more, because it makes their costs harder to predict (and usually higher). Comms providers are ruthless about exploiting the myth of high network friction to justify high prices and metering, and they generally get away with this for a while in the early stages of a new communications technology. But at least two things cooperate to change this over time, both actually effects of the widening gap between that mythical “high-friction” cost curve and the actual one.

One is that metering overhead rises as a proportional drag on on per-customer revenue (and thus profits) as the network’s coverage increases. Again, this is strictly predictable from the fact that the cost of service rises more slowly than the cost to meter. The other is that profit margins, as in any other sort of market, tend to get competed down towards actual cost levels. Telecomms vendors, like all other producers of non-positional goods, feel constant pressure to price in a way that actually matches their cost structure more closely.

Usually this means pricing by maximum capacity with no metering. Eventually, as link capacity reaches a level the average customer isn’t capable of saturating, flat-rate unlimited starts to make more sense.

Thus, communications-network prices have a very specific trajectory that’s repeated over and over with new technologies: from metered by transaction to billed by maximum capacity to flat-rate unlimited. The providers resist each change as ferociously as possible, because each one is accompanied by a shift to decreasing profit margins on increasing volume, but the underlying logic is inexorable.

118 comments

  1. “As your tech level rises, you stop having to do that, even; you find cleverer ways to squeeze bandwidth out of fiber, copper, or air by using denser encodings, better noise cancellation – better algorithms. The action moves from hardware to software and upgrade costs drop.”

    My initial thought on that is “Well, sort of.” As bandwidth requirements and capacity rises, hardware, especially the hardware architecture of a given node (router, switch, and really, they’re the same thing at a certain point), becomes crucial. It really is to the point where the hardware and the software on high-end networking gear is so tightly coupled that they’re becoming indistinguishable. It’s not that the software is supplanting the hardware, they’re becoming more of a gestalt than ever before. Eventually, however, you run out of backplane, or wattage, or physical interface capacity, and need to refresh the hardware.

    1. >Eventually, however, you run out of backplane, or wattage, or physical interface capacity, and need to refresh the hardware.

      Perhaps I wasn’t clear. I wasn’t talking about things that have backplanes here, but the link media, the physical layer of wire.

      I shouldn’t have said you get to stop upgrading the link media entirely, because that’s not so. On long enough timescales you do. The real point is that these upgrades are so rare that on the normal net-present-value horizon of cost accounting they never happen. Optical fiber is, of course, the main case in point here.

  2. “Comms providers are ruthless about exploiting the myth of high network friction to justify high prices and metering,”

    And cellcos *still* get away with charging at *all* for SMS, much less thousands of dollars and cut-off service.

    What’s that “cut the jumper and the VAX goes twice as fast” entry in TJF?

  3. Baylink

    They get away with charging because the customer push back isn’t as great. How long has SMS been around versus how long telecommunication has been around?

  4. And cellcos *still* get away with charging at *all* for SMS, much less thousands of dollars and cut-off service.

    Welcome to America, where nickeling and diming strategies have been the telcos’ greates telecommunications innovation in ten years.

    Europeans know the value of infrastructure; Americans seem not to.

  5. @BigFire: Yep. And to add to that the major wireless providers have added flat-rate pricing, largely in response to push-back from an ever-more savvy customer base. People love flat-rate pricing, and they’ll continue to demand it. Furthermore, price wars have already broke out, with the early flat-rate pricing set at $100, while smaller providers like T-Mobile are offering $69 flat-rate pricing, and even smaller providers like MetroPCS are offering $49 flat-rate pricing. Market pressure will continue to push all the vendors towards that magical $50/month that permeates ISPs. (This is no coincidence — it’s all about what the market will bear.)

  6. “Perhaps I wasn’t clear. I wasn’t talking about things that have backplanes here, but the link media, the physical layer of wire. ”

    Then it would be more accurate to say that at a certain tech point, one doesn’t have to replace the physical layer, but merely replace the driver of said layer. That being said, you were talking about software supplanting hardware, and there’s not a lot of software that runs directly on optical fiber that I’m aware of (grin), so I could only assume you were talking about the devices that *drive* that physical layer, the routing/switching devices.

    1. >Then it would be more accurate to say that at a certain tech point, one doesn’t have to replace the physical layer, but merely replace the driver of said layer.

      That’s correct, sorry for being less than pellucidly clear.

  7. I would also propose a qualification point to all of the above and that is:

    4. “The points above hold, until an inflection point is reached; either in capacity, or in technology; when the cycle starts again”.

    Presuming the network as a continuous thing, and the underlying technologies providing it as fungible. If you want to look at a technology inflection point and instead say “ok, that’s actually two different networks” then it works without the qualifier; but I think historically this isn’t always the case…. and after all, this is intended to be an iron law.

  8. @Chris

    We’re going to need a more precise formulation than “inflection point”. That is a simple mathematical statement about the derivative of a continuous function. You seem to be saying, “the system behaves as ESR describes, until it doesn’t behave that way”. Tautological much?

    I’m not sure if I’m going to agree or disagree with your argument, when and if you elucidate these bizarre statements, but at this point they aren’t even wrong.

  9. Usually this means pricing by maximum capacity with no metering. Eventually, as link capacity reaches a level the average customer isn’t capable of saturating, flat-rate unlimited starts to make more sense.

    Never underestimate the capability of the average consumer to saturate a link. If everybody got TB/s quantum-teleportation-to-the-home tomorrow, those links would be soaked with porn downloads and stolen copyrighted content by Memorial Day.

  10. Sorry, Eric, but these laws are not “iron laws.” In fact, many only apply to wired and unregulated networks.

    Shannon’s Law dictates that the capacity of wireless networks is ultimately dependent upon spectrum. Because one the Shannon Limit is directly proportional to bandwidth, one cannot realize the same exponentially increasing economies of scale in wired networks as in wired ones under the current spectrum licensing regime. Auctions of spectrum occur infrequently, and most spectrum is bought not to be used but to foreclose competition. Attempting to expand a wireless channel usually causes one to run into a competitor who will not sell or license for any amount of money. Therefore, there’s often a hard limit — imposed not by raw physics, but by government policy — on what one can do. This is yet another argument for the abandonment of the current auction/spectrum-as-property regime.

    1. >This is yet another argument for the abandonment of the current auction/spectrum-as-property regime.

      You think you’re refuting my iron laws. I think you’re explaining why the current regulatory regime is unsustainable, in terms very similar to what I’d use.

  11. I suspect Chris was referring to the phenomenon esr alluded to in http://esr.ibiblio.org/?p=2024#comment-254844 as the “inflection point”…that is, the point where the events that “on the normal net-present-value horizon of cost accounting […] never happen” get around to happening. That is, the point where the relatively-more-expensive stuff has to be either ripped out and replaced, or at least supplanted by newer stuff that isn’t a pure-software upgrade from the old stuff.

    Some will call the network after this point a different network than the one before this point, and some will assert they’re really the same. Both views are correct…they’re just held by people looking at different layers of the OSI model. Either way, the event tends to reset the cycle when it occurs, although not always all the way back to the beginning.

    (Chris should feel more than free to correct me if I’ve misinterpreted his meaning, of course.)

  12. Brett Glass: You are entirely correct.

    I would add, however, that spectrum starvation is only a problem in VERY dense areas at present, and that the FCC will start loosening up more spectrum by hook or by crook as people start complaining about poor service in major metro areas. They did it before, they’ll do it again, because the ultimate destiny of every FCC regulator is to go to work for somebody in the industry they regulate and pissing off your future employer isn’t an option.

  13. ESR or anyone else:

    Have any idea how broadband over power lines plays into this?

    I have heard that it could be good, but I have also heard that it causes way too much interference with other devices.

    Anyone know the score?

    ESR says: All I know about it is that the proposals have RFI implications that make ham-radio guys crazy. I do not have a judgment formed about whether this is sufficient reason to oppose it.

  14. Darrencardinal: I can tell you that Motorola has pretty much abandon it. It was too unreliable, and as ESR points out, the stuff was noisy as hell. They stopped selling the LAN product about 2 years ago and I think they abandon the WAN product about the same time.

  15. Excuse me for a moment while I point to the K5ZC license plate on the wall above my ham radio station.

    The fundamental problem with broadband over power lines (BPL) is that a power line makes a terrible transmission line for RF energy, and you need to send RF signals down the line in order to carry signals of more than a few tens of kilobits per second. This is because power lines are, at RF frequencies, essentially antennas. On top of that, you can get away with a lot of things at 60 Hz that will kill you dead at even 5 MHz. You can’t just get out your Kleins and twist two wires together and call it good.

    The only way to keep a BPL installation from interfering with the licensed users of the radio spectrum is to notch out the frequencies used. The HF spectrum is as full as any other part of the radio spectrum, and its users don’t like being interfered with any more than anyone else. It’s fundamentally impossible to do BPL without causing RF interference.

    As if that weren’t enough, the biggest, best known BPL installation, in Manassas, Virginia, never had more than a few hundred customers at most. It simply wasn’t fast enough or reliable enough to compete.

    BPL was a bad idea from the start, and the FCC should have given it the mercy killing it deserved instead of trying to hold it out as the solution to broadband access problems. By now, the market knows better, and the FCC is finally coming to its senses.

  16. > Europeans know the value of infrastructure; Americans seem not to.

    BWAHAHAHAHAHAHAHA!

    Is this supposed to be an ironic statement? Or do I have to point out the fact that you are usually driving on a grid in the US and usually not in places that have been heavily urbanized longer?

    ESR says: No, this is just stupid reflexive America-bashing. Which, coming from Jeff Read, is about as surprising as the sun coming up.

  17. Is this supposed to be an ironic statement? Or do I have to point out the fact that you are usually driving on a grid in the US and usually not in places that have been heavily urbanized longer?

    Hey, at least the roads in those places are well maintained. As are the trains and railways.

    And I live in Boston, which oftentimes feels like Merrie Olde England, particularly in its street layout… :(

  18. In Europe some telco’s charge for SMS’s, but others have flat rate plans. One example is Italy, where unlimited SMS plans are common. One effect of this is that about a third of all SMS messages contain viruses and a sizeable portion of the rest is Spam, which does put quite a workload on the Telco’s. That’s a good argument for still metering SMS traffic.

    The trend however, even within Europe, is for flat rate plans, however a lot of the most popular mobile plans are still prepaid plans, and there you really still have to meter. You can keep it simple though. A popular mobile internet plan is “X per day, for unlimited use each day you use it, 0 on days you don’t use it.”

    1. >That’s a good argument for still metering SMS traffic.

      No it isn’t. It’s an argument for rate-limiting SMS sends. Spammers have to spew thousands of these per second for the tactic to be either effective or annoying; that can easily be prevented without metering.

  19. K>One effect of this is that about a third of all SMS messages contain viruses…

    Hmmm. A virus in 144 (or whatever) ascii characters. I’d like to see that code ;^).

    Carrier: “About 98% if the email on the internet contains virii or spam, so you’ll have to start paying $0.10/email to make sure you’re not sending spam.”

    Customer: “Screw you.”

  20. I’d just like for my wireless carrier to use SPF on emails gated to SMS. That would slow the spams to a trickle.

    I think the viruses “in” SMS are actually hyperlinks in the SMS that, if followed will instruct a browser to download the virus.

  21. I seem to recall that there was a root vulnerability in iPhones some time back that could be exploited with a malicious SMS message.

  22. I think the viruses “in” SMS are actually hyperlinks in the SMS that, if followed will instruct a browser to download the virus.

    Which shouldn’t be a problem because most phones require apps to be signed, right? Or am I missing something?

  23. Jeff Read: Was it SMS or MMS? I could understand an multimedia message, but SMS would seem surprisingly short.

    Of course, there are some tools out there that do things on your phone based on an SMS message. Like: “application: action-or-notification parameter”. I’ve seen a few of these slip past a crashed app on my Android, so I guess that would be a possible SMS vector (in a round-about sort of way).

  24. Jeff Read: Thanks. I’ll delve into it tonight, as I’ve not paid much attention to this stuff until recently – my Palm Centro was secure for the same reason that DOS is secure… it was just too stupid to be a useful target :^).

  25. @Matt, that is exactly the point I was attempting to make.

    To Jess’s point, I made a classical argumentative error: I presumed an understanding of context, a common frame of reference, and shared knowledge base; rather than explicitly stating a frame of reference, and defining terms and conditions associated with that frame.

    In the context of structured logic, Jess is correct, I committed tautology (it IS a crime against both language and logic after all, it should be something one commits, like a crime or an atrocity).

    In the context of systems engineering and analysis however, the concept of “inflection points” (also called “break points” or “cascade points” depending on the specific situation) is well understood as a tangible phenomenon; elucidated succinctly by Matt above.

    1. >In the context of systems engineering and analysis however, the concept of “inflection points” (also called “break points” or “cascade points” depending on the specific situation) is well understood as a tangible phenomenon; elucidated succinctly by Matt above.

      I understood what you were driving at, but you should never use “inflection point” that way again. That term has a very specific meaning that does not imply a discontinuity in whatever curve you’re talking about.

  26. True enough Eric. It’s a bad habit, to fall into institutional or professional jargon. It creates misunderstanding and should be avoided as much as possible.

    Not only that, having degrees in related disciplines (double BS in aerospace engineering and computer science, and a minor in mathematics), I DO know better.

    Though I will note, the jargonal usage of it comes from structured mathematical analysis of systems (systems in the larger sense, as opposed to nodes, or boxes etc…). It is generally understood to be the point (or points) at which the cost/performance curve of a system being analyzed changes radically.

    It is not usually a literal inflection point (the curvature doesn’t necessarily change signs), but this is close enough to the literal definition to be “dangerous”, in that it can create great misunderstanding than an obvious malapropism.

    1. >Is it bad that I had to struggle very hard not to make a “derived” pun in that last post…

      I’m an ex-mathematician myself…

  27. > Is there really any such thing as an ex-mathematician?

    So Morgan, you believe in life after death. I thought you were an atheist pagan-because-it-works like Eric (as I understand it).

    Yours,
    Tom

  28. Massive thread jack, I know. But I couldn’t resist. For some reason, it’s really fun to discuss those subjects with Morgan/Eric.

    Sorry,
    Tom

  29. Is there really any such thing as an ex-mathematician?

    ErdÅ‘s seemed to recognize their existence; they were said (by ErdÅ‘s) to have “died”. In contrast, someone who underwent actual physical death had simply “left”.

  30. So Morgan, you believe in life after death. I thought you were an atheist pagan-because-it-works like Eric (as I understand it).

    Well, if Mr. Raymond were dead, he wouldn’t be posting on his blog, now would he? Also, when you’re dead they don’t really call you an ex-mathematician, they call you ‘the late mathematician’ or ‘a deceased mathematician’ perhaps, but not an ‘ex-mathematician’.

  31. > Well, if Mr. Raymond were dead, he wouldn’t be posting on his blog, now would he?

    True, but you claimed that the set of ex-mathematicians was actually empty. Therefore there must be no ex-mathematicians by virtue of death.

    > Also, when you’re dead they don’t really call you an ex-mathematician, they call you ‘the late mathematician’ or ‘a deceased mathematician’ perhaps, but not an ‘ex-mathematician’.

    The set of late or deceased mathematicians is a subset of the set of ex-mathematicians.

    Alas, I am clearly not a mathematicians, as I am unable to come up with a joke or witty comment utilizing set theory.

    So please, think of something humorous on your own.

    Perhaps involving snails, primitive steam engines, electric frying pans, and the failed matrimonial machinations of the typical Estonian grandmother.

    Yours,
    Tom

  32. True, but you claimed that the set of ex-mathematicians was actually empty. Therefore there must be no ex-mathematicians by virtue of death.

    Would you refer to, say, Dennis Hopper (since he’s at this point actually deceased) has an ex-actor?

  33. > Would you refer to, say, Dennis Hopper (since he’s at this point actually deceased) has an ex-actor?

    No. I believe in life after death.

    Yours,
    Tom

  34. No. I believe in life after death.

    And so does most everybody else. Hence, culturally, it is not really accurate, and in some circles could be seen as disrespectful. to refer to esr, at some point in the hopefully very distant future, as an ex-mathematician after he his deceased. Unless, of course, he prefers to be called that for some reason. But once someone is a mathematician, aren’t they really always a mathematician? In fact, aren’t programmers really applied mathematicians?

  35. The set of late or deceased mathematicians is a subset of the set of ex-mathematicians

    I don’t believe I’ve ever seen Gottfried Leibniz described as an “ex-mathematician”. I’m pretty sure that folks like him are listed in the history books as “mathematicians” even after they died.

  36. “Shannon’s Law dictates that the capacity of wireless networks is ultimately dependent upon spectrum. Because one the Shannon Limit is directly proportional to bandwidth, one cannot realize the same exponentially increasing economies of scale in wired networks as in wired ones under the current spectrum licensing regime.”

    What The Fuck???

    First, it’s not Shannon’s law, it’s the Shannon-Hartley theorem, or the fundamental theorem of information theory, or just Shannon’s theorem.

    Second, it’s about the theoretical maximum information transfer rate of a channel, for a particular noise level. The channel is defined by the bandwidth and signal power.

  37. @JF Sebastian:

    First, you have a minor quibble about Brett’s terminology, couched in righteous indignation. Fine, but you don’t say anything that would contradict what Brett said. In fact, I would suspect that Brett fully understands, at least qualitatively, what you have said, but that you have completely missed his valid point that it is, in some cases, regulatory capture by rent-seekers, rather than the laws of nature, that reduces the number of bits per second that can be delivered to wireless end users.

    Imagine the case of a single user connecting to a single network, in an almost perfect world with femtocells everywhere. Further imagine that the user’s small handheld device has but a single antenna. (MIMO could certainly tweak the equation a bit at the expense of some DSP and reduced battery life for the handheld, but that still just applies a constant factor that may be somewhat greater than one, but probably can’t get very large because of the small form factor of the handheld.)

    Yes, signal power constrains the channel by affecting the SNR (possibly lowering the SNR considerably in the case of a femtocell), but what Brett calls spectrum certainly sets an upper limit on what you call bandwidth — the amount of spectrum “owned” by a particular provider in a particular location determines absolutely the maximum bit rate the provider can provide to the single antenna on the handheld device in the absence of any interference.

    Even more to Brett’s point, if the femtocells cooperated and used MIMO themselves, they could not only use beamforming to focus a signal on the handheld device while simultaneously focusing different signals on other handhelds, but could even use active cancellation techniques to reduce the noise in the channel for both transmission and reception. So at the limit, Brett’s assertion that “capacity of wireless networks is ultimately dependent upon spectrum” is not at all an unreasonable assertion, even in a near-future world with a lot more antennas and processing, especially if you take the (increasingly valid) position that the primary purpose of wireless networks is to enable bidirectional communication to untethered handheld devices.

  38. # Patrick Maupin Says:
    > So at the limit, Brett’s assertion that “capacity of wireless
    > networks is ultimately dependent upon spectrum” is not at
    > ll an unreasonable assertion,

    Sorry to drop in unannounced, but that is one of those great assertions that while true is entirely unimportant. Like calculating the maximum computational power of the universe based on the energy available, or asserting that, even though human population of the earth has grown exponentially, and average wealth has also grown exponentially, that there is an eventual limit to the Earth’s human carrying capacity. True, but entirely unimportant.

    All physical systems impose some limit, but we are so far from the limit it is entirely unimportant for all practical purposes. The fact is that the limited carrying capacity of the RF Spectrum is constrained primarily by poor allocations and an inability to innovate due to muddy property rights and rights of renters. Where small amounts of rights are allocated innovation is abounding (such as in the cell phone segments), when property rights are almost full conveyed (for example in the junk bands where regulation happens by way of controlling the physical space and what devices are allowed in that physical space) innovation has gone nuts. In military bands, where the military has complete control, there is amazingly high levels of innovation. However, in most other areas Marconi would be considered an expert innovator.

    The best use of our radio spectrum is to sell it off to the highest bidder, give them full property rights based on appropriate criteria of power and bandwidth, including limitations of trespass and nuisance, and then use the money we save to pay down some of the unfunded obligations we have to users of social security and medicare. When people’s profits are at stake there will be amazing increases in efficiency of use of the resources, and all those social security and medicare funds would be invested in private industry creating an amazing slew of new jobs.

    Fat chance.

  39. Fine, but you don’t say anything that would contradict what Brett said.

    @Patrick Maupin: Actually, he does. He’s demonstrated that Brett doesn’t understand Shannon-Hartley theorem completely. I suggest taking a gander at a presentation called How wireless networks scale: the
    illusion of spectrum scarcity
    , which was presented to the FCC Technology Advisory Council in 2002. Summarizing rather quickly, the point of the presentation is that wireless wireless networks as we’re discussing them here are subject to multi-user information theory, which often has surprisingly counterintuitive results.

    For instance, you might not realize that using multiple paths and repeaters actually increases capacity or that channel sharing actually decreases latency and jitter. Or that spectrum capacity actually increases with the number of users.

    The problem is that you and Brett are assuming that wireless communication models will stay exactly as there, while this presentation is actually positing that dynamic cooperation and shared spectrum use are necessary to maximize efficiency.

  40. @jessica:

    It’s not “entirely unimportant” — if it were unimportant then a company which owned a small sliver of the spectrum in a particular locale would currently be able to deliver unlimited bandwidth, and that doesn’t happen, and both you and Brett agree that it’s because of the current regulatory regime. Your proposed solutions are probably different than his. I, for one, don’t buy your argument about natural monopoly — that’s the kind of thinking that led us to serious trouble for years with AT&T. For my money, the best example of spectrum usage these days is the sort of free-for-all that happens in the WiFi space. With similar arrangements (with higher power limits and perhaps somewhat more stringent rules of the road) in other bands, you’d find a lot of innovation that would drive price down, which is exactly the opposite of what happens when the government grants a monopoly on part of the commons (whether purchased outright or indirectly by buying politicians).

    @Morgan:

    For you to write “The problem is that you and Brett are assuming that wireless communication models will stay exactly as there” means that either I didn’t write very carefully, or you didn’t read very carefully, because I certainly didn’t make the assumption you accuse me of — I wrote about a near-perfect world with femtocells everywhere cooperating in beamforming MIMO. Although I can’t really speak to Brett’s state of mind, I certainly didn’t find your blanket assumption implicit in his post, either. Finally, for you to say “For instance, you might not realize that using multiple paths … actually increases capacity” means that either you don’t know what the acronym MIMO means, or again, you didn’t carefully read what I wrote…

  41. Jessica Boxe>The best use of our radio spectrum is to sell it off to the highest bidder, give them full property rights based on appropriate criteria of power and bandwidth, including limitations of trespass and nuisance, and then use the money we save to pay down some of the unfunded obligations we have to users of social security and medicare.

    How about:

    The best use of our water is to sell it off to the highest bidder, give them full property rights based on appropriate criteria of power and bandwidth, including limitations of trespass and nuisance, and then use the money we save to pay down some of the unfunded obligations we have to users of social security and medicare.

    or:

    The best use of our oxygen is to sell it off to the highest bidder, give them full property rights based on appropriate criteria of power and bandwidth, including limitations of trespass and nuisance, and then use the money we save to pay down some of the unfunded obligations we have to users of social security and medicare.

    Here’s a better idea. Let everybody use what they want (I can live with restrictions on VERY SPECIFIC frequencies for emergency services and military us) and let the market decide how best to use it. Anything that’s not specifically reserved for the two above list items, operates under Part 15 rules. EVERYTHING. This would create a significant market for new technologies and resolve MANY of the problems that currently exist in the wireless world. It would also create billions (trillions?) of dollars in income and put tens of thousands of people to work.

    As for unfunded liabilities, don’t give the weasels in DC a reason to NOT end them. If they think they have a silver bullet, it will only encourage them (as it did when Reagan allowed them to start spending Social Security money as part of the general fund). That crap has GOT to end and everybody has to know there is no magic hat to pull a $125T out of. It’ll be painful as hell, but it isn’t going to get less painful by putting it off.

  42. Patrick> …means that either you don’t know what the acronym MIMO means….

    In Morgan’s defense, the term MIMO has been so basterdized by the marketing droids that NOBODY knows what it means anymore :^). These days, anything with diverse antennas seems to qualify for the label “MIMO”. And beam forming is great technology, but it’s so encumbered by patents that it effectively doesn’t exist for anything except the most elite/highest price consumers (e.g. the DOD). In 2 years of searching, I only found 1 Korean manufacturer that made the beam forming equipment for the WiFi spectrum, and they didn’t sell in the US because of patent problems. There were a few in the GSM frequencies, but HOLY CRAP!!! the prices were incredible (even for carrier technology).

  43. @dgreer:

    Sure, I agree that beamforming technology isn’t available for the average Joe now, but in my original post I explicitly mentioned “beamforming” as well as “MIMO” (btw, simple diversity surely does qualify as MIMO even if it’s not as exciting or technically challenging as beamforming), so you’re really just helping to make my point that Morgan’s accusation that I was assuming things would stay exactly the same was incorrect.

    BTW, I would say I like the analogies in your response to Jessica about other potential “natural monopolies”, but I’m deathly afraid there might be politicians paying attention and taking notes. ;-)

  44. > What neither of you know about wireless would fill volumes.

    AM = Amazing magic.
    FM = Freaking magic

    Nobody knows wireless. It’s magic! Clarke’s theorem applies.

    Yours,
    Tom

  45. # Patrick Maupin Says:
    > if it were unimportant then a company which owned
    > a small sliver of the spectrum in a particular locale
    > would currently be able to deliver unlimited bandwidth,

    Thank you Mr. Strawman.

    > the best example of spectrum usage these days is the sort
    > of free-for-all that happens in the WiFi space.

    Curious that you should say that since it is a point I addressed directly in my original comment. The so called junk bands that WiFi occupies are successful for precisely the causes I advocate. These bands are absolutely regulated. You control the use of that sliver of spectrum in your home or place of business, because you control the transmission equipment in that location (notwithstanding the aforementioned issues of tresspass, which in RF terms is called interference.) So I agree. I advocate the privatization of all of the radio spectrum, and were we to do so, I imagine the benefits you mention would come about. But not because it is a commons, no rather, because it is not a commons.

    dgreer Says:
    > The best use of our water is to sell it off to the highest
    > bidder,

    Curious. You are aware that water is a commodity that is generally privately owned, right? You are aware that in cases where large bureaucracies control water, such as in Southern California, it is grossly over priced and badly misallocated?

    dgreer Says:
    > The best use of our oxygen is to sell it off to the highest bidder,

    Curious again. You are aware that oxygen, when captured, is privately owned? Goods that move freely between real property are not generally ownable, because owning implies the ability to control. If you could build something on your property to contain all the oxygen above it you would be entitled to do so. But you are just being silly, right?

  46. dgreer Says:
    > The best use of our water is to sell it off to the highest
    > bidder,

    My previous argument on this is not good, so I apologize, and will give it another shot.

    My argument in regards to radio spectrum is utilitarian, not philosophical, and as you know, utilitarian arguments are always contextual. So, just because I think radio spectrum should be managed by the means of private property does not necessarily mean I think anything else should.

    However, if you want to discuss water rights and oxygen rights, I would be happy to do so. The fact is that water has been mismanged by governments in the most egregious ways. I have no doubt, from a purely utilitarian point of view, water would be much better allocated by a property rights framework.

    Oxygen is not considered a limited resource and so it is not managed or regulated by either governments or private enterprise via property rights. It is not an economic good. Consequently, your argument is a straw man.

    I should also point out that not all property rights are the same as real estate property rights. Sometimes people own a share of something. This is especially the case for property that moves through real property rather than simply residing on it. For example, property rights in rivers and lakes does not allow you to suck up all the water in that river or lake for your own use. Rather it allows you a share of the water, and a limited right to pollute the water. It is not ownership of a particular bunch of water molecules.

    If your argument is that commodities vital to life cannot be safely owned, then you are clearly wrong. Food is a quantity vital to life and is abundantly provided by the free market. For sure government regulation messes things up there too, basically via corporate welfare. However, there is little doubt who owns the can of peas in my pantry, or the mango in my fruit bowl.

  47. dgreer Says:
    > As for unfunded liabilities, don’t give the weasels in DC a reason to NOT end them.

    Just to clarify, if you haven’t read what I have written on this before, I am not advocating using the income from selling resources, like spectrum, for ongoing expenses. My argument rather is that the debt owed to people who have paid into the system be converted to a real debt instrument. And these debt instruments be paid off in full. There are various mechanisms for doing this, but one way would simply be to set up a bunch of large corporations who were randomly given a bunch of government assets, and then shares in these companies distributed to people who were owed social security debt. These companies would be floated on the public markets and entirely private corporations owned by the social security debt owners. In exchange, the social security debt is wiped out, and all these government assets are put back into the private sector where they can be used much more efficiently.

    There are other ways to achieve the same end, but bottom line is that it will never happen. Debt default, collapsing economies, riots in the street, devauling the currency, these are all far more appealing ideas from the point of view of politics. And it is in the political economy, not the business and financial economy, that these decisions are made.

  48. @JF Sebastian:

    What neither of you know about wireless would fill volumes.

    Absolutely! (OTOH, there are probably fewer than ten thousand people on the planet that you couldn’t legitimately apply that statement to.)

    I never claimed to be other than an uneducated schmuck when it comes to wireless technologies. But, your righteous indignation over Brett “misusing” a term (when you knew exactly what he was talking about but was sure he didn’t) and then your later argument by dismissive insult show that you are absolutely unqualified to teach on this subject — even if you happen to know the material (as yet unproven), you don’t have the right temperament.

    @Jessica Boxer:

    There was a point made and seconded that the number of bits per second available for communication, e.g. to a small handheld device, was dependent on the amount of spectrum available.

    First you said that argument was true but unimportant, then you said it was a strawman. Which is it? (Frankly, in this context, I don’t believe it is either, because if the argument were either false or unimportant, that would mean that there is already sufficient bandwidth in the unlicensed bands for all uses, and then all your arguments about how we should reallocate additional spectrum fail.)

    I have to say, though that I find a part of your argument very confusing. In your second post you say “The so called junk bands that WiFi occupies are successful for precisely the causes I advocate” yet in your first post you say “The best use of our radio spectrum is to sell it off to the highest bidder”. I’m not very imaginative and have a bad memory, so personally I have a hard time squaring those two statements — did a huge consortium of equipment manufacturers buy the wifi band while I wasn’t looking?

  49. Merciful $DEITY.

    The kind of free-for-all advocated over the last several messages would destroy the utility of the RF spectrum, period. Standards go out the window – and so do those shiny new HDTV’s you’ve got in your living room, and every radio you own. Emergency communications becomes impossible. Things like ham radio, where the users can’t even dream of competing in multibillion-dollar spectrum auctions, disappear in a single bang of the auctioneer’s gavel.

    What people don’t realize is that we had exactly the kind of radio free-for-all they’re advocating in the early days. The loudest signals won – whether or not they were drowning out the weak SOS form the ship that was sinking fast…

    We have radio regulations today because we discovered we needed them, the hard way. We may not need them in exactly their current form now, but we need something much closer to them than what the anarchist libertarians are advocating. The FCC and the ITU are about far, far more than just handing out broadcast licenses.

  50. @Jay:

    I don’t think you really saw anybody argue for complete anarchy — Jessica said thing like “limitations of trespass and nuisance” and even when I was holding up WiFi as a shining example of useful innovation, I suggested “with higher power limits and perhaps somewhat more stringent rules of the road” for additional bands.

    Jessica and I apparently disagree over the best management practice. In these range wars, I’m closer to the side of the cattle ranchers, and Jessica is apparently on the side of the sheep herders. Unlike the wild west though, I think we now have the technology to actually verify that people aren’t misusing the commons, and I think it is foolish and short-sighted of us to sell off spectrum to organizations which can extract perpetual rents, in the same way that I think it is foolish and short-sighted to extend copyright terms whenever Mickey Mouse gets close to the public domain.

  51. Jay Maynard Says:
    > The kind of free-for-all would destroy
    > the utility of the RF spectrum, period.
    > Standards go out the window

    I was wondering if you had heard of a technology called WiFi?

    > Things like ham radio, where the users can’t even dream
    > of competing in multibillion-dollar spectrum auctions,

    And that is bad exactly why?

    > What people don’t realize is that we had exactly the
    > kind of radio free-for-all they’re advocating in the early
    > days. The loudest signals won – whether or not they were
    > drowning out the weak SOS form the ship that was sinking fast…

    Not only do you misunderstand the history of radio regulation, you also misunderstand what I am advocating. 911 services work perfectly well on privately owned phone systems. And they would continue to work without the regulation involved. The government and maritime distress systems can buy some spectrum for such purposes, just as they buy fire trucks and boats from commercial companies. No doubt they should charge some sort of rent to the commercial enterprises that use them, but that is a separate matter.

    I find the term free-for-all curious. The consumer is a very powerful regulator, one need only look at our government if you want to see free-for-all writ large.

  52. # Patrick Maupin Says:
    > I think it is foolish and short-sighted of us
    > to sell off spectrum to organizations which
    > can extract perpetual rents,

    But your OK with the organization called “the FCC” to extract perpetual rents? Why? Because the FCC has shown itself so capable of producing optimal use of the spectrum?

  53. > Things like ham radio, where the users can’t even dream
    > of competing in multibillion-dollar spectrum auctions,

    And that is bad exactly why?

    Because stifling ham radio will hinder the ability of the little guy to experiment and advance the state of the art, and because hams still play major roles in disaster communications when nothing else runs. Hams specialize in developing inexpensive, reliable communications systems that work under all sorts of adverse conditions, far more than governments or corporations can.

    There’s another question that needs to be asked: Which corporation are you willing to be beholden to? Under your regime, all of the spectrum would be owned by telcos, or Google, or Microsoft, or GE, or Disney. The investments required would be so massive as to drown out the little guy.

    The consumer is a powerful regulator only when there is meaningful choices for him to make. There would be none under your regime, just as there are none in getting broadband Internet access now.

  54. Jay Maynard Says:
    > Because stifling ham radio will hinder the
    > ability of the little guy to experiment and
    > advance the state of the art,

    All value is relative. Is this more valuable than other types of use of the radio spectrum? Are you seriously contending that ham radio operators make a more significant contribution to radio than, for example, Qualcomm?

    > There’s another question that needs to be asked:
    > Which corporation are you willing to be beholden to?

    And you’d rather be beholden to a largely unrepresentative, unresponsive government agency? If you have a problem with something going on in the world of radio, is your complaint heard in any significant way by the FCC? Do you think voting in our bifurcated democracy offers any significant representative control over the FCC?

    > Under your regime, all of the spectrum would be owned
    > by telcos, or Google, or Microsoft, or GE, or Disney.

    You are aware of the brutal level of competition between cell phone providers right? You are aware of how responsive many of them are to customers out of a desire to retain that customer? You are aware of the radically improving price performance ratio of cell phone bands? You are aware of the brutal competition in the FM and AM radio market? You are aware that it is free to the consumer? And that is companies that control tiny amounts of the spectrum.

    > The investments required would be so massive as to drown out the little guy.

    You think so? How much do you think a few MHz at a few hundred watts would cost for a few hundred square miles in the middle of rural Kansas? I think in my scheme the little guy would have a much better chance.

    > The consumer is a powerful regulator only when
    > there is meaningful choices for him to make. There
    > would be none under your regime, just as there are
    > none in getting broadband Internet access now.

    Exactly, and selling off the whole of the spectrum (as opposed to tiny slivers as we do now) would provide a much more competitive environment. Curiously, one of the problems it would solve would be greatly broadening broadband Internet access.

  55. @Jessica Boxer:

    Instead of addressing my question about how you square your two assertions that “The so called junk bands that WiFi occupies are successful for precisely the causes I advocate” and “The best use of our radio spectrum is to sell it off to the highest bidder”, you equate my statement that “it is foolish and short-sighted of us to sell off spectrum to organizations which can extract perpetual rents” with a wish on my part for the FCC to extract perpetual rents, when it was clearly framed as an argument to open up more spectrum to rent-free uses.

    To be clear, with the WiFi example we both love so much, there is very little, if any, rent-seeking going on. Most certifications are done by third parties, and there is no significant regulatory cost associated with building, selling, or using most Part 15 equipment. WiFi equipment truly operates in a commons, and IMO the FCC has actually done a reasonable job of enforcing the boundary of that particular commons.

  56. Patrick Maupin Says:
    > Instead of addressing my question…

    I’m sorry if I did not explain my pov clearly, my fault. Let me try again: fundamentally, WiFi does not operate in a commons. WiFi is a fair example of property rights as applied to spectrum. Although the FCC says “do what you want as long as you don’t exceed this power level”, what we have is something different in an important way. “power level” roughly corresponds to “radius”, so what this roughly corresponds to is “do what you want as long as you keep your signals within a given radius.” If I control the transmission equipment within a physical space, then I control the WiFi within that space, up to the radius outside my property line.

    Obviously this is very approximate. In my apartment there are about four or five WiFi transmitters in range, but it is a decent approximation to proprietary control of an important part of the spectrum.

    If investors can actually own a space then they can invest more into it, and make it more useful (for example by fracturing it into cells and connecting the cells with point to point links.) Since there is so much of it, and much of it has pretty similar characteristics, the most likely result would be comoditization, which is good for everyone, except the real rent seekers. (And here I use “rent seeker” in the strictest sense of using government enforced rules to extract rent that would not be possible in a free market.)

    I don’t object to some parts being set aside as a public commons, in much the way that we have with WiFi (public space that can be privatized, that is to say,) public spaces are good and useful, even if they don’t provide maximum utility. But the majority of space needs to be in private hands to maximize its utility.

    I should also point out that if we collect the “rent” with a one time payment during such a sale, and use it to eliminate large swaths of our debt or unfunded obligations, we will be paying ourselves out of that annuity for a long time to come.

  57. Jay: I’m pretty well acquainted with the history of which you speak, which is why I specifically mentioned Part 15. FCC has a role to play, I just wished it was MUCH smaller and less intrusive into the parts made by the main characters in this play (which are the ones creating, using, and adapting the technologies we actually need).

    Jessica: Yes, I’m fully aware of water rights… own them myself, thank you very much. My comparison to water and oxygen was that it is a fluid thing, not “real property”. You cannot hold spectrum in your hands. Even if you build a radio system that transmits on frequency X, it sends harmonics on a whole BUNCH of other frequencies, inducing interference proportional to the power level of your transmitter and design of you antenna. The part 15 rules are designed to minimize this cross-spectrum interference without being too draconian. The innovation that has come as a result of Part 15 (which includes virtually every electronic gadget you own, not just wifi, and allowed the development of the commercial microwave oven) came BECAUSE of the release of regulation in those frequencies.

    I’m not sure what you mean by “junk frequencies” as the ISM frequencies (900MHz, 2.4GHz, and 5.7 GHz among others) have both pros and cons as do any other frequency in the spectra. If a lot more where made available under Part 15 rules, there’d be a WHOLE LOT of new toys out there to play with within a couple of years, particularly if they are in NLOS ranges (e.g. the old analog VHF and UHF television channels).

    And I’m not sure why you think HAMs don’t contribute as much as, say GE or Motorola. Most truely gifted radio engineers, even today, start out as HAMs, just as most truely gifted programmers start out as Hackers. Ham’s have contributed more than their fair share to the use of radio technology. Even our beloved WiFi is based Packet Radio which was developed by (drum roll) HAMs.

    Patrick Maupin> btw, simple diversity surely does qualify as MIMO even if it’s not as exciting or technically challenging as beamforming

    Uh, no, it doesn’t. Multiple In Multiple Out means more than one transmitter and more than one receiver. Diverse antennae are neither transmitters nor receivers, they are antennae (your rabbit ears don’t make your TV MIMO). The marketing droids strike again. ;^) Beam forming requires a bit of math to accomplish what is basically interferometry, which makes the whole thing quite a bit more complex.

  58. Jessica Boxer: Your arguments make an assumption that the purchasers of frequencies will maximize use of that spectrum, but there is AFAIK nothing that argues for that idea, and in fact in practice quite the opposite that is true. Companies buy up frequencies and sit on them for decades without ever putting anything into them. The cell phone companies are famous for this sort of thing. The reason that there are so few mom-and-pop cell phone companies is because it’s damned near impossible to get most GSM and CDMA cellphone frequencies in any heavily populated areas because they have been grabbed up by the top 5-6 providers (even when the providers don’t service that area).

    If these spaces were Part 15, then it would be far more reasonable. Hell, even a “use it or lose it” clause would make it more reasonable. But, if you’re going to take the spectrum (which BTW belongs to ME TOO) and sell it to somebody to squat on, I’m not going to back that.

    I’m fairly surprised by you’re POV on this, seeing as you’ve typically been more on the side of the free market and what you are proposing looks much more like the existing regulatory regime, only more so.

    What if the FCC starts selling off frequencies in the nanometer wavelength range? What will you do when they sell 700-635nm to AT&T? Will you simply stop wearing red? That may seam like a ridiculous argument (and intentionally so), but it points out the ridiculousness of much of this argument. The FCC’s job should be to set rules that avoid interference in common use, and prevent interference (there’s a difference) in certain very specific cases (public safety, ATC radar, military, maybe a few others). I’d rather my TV’s signal be in Part 15 space than the current wasteful regime.

  59. @Jessica:

    Actually, I think that WiFi does operate in a commons. “Commons” does not imply no property rights — in point of fact, it usually implies quite complicated property rights. Despite the “tragedy of the commons” meme, historically, there have been many examples of good management of common resources, both cooperatively and through regulation, and the FCC’s limitation on transmit power lies solidly within the tradition. BTW, it is interesting that you try to show a connection between WiFi transmit power and property lines — the FCC has forcefully rejected this connection and won in court on multiple occasions.

    But whether the WiFi band is a commons or not does not really alter the basic argument. If there were nothing like the WiFi band available — if that spectrum had been sold to the highest bidder for an outrageous sum, I don’t think you’d be able to buy a USB WiFi adapter for ten dollars at Fry’s and just start using it. I don’t know how you define maximizing utility, but I find that pretty utilitarian. Absolute ownership of a space is not always required to spur investment. A huge investment in WiFi was made because many competing companies felt relatively secure that the spectrum would not be capriciously reallocated. (In other news, this rapidly became a self-fulfilling prophecy.)

    I argue that the FCC should work with industry to try to innovate more WiFi-like outcomes. You argue that we should sell something we all own to someone who will make their money back with profit by collecting rents from us. Sometimes this works, but as numerous examples of botched privatization schemes show, sometimes it doesn’t, and sometimes the average consumer is left much worse off after basic services are privatized.

  60. @dgreer:

    Ahh, I guess it was the marketing droids — my first introduction to diversity was with a system that actually had multiple receivers, not just multiple antennas.

  61. # dgreer Says:
    > Your arguments make an assumption that the
    > purchasers of frequencies will maximize use
    > of that spectrum,

    I might have used a turn of phrase similar to that, if so, I didn’t write precisely enough. It would be more accurate to say that they will maximize the value of the spectrum. There is a lot of capital investment involved and consequently they need to get a return on that investment. Part of the reason why cell phone carriers can sit on unused spectrum is simply because they have serious restrictions imposed on them by the FCC as to what they can do with it. If it can only be used for cell phone signals, then it might very well be the best use to them to sit on it. However, if they can use it for whatever they want the market insists on getting appropriate value out of it. Even large fat cat corporations don’t leave $10 billion investments fallow.

    > You’ve typically been more on the side of the free
    > market

    Property rights are at the very core of free markets. Without them, property is just on loan from the government. The problem with the spectrum is the tragedy of the commons. The solution we have is government regulation of the commons. My solution is privatization of the commons. This is pretty much the two standard approaches to the commons, state oriented or libertarian.

    This whole argument comes down to the same philosophical positions that all political arguments do. One position is that big corporations are evil money grabbing and vile, and that governments are looking out for the good of the people. The second position is that big corporations are self interested, but that their self interests largely correspond with that of their customers through the free market, and that governments are both incompetent and self interested also, and that their interests do not line up well with the interests of their “customers”. Voting is very different than writing a check.

    I was reminded of this last week when I had the experience of calling both my cell phone company and visiting the department of motor vehicles. Would anybody like to guess which interaction was more pleasant?

    > What if the FCC starts selling off frequencies
    > in the nanometer wavelength range?

    Such bandwidths are not considered scarce, so they are not economic goods.

  62. @Russell:

    A good starting point for learning about early radio “wars” is the “Wireless Ship Act of 1910”

  63. Jessica Boxer> Even large fat cat corporations don’t leave $10 billion investments fallow.

    Of course they do! Have you never heard of investment? They buy a chunk of juicy spectrum and sit on it for 10 years, knowing that there will be increased demand and they can sell or trade it for a profit. Again, the real property argument doesn’t hold here because radio spectrum IS a commons. You can’t broadcast signals inside you property line without sending signals outside you property line.

    If you want to abolish the FCC (which I’m willing to do), then another, private mechanism will spring up to do essentially the same thing as enforcement of Part 15 and coordination of spectrum use. The coordination thing already happens in heavily congested wireless areas like the SF Bay area.

    >Voting is very different than writing a check.

    Depends on the size of the check ;^).

    Might I suggest that there are more than just the two extremes you mention? Might I also suggest that you’re insistence on the SALE of spectrum is a usurpation of my rights? If bandwidth is property, then I claim all the bandwidth in all the frequencies which cross my property lines. I want rent from every radio station, satellite, etc. that I did not explicitly give rights to.

    Oh wait, that system’s unworkable isn’t it? Then perhaps bandwidth should be treated as something other than real property. It doesn’t ACT like real property. You can’t put a fence around it and say this is mine, that’s yours, get off my property. Perhaps it requires a bit more thought than just chopping it up into arbitrarily sized blocks and selling to the highest bidder?

    As a resource of the People of the United States, perhaps it’s not unreasonable that we want to see more done with it, and see more of it opened to innovation that generates trillions of dollars in business, rather than a few billion in sales. As I said, if you support the current chop it and sell it scheme, then at least add a use or lose it clause to the sales contract. If Joe’s Cell Phones for Less wants to open up Kansas City, he should be able to demand the right to buy or lease squandered resources (at a fair market value) from the squatter and start his business. This would encourage better use of the resources. As I said, a better use is either Part 15, or something similar to the proposals made for WISP licensing in the 3.5 GHz spectrum (where transmission sites are licensed to specific frequencies and powers and new appropriate seperation in spectrum and distance must be maintained).

    >Such bandwidths are not considered scarce, so they are not economic goods.

    They are no more or less scarce than any other frequency. The only difference is the information carrying capacity of the wavelength which is a function of it’s frequency and band width.

  64. Patrick: If you ever get a chance to play with them, take a look at Motorola’s PTP-400,500,600 series equipment. These are the old Orthogon radios (and Moto’s new versions) and they are REALLY COOL. My longest shot with one of these was 26 miles maintaining >70 Mbps throughput (35 up and down) on a pair of 600s. The newer 500’s are supposed to be a bit better. They use 2 radios and two antennas with the assumption that you are using H/V polarization between the two radios (and it’s specified which is which… if you mix them up there’s a potential for problems as I found out :^). Unbelievable machines. Dead simple interface, basically just a bridge, and REALLY solid.

    To make it even cooler, there’s a tool you can download from Moto that you put in the Long/Lat of the two points you want to link, put in the radio type and antenna characteristics, and it tells you what you can expect to get for link status and speed – it’s really accurate: if you aren’t getting what you expect, you installation is wrong or you missed something in you LOS path. I’ve never heard of any other reason for this.

    Because of the dual polarity, these things can achieve really good NLOS connections as well. I generally hated dealing with Canopy equipment, but I gotta tell you, these Orthagon radios were just incredible.

  65. # dgreer Says:
    > Of course they do! Have you never heard
    > of investment?

    So what precisely is your contention? That the fat cats won’t utilize the spectrum they bought quickly enough? Because you feel that government regulatory agencies are particularly quick at getting things done? Let me ask you this: if you were the CEO of MegaBucks Corp., and had spent $1 billion on a big chunk of spectrum. You are not only MegaBucks, but secretly also EvilBucks, and it is your intention to sit on the spectrum for a while thinking you will make it more valuable. Mwaaahhhh. Hey, why not use it in the meantime? No one else can use it, and the time it is fallow isn’t going to come back to you. So here is the new evil plan: keep it off the market to drive up its value, and get rent out of it in the meantime. After all, a productive resource is a more valuable resource. Yes, that is it, that is how to screw money out of everyone so I can eventually take over the world. Has anyone seen my white cat?

    > You can’t broadcast signals inside you property
    > line without sending signals outside you property line.

    You can’t water a garden without some of it leaking out into your neighbors. So what exactly is your point?

    > If you want to abolish the FCC

    Not necessarily. People have a right to enforce their property rights and something like the FCC might be a useful police mechanism for that purpose.

    > Depends on the size of the check ;^).

    I didn’t quite understand. Are you claiming that a very small check is less important than a vote, or are you claiming that your individual vote does have a significant impact on you personally? Voting is a vanity (though a necessary one), spending money is a directly self accountable action.

    > Might I also suggest that you’re insistence on
    > the SALE of spectrum is a usurpation of my
    > rights?

    This is a fairly complicated question, and I am not unsympathetic to your point of view. I’d love to debate the nature of property with you, but I am afraid that lack time requires me to take a pass. I think it would require me to write a great deal more than I am willing to do.

    > that generates trillions of dollars in business,
    > rather than a few billion in sales.

    Do you really think the value to society ends with the check received at the department of revenue?

    > a use or lose it clause to the sales contract.

    A use it or loose it clause means you don’t own the property. It depends on the specific details of the clause, and there is precedent for this, western homesteading required the homesteader to actually develop the land to make the property fully his own. However, it would not be my choice. Often the devil is in the details. What does “use it” mean? What does “lose it” mean? For sure in any sale of government assets it is possible to try to engineer the future with complicated conditionals in the sales contract. These are not a terrible idea, the problem is that the bureaucrats who would write such clauses are rarely very good at predicting the future, and it is an amazingly dangerous opportunity for pompous, political posturing.

    Nonetheless, as I said initially, the political climate would never allow such a transaction to take place, so arguing about the realpolitik of it seems like a hypothetical on a hypothetical raised to the tenth power.

    > They are no more or less scarce than any other frequency.
    > The only difference is the information carrying capacity of
    > the wavelength which is a function of it’s frequency and band width.

    Scarcity refers to contention for the use of the resources, not to its existence or its information carrying capacity. Property rights arise out of real conflict, not theoretical conflict.

  66. By the way, the title of that book is significant for a reason that bears on this discussion: Hams were first given permission to operate on wavelengths “200 meters and down” (shorter; 1.5 MHz and above in frequency) because it was thought that they were otherwise useless. Hams proved the doubters wrong. Hams have made innovations in every phase of the communications art. There’s a reason one early TCP/IP package for DOS was known as KA9Q: that’s Phil Karn’s ham radio callsign. You’ll find more hams in companies like Motorola and Intel and AMD and HP and IBM – the leaders of technology industry – than you will in any other industry. Kill ham radio, as you are proposing to do, and you will irreparably harm innovation.

  67. Jessica: You are wrong again here, and if you stop and take a breath, you’ll figure it out. You’re stretching your arguments to fit a set of circumstances that don’t exist. What you have posited is almost EXACTLY what the FCC does. It’s not working, as I and a number of others with experience in that realm have tried to tell you. Auctioning off nation-wide spectrum to the highest bidder is a BAD idea, has been a bad idea for decades, and won’t get any better just because if fits with you’re model of the universe.

    >Scarcity refers to contention for the use of the resources, not to its existence or its information carrying capacity. Property rights arise out of real conflict, not theoretical conflict.

    The “scarcity” is artificial, as you’ve been told repeatedly by several knowledgeable posters. It’s scarce EXACTLY BECAUSE the FCC has bundled it up and sold it to the highest bidders, who are largely letting it go unused because they don’t need it or it doesn’t fit with their current business plan.

    I might remind you of Eric’s post a few months back that the government creates monopolies. This is JUST such a situation. It’s not impossible for a small business man with a million bucks to build a cellular network, but these entrepreneurs have almost no chance of doing so because they cannot get spectrum. This is one reason I’m seriously rooting for the WiMAX folks because they have a chance to route around this damage and build something that actually competes with the big cell phone providers.

    And your supposition that spectrum buyers will use that unused spectrum to make extra money is bogus, because they already use other spectrum for which they are licensed and to use the new spectrum involves a whole lot of work on the front-end, not to mention a complete engineering of a new radio network. Even after you own spectrum, you have to do lots of work in locating you transmitters, setting power levels, optimizing the network, etc. to make a system useful, whether it be for VHF mobile/handheld networks or microwave voice/video/data. There’s a reason the guys who build this stuff are called radio “engineers”, and why to those outside the discipline call it “F’ing Magic”.

    Oh, and just in case somebody thinks I claim to be part of that very exclusive club, nope. I’m not a radio engineer. I know several and have worked with them, and I wouldn’t do that on a bet :^). And you HAM guys are almost as crazy as the REs :^).

  68. dgreer: Personal pet peeve: “ham” as in “ham radio” is not an acronym. Nobody’s quite sure of the etymology of the term, but it’s a word, and should be capitalized as one.

  69. Jay: Thanks for the correction. I’ll try to remember that (and if I forget, feel free to kick me about it again until I remember :^).

  70. dgreer Says:
    > You are wrong again here, and if you stop and
    > take a breath, you’ll figure it out.

    I assure you, my respiration is within normal parameters.

    > What you have posited is almost EXACTLY what the FCC does.

    Oh really, so that bandwidth that Verizon “owns” can be re-tasked to TV signals if they want?

    > The “scarcity” is artificial,

    The scarcity of bandwidth is no doubt increased by the present situation yes. However, different people want to use different parts of the spectrum for different uses and everyone can’t get what they want. That would be true under any conceivable scheme, and that is the very definition of scarcity. There is no contention over the color red, except when Melinda wears a the same color of red dress as Susie to the prom. That is certainly a real problem, and I think even libertarianism has no known solution to the problem of high school politics.

    > I might remind you of Eric’s post a few months back that the government creates monopolies

    From my memory, although Eric and I are usually on the same side of an issue, he does not agree with me here, but I won’t put words in his mouth. However, such a situation does not create a monopoly at all. Even today, with the corporate welfare driven system, we have dozens of suppliers of cell phone networks, dozens of radio stations, and hundreds of broadcast TV channels. There is a lot more than one seller in the market.

    > There’s a reason the guys who build this stuff are called radio “engineers”,

    Being patronizing and mocking is undoubtedly an effective debating tactic, I suppose I am prone to it myself. Nonetheless, your argument is basically that the “extra” spectrum would not be used because it is to expensive to build it out. What isn’t clear to me is that if DGreer Engineering LLC owned the same spectrum rather than BigBadEvil Corp. why DGreer would be able to make it economical when BigBadEvil couldn’t. But perhaps I am missing something obvious.

  71. # Jay Maynard Says:
    > Hams were first given permission to operate on
    > wavelengths “200 meters and down” (shorter; 1.5 MHz
    > and above in frequency)

    Sorry, Jay, I haven’t read the book you cited, though I am sure it would be an interesting read. However, based on what you say above, presumably Ham radio guys used to have a lot more spectrum to play with than they do now. So I guess, even in my scheme, if we left a little room for the public square, your guys would be happy?

    > Kill ham radio, as you are proposing to do, and you will irreparably harm innovation.

    I am not proposing that particularly, I am just saying that everyone has to compete in the market place to show the value of what they do. As I have said a couple of times, although an inefficient use, having some public square set aside in the spectrum might be a reasonable idea. But the vast majority needs to be controlled by private industry.

    When airplanes were first invented, there was huge innovation in the realm of the individual inventor. However, now that it is a mature industry we expect Boeing and Lockheed Martin to do the innovation, not Orville and Wilbur. I imagine the same is true today. Who in the Ham radio industry is producing anything approaching the innovation in the radio systems as we have seen used in cell phones, bluetooth or WiFi? These amazing, exponential improvements come from large corporations, not mad geniuses getting high on flux fumes in the basement.

  72. Jessica Boxer Says:
    > Eric and I are usually on the same side of an issue, he does not
    > agree with me here,

    Sorry, let me clarify this. I 100% agree with ESR that monopolies are created by governments. Where I think we disagree is over the idea of making the radio spectrum into a property rights issue.

  73. presumably Ham radio guys used to have a lot more spectrum to play with than they do now.

    Yes, once upon a time. As the spectrum became more and more useful, hams got to use less and less of it. The ham radio bands as they exist today were laid out in the aftermath of World War II, with only a few exceptions. Hams have pieces of spectrum from 1.8 MHz to 250 GHz, as well as everything above 275 GHz. Different frequency ranges have markedly different characteristics, and as long as hams can experiment in different parts of the spectrum, they can fulfill that part of their purpose.

    So I guess, even in my scheme, if we left a little room for the public square, your guys would be happy?

    One poster to the ham radio newsgroups called the ham bands the National Park of the MInd. I hate that title, but the concept has merit.

    But the vast majority needs to be controlled by private industry.

    WHY??!
    You have yet to do more than give impenetrable arguments straight out of academic libertarian theory.

    When airplanes were first invented, there was huge innovation in the realm of the individual inventor. However, now that it is a mature industry we expect Boeing and Lockheed Martin to do the innovation, not Orville and Wilbur. I imagine the same is true today.

    Then you are as woefully uneducated about the state of the art in aviation as you are the state of the art in radio. Boeing and LockMart haven’t innovated a damned thing in 50 years. They’ve taken innovations generated by guys working on aircraft in their garage and made them applicable to the kinds of airplanes they build. Walk around the EAA AirVenture show at Oshkosh the last week of July, and you’ll see far more real innovation in aviation than you ever will walking around the Skunk Works.

  74. Oh, I almost forgot:

    Who in the Ham radio industry is producing anything approaching the innovation in the radio systems as we have seen used in cell phones, bluetooth or WiFi? These amazing, exponential improvements come from large corporations, not mad geniuses getting high on flux fumes in the basement.

    Horse exhaust and hogwash. The very advances you cite were invented by hams. The Motorola engineer who is credited as inventing the cellphone explicitly cited ham radio as the source and inspiration for his idea. As dgreer noted, WiFi is nothing more than an enhancement of packet radio, which was developed by hams in Hawaii. Ever hear of ALOHAnet? Look it up sometime.

  75. Jessica Boxer> Being patronizing and mocking is undoubtedly an effective debating tactic, I suppose I am prone to it myself. Nonetheless, your argument is basically that the “extra” spectrum would not be used because it is to expensive to build it out. What isn’t clear to me is that if DGreer Engineering LLC owned the same spectrum rather than BigBadEvil Corp. why DGreer would be able to make it economical when BigBadEvil couldn’t. But perhaps I am missing something obvious.

    Perhaps I was a bit snippy there. My apologies.

    To clarify, when “BigBadEvil Corp.” (where did such a name come from anyway? I don’t think I proposed it.) buys up a nation-wide spectrum block, it locks out Dgreer Engineering LLC from using that spectrum IN ANY WAY. Whether they bought that block for future expansion, or simply to lock out competition, the result is that Dgreer Engineering can’t bring it’s super-wiz-bang-widget to market, employ hundreds or thousands of people to make, assemble, test, ship, sell and maintain the wiz-bang-widgets, and possibly create a completely new commercial sector of wiz-bang-widgets. The argument for the FCC “managing” and “selling” the spectrum is that it is maintaining the public interest in a common resource (wireless electromagnetic frequency ranges), but instead they’ve allow a potential high growth new sector of the market to wither on the vine because they cannot bring the product to market. Further more, if Joe over the cell phone shop wants to sell his cell phones, the mechanism to prevent him from entering the market has been graciously provided by the FCC. Joe, and the hundreds of others like him (and believe me when I say there are a BUNCH out there that would like to get into the cell phone carrier game and could afford it if the spectrum were in a Part 15 style box) would provide the competition that is necessary to drive cell service prices to the more appropriate commodity levels that they should be. This already happens with a few regional providers that were lucky enough to get in on some regional sales (or like Pocket was able to buy the stuff at auction after the .BOMB era), but it happens far to infrequently to make a difference with the national carriers, and because these are only regional licenses it does not allow these carriers to move into larger markets to compete for more lucrative networks.

    In short, the current oligopoly is maintained by the FCC “bandwidth auction” scheme and this must change to either a per-tower licensing scheme with assurances of non-interference with near-by towers, or by a Part-15 scheme where interference is highly unlikely due to restrictions on the gear (and actionable when people willfully violate the rules).

    >When airplanes were first invented, there was huge innovation in the realm of the individual inventor. However, now that it is a mature industry we expect Boeing and Lockheed Martin to do the innovation, not Orville and Wilbur.

    Uh, once again, you appear to be making assumptions based on a lack of experience. Virtually ALL innovation in civilian aircraft takes place in the Experimental category first and flies there 5+ years before going through the appropriate testing and certification to be included in production aircraft. If you go to an EAA regional fly in and look inside some the the new Experimentals, then look in some of the new production aircraft, you’ll see a HUGE difference in that production aircraft is WAY behind what a lot of amateurs are doing. The only exceptions to this general rule are the new S-LSAs which have a significantly lower barrier to entry into the market and much more liberal rules on the hardware they can use. I think that’s because the EAA helped write those rules to balance the desire of the FAA to control everything in site and the desire of individuals to have access to reasonably priced aircraft for sport aviation.

    And as I stated above, Wifi as a technology was built on top of packet radio, a ham technology. Hams do a LOT of experimenting that eventually gets picked up by industry (and is the primary training ground for most radio engineers). Hams are NOT CBers and you should know the difference before belittling them.

    These are really poor blanket statements, they are beneath you and you know enough to know better. Would you also have us believe that innovation in software only comes from Microsoft, Apple, Oracle, and IBM? Of course not, then why do you think other engineering areas are any different?

  76. @dgreer:

    That sounds like a lot of fun; unfortunately, I am not in a position to be playing with that sort of stuff right now. FWIW, one of the things I do is chips for POTS, so right now the highest frequency analog signals I get to play with at work are around 8 KHz, and in my job that’s considered “wideband” ;-)

    (That’s a slight oversimplification because of delta sigma modulation and out of band noise measurements, but I think you get the idea.)

  77. The only exceptions to this general rule are the new S-LSAs which have a significantly lower barrier to entry into the market and much more liberal rules on the hardware they can use.

    And every S-LSA on the market today was developed from, or as, a kit-built experimental.

  78. Jay: Actually there are two classes for most S-LSAs, Kits converted to production S-LSAs and European and Canadian “Ultra Light” planes (not FAR 103 compliant, but following European Ultra light rules which have similar specs to those developed for S-LSA). Although American manfacturers are finally starting to catch on with Piper and Cesna about to release their S-LSAs (or has Piper already put one out? I don’t remember).

  79. Piper has bought Czech Air Works, the SportCruiser manufacturer, lock, stock and barrel, and is selling their aircraft as the PiperSport. I don’t know if they’ve actually delivered any, but it’s probably not that far off.

    I don’t know much of anyone who thinks very much of the Cessna Skycatcher. The thing is heavy, leaving damned little useful load – little enough that a student and an instructor have to fly on much less than full tanks, and if they’re typical American males, have almost no room for fuel at all. Cessna has delivered production Skycatchers; the first went to the wife of the chairman of the company to learn to fly in.

    Even the European/Canadian microlights are pretty much kit-built experimentals built in factories, usually from the very same kits, and sold as S-LSAs. Note that a microlight typically has a 450 kg max gross weight, and so the aircraft take some reworking to be able to handle the US 600 kg max gross – and they all have been uprated, since an aircraft with a lower max gross is going to have problems competing.

    This doesn’t invalidate the point, though. None of this innovation comes from the big manufacturers. The Skycatcher is hardly innovative, and it’s the only S-LSA to be developed by anything resembling a large aircraft manufacturer.

  80. Let me address Jay’s comments, but for the most part the same principles apply to DGreer’s comments also.

    # Jay Maynard Says:
    > You have yet to do more than give
    > impenetrable arguments straight out of
    > academic libertarian theory.

    I do believe I have gone through this already, but since I am not being clear let me take one more whack at the pinata.

    Property rights are the core of the free market system, and consequently one of the roots of all our wealth, so evidence in favor of this would be: “Western civilization.”

    However on a more prosaic level, your co-participant in this debate provides a good example. He is concerned that MegaEvilCorp is going to buy up lots of spectrum and sit on it. I told him they would most likely realize the value of that asset by using it, and renting it out. DGreer tells me that building it out would be too expensive. I suggest that QED would be the appropriate acronym at that point.

    > Boeing and LockMart haven’t innovated a damned thing in 50 years.

    This comment reminds me of a thread on this blog a while ago about the putative vacuity of marketing. This comment reminds me of that same engineering attitude that I lamented back then. The hubris in the preceding statement is breathtaking. I think I failed to convince you back then, but I will simply reiterate again, there is a huge difference between coming up with a wild ass idea and making a commercial product.

    It reminds me of a meme in our society too. The idea of the mad genius coming up with an idea, getting a patent, and making billions of dollars. Didn’t they just make a moral outrage movie about that recently concerning the guy who invented the intermediate windshield wiper? This meme is simply wrong on so many levels.

    Like I say, it takes more than a wild ass idea and a prototype to make a commercial product. What in fact it takes is millions of innovations small and large, both in engineering, design, ergonomics, aesthetics, manufacturing, quality assurance, operational systems, marketing, sales, capitalization, and management. To think it is all about engineering is parochialism gone wild.

    The idea that Boeing built spectacular aircraft like the 747 or the 787 or the KC-767 without being innovative is simply not credible.

    And to bring it round to the original matter under discussion. All that massive innovation makes it possible for you to fly in comfort from New York to Tokyo, watch a movie or two on your iPhone, and call your spouse via the on board telephone (bouncing off communication equipment in space.) It is made possible by big bad corporations who provide the environment where all this innovation in thousands of different disciplines can be brought together, marshaled through the perilous process of product development, and brought to you, for a day’s wages with a smile and a thank you for your business.

    That is why we need big corporations, that is why they need property rights, and that is why we would all be better off without the true free-for-all that is our current spectrum allocation and management process.

    Nonetheless, I appreciate both of your thoughts on this subject.

  81. Jessica: I used to think you had two clues to rub together. Not so sure anymore.

    Jay>…straight out of academic libertarian theory.

    Uh, I don’t think so. I don’t remember ever reading that libertarians are blind commercialists. Jessica has taken all the Marxian Capitalist definitions and is trying to live up to them with her posts.

    Her bandying about obnoxious names like “EvilMega Corp” is kind of a pointed clue. She seems to think anybody who thinks small businesses and those unable to pay millions for worthless government approvals to do things that they should be able to do whenever they want as long as the don’t harm others are a bunch of tree hugging socialists.

    She seems to forget that the root of “libertarian” is “liberty” and is focused on INDIVIDUAL liberty, not huge business and regulatory capture of entire industries.

    Sad, she seems so bright otherwise. :^)

  82. dgreer: I happen to know quite a bit about the Zenith CH601/650 Zodiac, because I used to own one before the Great Dempression destroyed my finances. I’ve seen a 640 in the flesh, although I’ve never flown it; there was one at the AMD factory while I was there taking delivery of N55ZC. It seemed a nice enough airplane. It’s based on the CH2000 Alarus, a part 23 certificated aircraft that seems to be a decent trainer.

    Given all of the problems around the Zodiac, though, I have to wonder about Zenair’s long-term stability. If you do go with a 640, I’d recommend buying the entire kit at one time.

  83. Jessica, not everything in Western civilization is built on selling things. In particular, the radio spectrum does not lend itself to this approach for one simple reason: Radio waves do not respect international boundaries. Even if the US were to do away with the FCC and sell off spectrum to the highest bidder, there’s no assurance that any other country, let alone every other country, would follow suit, or even recognize what we have done. This is the reason the radio spectrum is divided into services, of which the broadcast and cellphone service are only two of many, and each service has its own chunks of spectrum. That’s intended to keep people from interfering with each other and to make sure that, for example, George Soros doesn’t buy up a chunk of spectrum for broadcasting in the US and wipe out police communications in Mexico.

    Boeing’s innovations are in engineering, and in particular ways of manufacturing. They’re always built on top of things that the guys in experimental aviation pioneered. The 787 wouldn’t be possible without Burt Rutan in his garage building the Vari-Eze. Rutan figured out how to do composites in aviation. Boeing simply figured out how to make them work in that one specific application. Their innovations are incremental. Rutan’s were and are fundamental. That’s the difference. As I mentioned, I used to own an S-LSA, designed and originally built in some guy’s garage before it was turned into a commercial product. You deride the meme; I live it.

  84. @Jessica:

    I told him they would most likely realize the value of that asset by using it, and renting it out.

    And you’ve been told multiple times that companies that already have spectrum often buy up other spectrum to sit on. Can you say “anti-competitive”?

    DGreer tells me that building it out would be too expensive. I suggest that QED would be the appropriate acronym at that point.

    Nope. Not at all. “Too expensive” as decided by a single large entity is often orders of magnitude smaller than an aggregate “too expensive” as decided by a bunch of smaller entities, especially when “this spectrum is already earning its keep by just being a barrier to market entry” is factored in. And as far as the rest of your comment goes, I would suggest that, if you stop viewing the world in black and white (e.g. LoneEntrepreneur vs. MegaEvilCorp, or whatever your words are), you might realize that small to medium sized businesses account for a hell of a lot of innovation these days. But, even if you insist on viewing the world in black and white, perhaps you should reflect that the market-cap leaders Apple and Microsoft are both around 35 years old, and were both founded on shoestrings. The same goes for market leaders in other industries besides computers, too, even highly capital intensive ones.

  85. It’s worth pointing out in this discussion that AT&T has now announced that no new customers will be allowed to sign up for unlimited data plans. No word on whether other carriers will follow suit, as yet.

  86. Sprint is still loudly touting their $99 plan, and they have added a $10 charge for the EVO and other 4G shared access point systems. Not bad, especially if you can drop you DSL link as part of the deal :^).

  87. Her bandying about obnoxious names like “EvilMega Corp” is kind of a pointed clue. She seems to think anybody who thinks small businesses and those unable to pay millions for worthless government approvals to do things that they should be able to do whenever they want as long as the don’t harm others are a bunch of tree hugging socialists.

    Yeah, I’ve noticed the same thing. :)

    @Jessica: Yes, there is a difference between a commercial product and a wild ass engineering idea. The thing is that you make it sound like there’s nothing in-between. Like it or not, the fact is that most innovation in this country does not happen at “EvilMega Corp,” but actually happens by small business entrepreneurs. Do you think Apple Computer, Microsoft, Ford Motor Co., Dell, Google, and HP were always big? Every single one of those companies — all of which have contributed innovations to their respective states-of-the-art — which I have named started in someone’s garage, basement or dorm room. And all of them were able to produce a commercially viable product despite being small. And they are the rule, not the exception. The next Apple, Google or Microsoft won’t be born of big company ideas, no, they’ll start small and start with the ideas of one or a handful of individuals.

  88. Morgan Greywolf> Yes, there is a difference between a commercial product and a wild ass engineering idea.

    I would even go so far as to say the about 90% of the stuff that is actually innovative started out as “a wild ass engineering idea,” even inside of “EvilMega Corp.”

    PARC
    Bell Labs
    Skunk Werks
    Google Labs

    Every one is a collection of really smart engineering and science types working on pet projects that eventually lead to things like fiber optics, DSL, Cable Modems, the GUI, the SR-71, and Android to name a VERY few. Wild ass engineering ideas have a tendency to change the world, like when a network administrator and hacker sees a need for a way for a bunch of scientists to easily create documents that link to other documents of previous work and an interface to easily follow those links. And sometimes they even get the recognition they deserve, like a knighthood :^).

  89. @dgreer: Good point. In addition, virtually everything associated with the modern personal computing experience — WIMP GUIs, CD-ROMs, Ethernet, PostScript, laser printers, WYSIWYG, even PDAs and smartphones — got its start at Xerox PARC, which was pretty much an environment where they gave the engineers a huge budget and told them to go nuts. And they did so with aplomb! Of course, Bell Labs is notable for its involvement in Unix, Plan 9 and Inferno. Those guys are now over at Google, writing things like the Go language. So yeah, even the “EvilMega Corps” realize that giving engineers a bunch of money to work “wild-ass engineering ideas” is a good thing.

  90. # dgreer Says:
    > Jessica: I used to think you had two clues to rub together. Not so sure anymore.

    Well, that’s nice. Like I say, mocking is undoubtedly a good debating tactic. But it doesn’t really make for an enjoyable conversation.

    > Her bandying about obnoxious names like “EvilMega Corp”

    I think you are missing the point. This is a pretty common technique in debate, namely accept the worst assumptions of your opponent and then demonstrate that despite these terrible characteristics, a beneficial outcome still arises. Anybody who either knows me, or has read more than three comments I have made in this blog or elsewhere, would know very well that I do not think corporations are evil, or that small businesses are bad. Neither would they think for a moment that I am in favor of the abuse of government power whereby wealthy corporations can buy unfair advantages from governments.

    However, you seem to be under the impression that an artificially created monopoly is the same as the protection of property (whether that property be real estate, movable, or one of the many other types of property.) They are very different in very important ways. However, I don’t think it is appropriate or useful for me to write a twenty page diatribe on the nature of property. There are many people much smarter than I who have done so and done so much better than I could.

    > She seems to forget that the root of “libertarian” is “liberty”

    If that is what you mean by liberty, then I plan to use my liberty to spend the money in your bank account. Please forward the number and password. Woo, hoo! New car for me!!

  91. # Jay Maynard Says:
    > Jessica, not everything in Western civilization is
    > built on selling things.

    I never made any such claim, in fact, I very carefully did not do so. I said that private property is one of the roots of our prosperity, not the only one. However, I invite you to remove one of the legs that holds up the Eiffel Tower, and see what happens. (Just to be clear, since I am feeling so misunderstood here, that suggestion is a rhetorical device, I am not actually advocating this.)

    > In particular, the radio spectrum does not lend itself
    > to this approach for one simple reason: Radio waves
    > do not respect international boundaries.

    I am not advocating selling the “radio spectrum”, I am advocating selling ownership of the right to transmit with a certain level of power at certain frequencies. Such a right does respect international boundaries, except insofar as the obvious fact that any use of property can lead to some pollution on the adjacent (and sometimes more distant) properties. The farmland in southern Texas bleeds its fertilizers into the Rio Grande, and thence into Nuevo Leon. Governments, and international treaties rightly regulate the degree of pollution permissible, and leave it up to property owners to decide how best to keep within these permissible limits. Mr. Soros had better do too that or he will be in big trouble.

    > Their innovations are incremental. Rutan’s were
    > and are fundamental. That’s the difference.

    Please explain to me the difference between incremental and fundamental innovation? Perhaps Newton’s Principia Mathematica might count as the latter, but I seem to remember something about standing on the shoulders of giants.

    Someone else cited Xerox PARC’s Alto. Undoubtedly this was an amazing piece of work, done under the auspices of a huge corporation. However, Alto was built on thirty years of work that preceeded it in bitmapped graphics, the mouse the menu and many other things. And, what is more, it sat in the lab unused until Steve came and set it free. So, it seems to me that all innovation is incremental. Though, to be honest, I have kind of forgotten why this is even part of the discussion. Oh, yes, the claim that Boeing hasn’t innovated “a damn thing” in decades. I’m afraid your examples don’t convince me that that statement is any more unconscionable.

  92. Morgan Greywolf Says:
    > Like it or not, the fact is that most innovation in
    > this country does not happen at “EvilMega Corp,”
    > but actually happens by small business entrepreneurs.

    I don’t know if that is true or not, certainly a lot of important work goes on in small businesses. Where exactly did I say any different? Do you agree with Maynard’s claim that Boeing hasn’t innovated “a damn thing” in decades?

  93. Jessica Boxer>This is a pretty common technique in debate, namely accept the worst assumptions of your opponent and then demonstrate that despite these terrible characteristics, a beneficial outcome still arises.

    Yes, it’s called a “straw man” and is a logical fallacy.

    >I am not advocating selling the “radio spectrum”, I am advocating selling ownership of the right to transmit with a certain level of power at certain frequencies.

    Sorry, I fail to see the difference. If you are unable to transmit in the spectrum, then it is mostly useless to you. That’s like saying you can buy a piece of property in the middle of my property, but you cannot have an easement to access it. This is such a common problem that it is illegal in most States.

    You seem to feel that we should simply sell off all commons to the highest bidder. My (repeated) argument is that this is not the optimal solution. You have been provided with ample evidence in this argument to support our side of the argument, and you’re only response is to accuse me and others of being snippy and present illogical arguments as to why I and others who disagree with you MUST give up our rights to (to use your snarky term) EvilMega Corp.

    Your argument assumes that I and others have no rights to our share of the spectrum, thereby you deny us OUR property rights and transfers those rights via the government to the highest bidder. Your solution is to steel from me to allow MegaEvil Corp to have a virtual monopoly via regulatory capture. On it’s face, your argument is theft, plain and simple. By your own reasoning, the protection of private property, your argument fails.

    THAT is what Jay, Morgan, and I have all been trying to tell you and present to you a more reasonable approach to REGULATION. The fact that you cannot accomplish this fairly simple logical progression is why I say you have no clue in this argument. No mocking necessary on my part, your arguments are self-mocking.

    I’m sure you’ll circle around to come at me from another direction because you don’t want to admit defeat, but when myself and two of the brightest folks on this site (I certainly don’t count myself in their class), two of us having a fair amount of APPLIED KNOWLEDGE in this SPECIFIC AREA have told you you’re wrong, and WHY you’re wrong, don’t you think it might behoove you to stop for just a moment and consider your arguments? I know that if something I say is contrary to people who speak with a great deal of experience, then I certainly stop arguing and ask questions to find out where my limited understanding differs from theirs and adjust my mental model accordingly.

    And before you go there, no, not everyone who is experienced knows what their talking about ( a lot of radio guys truly believe in FM and operate by intuition and dictate), but it’s not hard to find the ones who are kooks and weed them out. I don’t see where either Morgan or Jay appear to be kooks (I, of course, could easily fall into that category on so many levels :^).

    It’s for this very reason that I previously admonished you on your naivety and said I’d be very happy to discuss these matters with you in 10-20 years. By that time, you will have learned the lessons above (I would hope) and will be much more interesting to talk to. As long as you think that people who attack your ideas are attacking you personally, people will continue to see you as childish be hesitant to accept your arguments as well reasoned because they appear to be more statements of faith and you appear incapable of accepting criticism.

  94. # dgreer Says:
    > The fact that you cannot accomplish this
    > fairly simple logical progression is why
    > I say you have no clue in this argument.

    Wow! You are a angry person! I don’t really quite know what to say. You comment seems to be basically free of any content beyond invectives, appeal to authority, calling me a young whippersnapper, claiming I am unresponsive (despite having written thousands of words), naive, and telling me to grow up. I’m sorry, I don’t find any of these arguments particularly convincing, and I find discussing this matter with you more and more unpleasant. For the record, it is only you who has been rude and made personal attacks. Although I disagree with Morgan, Jay and others, they have all managed to make their case without resorting to such tactics. So please, don’t paint them with the same brush.

  95. For that matter, many of the same principles ESR propounds here apply to physical transport networks – consider the rise of public mail systems and the flat-rate postage stamp… though, strangely, it’s taken centuries for flat-rate pricing to move into package shipping.

    For that matter, the invention of the shipping container was really just the conversion of industrial shipping to a packet-switched system. It would be interesting to consider what conditions have to hold for these rules to apply to a physical transport system.

    1. >It would be interesting to consider what conditions have to hold for these rules to apply to a physical transport system.

      I don’t think the same mathematical model applies. Because you’re moving atoms around, and atoms have lots of mass, costs really do scale up with link capacity rather than the count of routing nodes. Packet switching, to the incomplete extent it is implemented by containerization, is an improvement, but it’s an improvement orthogonal to the basic capacity-cost function.

  96. Probably correct – certainly otherwise, as I alluded to, package shipping should have gone to a flat rate a century ago. Maybe it would still apply if/when the transport links have sufficient bandwidth that they never run at capacity?

    The thought occurred because I had read (some time back) of the the rapid success of fixed-rate postage in Britain.

    Wikipedia:

    The Great Post Office Reform of 1839 and 1840 was championed by Rowland Hill as a way to reverse the steady financial losses of the Post Office. Hill convinced Parliament to adopt the Uniform Fourpenny Post whereby a flat 4d per ½oz rate (equivalent to £1.18/kg for heavier items) was changed regardless of distance. The rate went into effect on 5 December 1839 but only lasted for 36 days.[2] This was immediately successful, and on 10 January 1840 the Uniform Penny Post started, charging only 1d for prepaid letters and 2d if the fee was collected from the recipient. Fixed rates meant that it was practical to avoid handling money to send a letter by using an “adhesive label”, and accordingly, on May 6, the Penny Black became the world’s first postage stamp in use.

    1. >Maybe it would still apply if/when the transport links have sufficient bandwidth that they never run at capacity?

      Alas, I doubt it. Remember the energy cost to move mass. I should have said that costs scale up with link utilization, not just maximum capacity.

  97. The Monster Says:

    May 29th, 2010 at 7:41 pm
    The set of late or deceased mathematicians is a subset of the set of ex-mathematicians

    I don’t believe I’ve ever seen Gottfried Leibniz described as an “ex-mathematician”. I’m pretty sure that folks like him are listed in the history books as “mathematicians” even after they died.

    I lol’d

  98. As the lead tech for an ISP, I’m perfectly willing to, for a fee of course, provide the technology to meter customer usage for any company who has problems with the cost to meter service. It’s fairly straightforward to set up and administer. Flat rate appeals to the consumer because it means bills are the same rate each month but there are ALWAYS abusers of flat rate services. ALWAYS. Metered service is the right answer, the real question is what the cost per unit of service should be. There is a real mess in how bits are charged for in the market, THAT’s what needs to be addressed. Get the price per bit down and nobody will complain about paying for metered service.
    I will argue that the cost to meter is so low as to fade to insignificance on a per customer basis (been there, done that, have the code to do it). The cost to provide service is much higher.
    As far as the people to run the network, I again take issue with the OP. You need 6 engineers to cover 24×7 shifts. Given proper network design and network monitoring design (something that is SORELY lacking in most networks), there is no reason that those 6 people can’t effectively manage a huge network. You do, of course, need additional staff to deal with customers but, really, you don’t need that many people to keep a core network running if you have proper design and operational procedures in place.

Leave a Reply to Jeff Read Cancel reply

Your email address will not be published. Required fields are marked *