interesting-people message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: [IP] More on A WSJ Letter to the Editor They Wouldn'tCare toPublish(which I mostly agree with)

  • From: Dave Farber <>
  • To: ip <>
  • Date: Wed, 08 Jan 2003 08:34:34 -1000

Title: approve:tippie &nbsp;More on A WSJ Letter to the Editor They Wouldn'tCare toPublish (which I mostly agree with)

------ Forwarded Message
From: Bob Frankston <>
Date: Tue, 07 Jan 2003 21:04:13 -0500
To:, "'ip'" <>
Cc:, Jeff Pulver <>, David S Isenberg <>, Daniel Berninger <>, David Reed <>
Subject: RE: [IP] More on A WSJ Letter to the Editor They Wouldn'tCare toPublish (which I mostly agree with)

[Dave -- I hate repeating myself but I guess I have to ... though each time I try to tune the remarks to the particular issue. This version is a little different since Voice over IP is now coming into its own. I realize that my metaphors might seem provocative but it's more a reflection of the extreme distortions that already exist than any exaggeration. Dan Gillmor's observations about the inconsistencies in the Telco whinings demonstrate this. It also got a bit longer than I planned but that is necessary since those who don't "get" connectivity need to have the concepts made explicit.]

The real tragedy is that the larger issues are lost in the sound and fury over artificial minutia. We have a simple problem -- one company owns the facilities that are used by it and its competition to build service. There is no such thing as real prices and costs in the absence of a marketplace. The FCC and Congress recognizes this and attempts to apply complex models in order to set prices and rules in lieu of a marketplace. If there were no alternative this would be a noble effort -- flawed but noble.

I think Joseph agrees with this but then says "Competition is best spurred by investment in new technology rather than in arbitrage schemes. Things would be simpler, and probably better for the industry and the public (if not for all the competitors) if interconnection standards were specified and enforced and all governmental restrictions preventing competition removed, including the UNE program."

I agree with the sentiment.

The problem is that this isn't viable as long as the "standards" are premised on a flawed model of telecommunications. We also don't need a new boondoggle with protected monopolies buying new fancy equipment when the existing facilities are used at less than 1% of capacity. There will be a need for new equipment once we give pent up demand an opportunity to assert itself.

We don't have to settle for just rearranging regulations when we can have a real marketplace. The key is to first understand what connectivity <> &nbsp;is. Rather than locking the providers into a narrow definition of a standard we should give them free reign well beyond what is possible even with faux deregulation. But this is only possible if their incentive is to provide connectivity rather than preserving their current highly profitable value-based offerings.

DSL provides a great example of problem. In an effort to assure competition we have defined standards in terms of an outdated implementation of signaling over copper. DSL can very quickly evolve to delivery many times the current capacity without distances limits at a lower price. In fact, why do we still use analog signaling over copper at all? And, though we can repurpose existing copper, why do I see new copper being installed rather than fiber?

What we need is a marketplace for connectivity.

We do not want nor need "broadband policy". It's just repeat of the ISDN debacle -- innovation was forced to build upon POTS because the incumbents controlled the definition of ISDN.

I already using Vonage <> &nbsp;and FWD <> &nbsp;and it works just fine without any special protocols (and without a telecommunications industry). More on this at <> . I want to emphasize that VoIP is not about any particular protocol nor standard. I can use the bit stream however I want. The use of such standards is optional and that's the real strength of connectivity and the end-to-end argument.

Aside from putting a lie to the existence of the telecommunications industry, VoIP is also a dramatic demonstration of why the carriers must not be allowed to define broadband. They complain about how there are no applications that really demand broadband and they keep trying to come up with just-so stories that seem to demand lots of bits. It's like trying to define the uses of PC's in 1975. Yet VoIP doesn't need much speed, just connectivity and we can have connectivity everywhere very quickly. We sort of do with modems and it is only a small step to stop second-guessing the CO Codecs and simply treat the two ends as a form of DSL even if it were not much faster than the modems (maybe 10x?).

The real problems with VoIP are the very interface standards Joseph is asking for. Any bridging to the PSTN requires meeting a complex set of regulatory burdens that are based on a 1930's models of telecommunications. It's like requiring all email carry a $.35 stamp (or whatever it is, I just have my computer print the stamps the few times I have to send paper mail - it will soon be $10 as the idea of sending email by paper becomes retro). An IP-based implementation would be far superior to the current E911. It could have a heartbeat to detect broken connections and automatically alert multiple observers and needs very little capacity -- just connectivity. We also have the problem of the e-rate which is threatened by VoIP even though its purpose is to fund such technologies.

Not only does all this sound and fury make it seems as if the the Regulatorium were dealing with reality, it creates more problems that need to be addressed while frustrating real competition. It's like the fireman who sets fires in order to do heroic rescues. The fires may be unintentional but they are also unnecessary.

We have an industry whose ROI is determined more by managing the regulatory process than technology and the technology itself is supposedly mysterious and unknowable. Today's Boston Globe had a nice column on Galileo <> . He may not have really said "'And yet it moves." but I identify with the feeling. Many of those most involved in policy do not have the technical expertise, or even believe they can have it, to evaluate the concept of connectivity. It is simply dismissed as a fantasy like perpetual motion or the idea that we Earth orbits the Sun. And that's the crux of the problem. The policy makers demand more policy rather than real reform.

The incumbent companies themselves actually recognize that connectivity is happening but their response is to try to find a new way to stay in a high value service marketplace. They should learn from Sears which, after years of trying to come to terms with a new marketplace, had to buy Lands End. Actually, we should learn from Sears and recognize that coddling them only delays in the inevitable.

So I have repeat my standard advice -- we must have a marketplace in connectivity and a remedy period in which the (incumbent?) service providers must have no ownership stake in the transport business.

What is different new is recognizing that VoIP is a demonstration of why broadband is so very dangerous. The incumbents are defining broadband as a high speed delivery system just like television. VoIP shows that low speed constant connectivity is where the real value is. It is easy to incrementally add speed as the market develops so there is no reason to even worry about it. At least, once the providers are incented to provide connectivity even if it means voice services are priced at a real market price of $0.00/minute above the cost of connectivity. At least that's what I pay for calls not bridged to the PSTN.

What we should wonder about is why my cable modem IP connection is less reliable than my voice or video services over the same wire and why my DSL service is slow because I am "too far" from the CO (though my two lines have different rated speeds!). If I have the IP I can do the video and audio myself. It doesn't work the other way and as long my connectivity provider has any stake in the video or audio content then we have a problem.

Arguing about the price of copper is an expensive diversion and foretells the more desperate efforts as we reach a critical point and VoIP simply becomes the norm. At least, that's my two cents {do I need to point out what US pennies are made from?}


Bob Frankston <>

From: [] On Behalf Of Dave Farber
Sent: Tuesday, January 07, 2003 17:58
To: ip

------ Forwarded Message
From: "Joseph H. Weber" <>
Date: Tue, 07 Jan 2003 17:49:49 -0500
Subject: Re: [IP] A WSJ Letter to the Editor They Wouldn't Care toPublish (which I mostly agree with)


Thre's always a lot more heat than light in the discussions of these issues. The costing standard - based on the theoreticl costs of building a new network, requires so many assumptions that the results are almost meaningless. It's plainly not an objective "fact" if those who benefit from low prices can always "prove" that the costs are low, while those who benefit from hight prices can always "prove" the opposite. All we can be sure of is that (1) the actual costs on the books of the telephone companies are much higher than the prices usually specified for unbndled network elements (UNEs) and (2) the most popular and controversial of the UNEs, the so-called UNE-P, is a combination of loop, switching and transmission that is functionally indistinguishable from resold service. Furthermore, the legislative standard for pricing of resold services in the 1996 Act (retail minus avoided cost) leads to much higher prices than is ordinarily charged for UNE-Ps. &nbsp;Thus the current arrangement has two markedly different prices for the same service, and they can't both be right. This is but one example of the kinds of problems this regulatory regime exhibits.

Everybody gives lip service to increased competition, but the devil is in the details here. If competition merely means that one company is riding on the facilities of another, with little opportunity for technolgical and service innovastion, is that really what we want? If competitors cannot economically supply services on their own to certain market segments, perhpas those market segments should not be artificially made "competitive." Nobody is arguing about the need for easy interconnection, which is of course necessary, but the requirement that one company allow another to use its facilities probably goes too far. I realize that this is in the 1996 Act, but it's well known that was a Christmas tree. I have always thought that the requirement to interconnect at any "technically feasible point" is a pernicious prescription that can and will lead to all kinds of problems, technical and economic.

Competition is best spurred by investment in new technology rather than in arbitrage schemes. Things would be simpler, and probably better for the indsutry and the public (if not for all the competitors) if interconnection standards were specified and enforced and all governmentqal restrictions preventing competition removed, including the UNE program. The result would be a lot less time spent arguing before commissions and courts, and a lot more rapid movement toward a real competitive environment.

Joe Weber
Strategic Policy Research
Bethesda, MD

------ End of Forwarded Message

------ End of Forwarded Message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help


Powered by eList eXpress LLC