Last Modified:                                                                                                



Homepage / Publications & Opinion / Interviews (post 2002)

Peter Cochrane Tracks Decline of BT and Rise of 802.11b in the UK
http://www.cookreport.com/
The COOK Report on Internet June - July 2002
Whither No License Wireless??
Exploring the Clash between an On Going Radio
Revolution, Regulators and the Local Loop
Volume XI, Nos. 3-4, June July 2002
ISSN 1071 - 6327

Spectrum Auctions Seen as Cause of Collapse of Telecom Industry in UK - Given Leadership Cochrane Points Out How Wireless Can Replace Local Loop

Editor's Note: Editor's Note: Peter Cochrane was Head of British Telecom Research from 1993 to 1999 when he was appointed Chief Technologist. In November 2000 he retired from BT to join his own start-up company - ConceptLabs - which he founded with a group out of Apple Computers in 1998 at Campbell CA, in Silicon Valley. A graduate of Trent Polytechnic and Essex University, he was the Collier Chair for The Public Understanding of Science & Technology at The University of Bristol from 1999 to 2000. He is a Fellow of the IEE, IEEE, Royal Academy of Engineering, and a Member of the New York Academy of Sciences. We interviewed Peter on February 28, 2002 and by email into the month of March.

COOK Report: It seems useful to explore how we may find our way amidst increasing chaos such as the passage in the House of the Tauzin-Dingell Legislation and the impulses of our bureaucracy to behave in more unpredictable ways.

For example, I heard from Dewayne Hendricks that on a trip to Washington at the end of February he had learned that the FCC was under pressure to reconsider the part 15 statutes on unlicensed spectrum. [See information on the Interdepartmental Radio Advisory Committee (IRAC) on page of this issue.] Perhaps this is an indication of the desperation of the local exchange carriers because, if any reasonable dent can be made in the local loop, it could well be by 802.11b wireless networks. Get rid of Part 15 and kill that possibility.

Operating in the Midst of Chaos
Cochrane: Well I can't really speak to what you Americans might do in a case of an attempt to overturn Part 15. But let me say that the great thing about the UK was that early on we were not allowed to have either a private radio or television stations or citizens band radio. The result of this prohibition was simple. We ended up with trawlers beyond the 12 mile limit with 100 ft. masts on them broadcasting to the mainland. People went to US and shipped back boxes of CB radio equipment. In short order the UK had illegal CB radio! But then in a fit of pique the UK licensing Authority said: "OK then you can have a CB radio, but we will simply change the modulation mode from AM to FM."

COOK Report: What happened?

Cochrane: Everyone simply junked their American AM equipment and went out and bought new FM CB radios.

COOK Report: This reminds me of the WTO. Spread spectrum radio certainly is manufactured outside of the USA. I would think that under the rules of the WTO it would be hard to ban the import of spread spectrum radios.

Cochrane: Not enough people know about the origins of spread spectrum. At the end of WWII the creation of a signal that could be received by a party with sufficient information and knowledge but that could not be heard by the enemy was an important objective. Why didn't you see applications after the war? The answer is simple. To generate the signalling and do the spreading you needed more equipment than would fit into my home.

COOK Report: In other words you could put it on a military ship but civilian use was not very practical.

Cochrane: One of the things for which I've been arguing in the UK and have totally lost out on because no one in the industry would stand up with me is a sane treatment of spectrum auction. The politicians involved simply aren't capable of getting it. They are totally conditioned to selling analogue TV and radio channels that are made by slicing up the spectrum into small amounts of bandwidth. Of course slicing frequency into channels is last thing you want to do with spread spectrum for something like a third-generation wireless. For that you should want the widest band possible. For example, something 100's of MHz wide. Instead what they finished up with for third-generation was, if I remember correctly, something like 60MHz. If they knew what they were doing, they would have had something like 300MHz.

COOK Report: Meanwhile, what is really the situation with BT? How old is the copper plant? Given its age does it really make sense to put much effort into DSL?

Cochrane: Before you make a judgment about DSL in the UK, I think you need to realize the precarious nature of the current situation. Just 18 months ago, all the telephone companies and all the cable companies were running a healthy profit. Today, by and large, they're all running at a huge loss and in a major difficulty.

COOK Report: What happened?

Cochrane: How about the government sucking $34Bn out of the phone companies?

The Spectrum Auctions Disaster
COOK Report: So you are talking about the 3G licenses?

Cochrane: Yes the industry has been raped of $34Bn and you can't take that amount of money out of the British telecom sector, or over $200Bn globally without seriously impacting the entire organizations. For example, the notion that the Mobile arm of BT is somehow disconnected from the rest of the company and that, if you rape that portion, it can be done without affecting the remainder is complete nonsense. After all BT provides a lot of the wire line and fibre-optic connection for the mobile operator side of Orange, Vodafone and all the other fixed and mobile operators. Moreover, it provides a switching capacity as well.

COOK Report: I thought BT was practically the only phone company. How many do you have in the UK?

Cochrane: We have about 150 licensed operators - with BT having about 60% of the market. While most of the other companies are regional operators, you must know that in the early nineties, the American cable companies came to the UK in order to do an experiment. This experiment was to build a cable TV and telephony infrastructure at the same time. They did it because they wanted to be able to go back to the US and say we did telephony in the UK, why can't we do it in the US? They invested about £10Bn in the UK. They (Telewest is one) make all their money out of telephony and make nothing from TV.

COOK Report: What you're saying then is that things were at a relatively even keel until this third-generation wireless auctions madness struck. That having been said how would you summarize the public and political understanding of what happened? How widespread is the knowledge of what it has done to the future financial viability of BT?

Cochrane: From the UK government there has been a total denial. At a public conference held on the issue I stood up and said "the industry has been raped of $34Bn." The response from Rothschild in defence of the government was: " oh, no, no, no. This was not a rape. It was an orgy." But it was the way the auction process was handled that led to the runaway bidding that got totally out of hand.

COOK Report: It sounds the like the people close to the action understand full well what happened, while the British government does not.

Cochrane: Within the EC there is recognition, at least in private that the politicians have goofed. The EC folk are desperately trying to find a face-saving way to enable them to put the money back into the industry. However, Mr. Brown the UK Chancellor of the Exchequer, has said quite emphatically that in the government will not put the money back the UK industry including BT. Since it is highly unlikely that the so dearly purchased 3G spectrum will be used for the purpose for which was bought, there has been some informal discussions of giving some of that spectrum to unlicensed 802.11b applications.

COOK Report: But presumably BT is not too keen on doing that?

Cochrane: I don't think that BT or the cable companies or any of the wire line industry is very interested in wireless. I find it fascinating that just a year ago spokesman from BT said that not only did they have no interest in 802.11b but that they also did not see it as a serious competitor with their business. In the meantime, consider what has happened. I have the sneaking suspicion that Mr. Blair, having made this statement and that this nation will be the most wired nation on the planet by the year 2004, has put his foot on a little pressure pedal to get BT Chairman Bland to rollout ADSL because compared to the rest of Europe the UK is as sick as a parrot.

COOK Report: Can ADSL be rolled out? Will the copper plant support it?

Cochrane: Let me put it this way. The UK will be the most wired nation on the planet by the Year 2004 because Tony Blair will stand up and say so. What the truth is at that point will have nothing to do with it. But politically the statement will be made. I suspect that political pressure is being put on the companies to get ADSL wide band out there in order to catch up with the French and in particular with the Germans.

COOK Report: When the politicians ask for an outmoded technology like ADSL, is it ignorance on their part or a cynicism?

Cochrane: I've yet to meet a politician who understands what word "wide-band" means. I have yet to meet a politician who uses email without having his secretary print it out first and hand it to him on paper. I think that ignorance is a big part of the problem. But I can grant them one small thing. They have figured out that it is important and that they better do something about it.

Assessing the Damage
COOK Report: BT is publicly held? How is its share price holding on? What do its profits look like?

Cochrane: It has a debt of £13.5Bn. In 2001 it made a profit of about £500M. It has managed to get its debt down to a £13.5Bn figure from about £18Bn by selling off assets. Three years ago it was making profits on the order of three or £4Bn UK£ a year.

Marconi in particular has also suffered through what I perceive to be as almost as big of a disaster as Enron. They have gone from having a £1Bn cash mountain to a debt of over £3Bn in 18 months. They were caught by the "dot.com" bust and the downturn in the market. And they had been buying at the top of the market and lost out very significantly.

COOK Report: But what was the effect of the 3G licensing on Marconi?

Cochrane: Think of it this way. When you take $34Bn out of the wireless side of UK mobile operators, what happens to budgets and capital expenditure? They immediately stop investing. Capital expenditure goes to zero. And immediately you are not buying any more fibre any more towers any more radios and so on and so forth.

COOK Report: Marconi made radios; transmitters and tower's didn't it?

Cochrane: That's right. You suddenly find that the whole sector goes into a tailspin. And I think that the economic impact of the 3G licensed debacle on the industry has slipped off the radar screen and has been responsible for triggering a major collapse of the telecoms market in the United Kingdom.

COOK Report: So our analogy to your 3G problem would be our fibre over capacity?

Cochrane: We also have a fibre over supply. In fact we have more fibre per capita than just about any nation on the face of the planet.

COOK Report: And given accommodation of the two factors in the UK, it sounds like you're saying is that the telecom downward spiral is bound to continue until we run headlong into massive reorganization - most likely by way of bankruptcies. In other words that people don't understand that the communications industry, virtually on a global basis is no longer fiscally viable. And furthermore that the current direction is likely to continue until enough people understand what is happening and the reasons why in order to get policy-makers to react intelligently.

Cochrane: Let me flesh out the scenario a bit. When the long lines guys looked at the growth of the Internet and at the growth of the "dot.com" boom and forecast the amount of capacity that would be required, they were also looking at mobile and 3G wireless and thinking that the demand could never slacken. Now let's look at why demand did slacken, and why the "dot.com" boom never took off.

Bankruptcies will sort it out
One of the major reasons was inadequate bandwidth to the home. You cannot have the "dot.com" economy without broadband to the home. The stalling of broadband installation has precipitated many things including the 'dot.com" bust. Of course, there were other factors involved. Greed being one. What we are likely to see now is a shake out where in each nation on the planet there are probably going to be four of everything. Four mobile operators. Four major ISPs, four local loop providers four inter city carriers and not 400 in each of these categories.

COOK Report: What can we hope for? Are we faced with another 2 years or more of bankruptcies?

Cochrane: I think bankruptcies will sort it all out. I don't think there is the understanding necessary for other kinds of change. What is even worse is that I don't think there is even the ability to understand. As a species we have an ability to cope with 3 or 4 dimensions or variables at a single time. What we do is take a problem that has 75 variables, pick the largest four, stick them into a formula and come up with a single answer.

This process works for things that are reasonably well behaved and more-or-less linear. When you get into a chaotic situation, you can pick one of the most insignificant variables in the set of 75, something otherwise insignificant like the 3G licenses, which can set off perturbations that send whole system spinning toward collapse. I believe that is what is happening here.

COOK Report: We have then a crash in progress. The question is where are we headed? Note the interest that Dave Hughes found in Wales during his visit from February 17 to the 23. He found wide numbers of groups ready to run their own community 802.11b networks if they can find a means of net connecting them. Now if the commercial companies are collapsing, isn't the general push of Internet technology to bring assets into the hands of the end users at the edge of the network? One outcome may be the growth of municipally owned and operated telecom at the local level. I imagine that the last big local Telco standing would like to keep such municipal competitors from becoming viable. How do you see this conflict unfolding? What can they do in Wales to get access to fibre?

Cochrane: Consider that most governments on the planet have spontaneously come up with charges for the use of "fresh air" in other words the radio spectrum. Consider that they choose to regulate use of that spectrum. I think that, with a suitable act of parliament, fibre that is not being used and has been purchased by a particular company could be commandeered and turned over to public use. If we have 96 fibre cables, only eight of which are being used, I think that there could be an argument for: "if you are not using it you are going to loose it."

Access to Fibre
COOK Report: So do you have some kind of national fibre authority like the Tennessee Valley Authority in the US during the depression?

Cochrane: I imagine that such a thing could be a solution.

COOK Report: In the meantime what can the people in Wales do? Start lobbying their representatives for an act that forces the availability of the fibre? How doable is something like this in the UK? Has anything like this happened in the last 25 to 50 years?

Cochrane: Let's see. The railways were private companies and they were put into public ownership. The coalmines were privately owned and they were taken into national ownership. The telephone company was nationally owned and made into a private company. Even AT&T was part of the American government and run by that government for a short period during WW II. So I would say that anything is possible.

COOK Report: And the reverse of some of that happened in the UK under Thatcher. But your point is that if enough people demand change, something can happen?

Cochrane: Yes.

COOK Report: Who controls the fibre in the UK? Is almost all of it owned by BT?

Cochrane: No. It is under the control of many companies. Some is owned by Level 3. Some by NT and Telewest, and about a dozen other cable companies. The power utility has fibre installed. Energis is a company with a large power utility-based network. Even the railroads have their own fibre. All of these folk have spare capacity because whenever you put in a fibre network, the cost of the fibre is insignificant compared to the cost of burying it. If you are going to dig the hole don't put two fibres in. Put 100 in. And team up with others to share the trench.

COOK Report: In Canada they have developed this to an art form where they talk in terms of doing almost exclusively shared condo style builds. Is this happening in the UK?

Cochrane: Yes. We have universities and town councils installing their own fibre into the ground for their own use.

COOK Report: It sounds like you are saying that if you understand what to do, you can do some good planning.

Cochrane: Let me give you a model. Supposing that in one reasonably concentrated geographic area, you had a thousand people willing to put money into getting better access. High-speed access from a satellite costs something like £30k/year. You need an up down link and £30k/year ($45) per household. This gets everyone Internet access. Every household installs its own 802.11b equipment at a cost three to $400. The community will have to band together to buy the earth station. That's a really viable proposition.

A Community Approach to Asset Based Telecom
Under that model you bypass everyone. Now you can then go to the local cable company. You tell them we are ready to buy a 10 or 100 Mbit/s circuit into the Internet.

COOK Report: The issue then is one of having a knowledgeable General who can lead an army?

Cochrane: Yes, if you have enlightened businesses in the area, they can also trigger the whole thing. Can't you just imagine a plant somewhere in Wales? Part of the community charter of such an organization might be to get into that space on behalf of its employees and community in which is operates. The only problem is were such a high-tech company in the pocket of the Telcos, it would be hardly likely to stick its neck out and say "oh by the way, here's how we will help you avoid being a customer of the phone company." As a result you may have a kind of incestuous relationship problem.

COOK Report: But then again the folk in Wales could take an inventory of local corporations and figure out which ones might actually be willing to help.

Cochrane: You must remember that you might not have a critical mass of people, or you may be short of money. Nevertheless, this sort of thing has been done before in the USA. It was called Community antenna TV, and eventually it became Cable TV.

The Lottery in the UK is similar to the ones in the United States. The profits that are realized are fed back into the community for worthy causes like the building of new hospitals or the creation of schools for the handicapped. The building of bridges. The providing of social facilities for disadvantaged groups. And perhaps the creation of community owned networks by means of which to access the Internet.

COOK Report: It sounds like what you need to do is to get a bunch of capable 'Generals' like Hughes together and set them loose on instructing and leading their communities. There is a recent paperback book called Optical Networks published as a Wiley research report that came in my mail in late February. This book is really a handbook and guide for building community owned asset based optical networks.

Cochrane: I have seen some wonderful websites in the UK along the same lines. I think you are right. There is an Army building. It is one that needs a General to orchestrate its direction and make things happen.

Spread of and Business Model for 802.11
Now the other day in London I got in a taxi and drove down the length of Oxford Street. There is contiguous 802.11b for the length of that mile and a half long street. You can now walk into just about any computer store and buy the necessary equipment.

COOK Report: What the business model behind these networks?

Cochrane: Certainly. I had not heard much about Boingo before you sent me your write up about it. That is really a pretty good model. It is nothing more and nothing less than a bandwidth brokerage. I have 802.11b in my home. If you pull up on the street outside I have no means of providing you with service other than for free.

COOK Report: But you can, if you choose, lock other folk out?

Cochrane: Yes but people generally don't do that and I am quite happy to let you have service on my system for free. Now if the traffic begins to pick up I will have to do something and I may say to myself: can I make some money out of this? The problem for me is that I certainly don't want to have to set up a billing system. I don't want to have to get involved in collecting money. But if Boingo is around and looks reasonable I will register with them. If they want to give me some money for outsiders who come along and want to use my system, let them go ahead and do it.

I like this. It seems to me that Boingo is a meta-layer. It is not providing points of presence because you and I are doing that. What they are saying is that they will manage our radio resources for us and will collect the money. And customer roaming the UK or the US, signs on with Boingo. They put in their 20$ and that gets them 200 minutes of Internet time or whatever the number is. They can come sit outside my house spend the entire 20$ and I make 10$ on the deal or whatever the percentage is.

COOK Report: Well you hope you make 10$. Last time I looked on their web site there was nothing that I could find regarding their revenue sharing model. But, unless they can make people sign rigid non-disclosures, I'd think that would become known before long.

Cochrane: Correct.

COOK Report: My first reaction was that this was a way to collect money from people who could have taken advantage of 802.11 nets themselves for free, if they had only known how to do so. That it was charging people for showing them how to grab other folk's bandwidth for nothing. But just so long as you have an opportunity to become part of the deal, if your bandwidth is becoming heavily enough used, your option is to buy more and you should be getting enough from Boingo to enable you to do so.

Except in the US one problem with individual 802.11b nets is that many individuals run them off cable connections and that they would have a problem increasing their bandwidth because the cable companies don't like them doing it anyway. In such a situation it might be a bit difficult to increase upstream bandwidth. But take your Oxford Street example. Is there an opportunity for an upstream bandwidth aggregator there since Boingo doesn't do this? Could someone put in a pop and invite 802.11b providers to put their radios in the pop? Is that another possible evolution?

Cochrane: I think all of this is up for grabs. I think there is almost a kind of Napster opportunity here where someone can come along and grow from 0 to 90 employees and 0 to 90M income in 18 months. I look at people who want billing machines to charge by the minute. Wrong. What has accounted for the phenomenal growth of mobile has been the prepaid block payment plan. People don't want to deal with billing but to buy a block of time. Everyone buys the £15 Internet access plans that allow you to go on anytime day or night and not to have to care what the phone bill is.

COOK Report: It is unmeasured usage even in an environment, which normally has always measured the usage. Right?

Cochrane: Correct.

Local Loop and the Future of the Telcos
COOK Report: If you had enough sufficiently robust 802.11b clouds, how could this play out in enabling the replacement of the local loop?

Cochrane: How about slipping IP telephony into those clouds? You could do telephony over 802.11b. This is what is terrifying the Telcos. It is also terrifying the mobile phone operators who have stopped to think about it. The other thing that you can very easily do is put a very small mobile phone base station into your PC so that you go mobile phone to PC to IP telephony.

COOK Report: What do you see coming out of the alleged voice telephony readiness of Windows XP?

Cochrane: I think there is a level of ignorance here that goes like this. You have Directors, Executive Directors and Managers in Telcos who are using something colossal, a 5 ESS switch with a life of 35 years. You then get companies coming along in the IP space with something the size of a shoebox, which is IP telephony. Of course nothing could be further from the truth, but it is a great sales pitch none the less.

COOK Report: There was one up for auction on Ebay the other day. A $5M device with a starting price of $350,000.

Cochrane: Precisely. That helps to indicate the nature of the problem faced by the industry. Everybody can see the cheaper cost for IP telephony while seeing that at the same time for it to work you have to waste colossal amounts of bandwidth! There is no free lunch here. You can't suddenly get rid of all your 5 ESSes and their point-to-point switching by suddenly replacing it with a packet switched network. If you don't sweep aside the circuit switched network entirely you have to do things to it that make it look like the circuit switched network to get anything like the same performance. To reiterate: with IP telephony to get any thing like the performance that you are enjoying now on this phone call you need to waste a huge amount of bandwidth.

What has changed? Bandwidth now costs nothing. I mean seriously costs nothing if you have the right infrastructure, which is lots of physical fibre all over the place. We do indeed now have this. But stop the clock. Don't start saying we have too much fibre. I think that if broadband (IP telephony and 802.11) in the local loop really started to take off, we are going to go from a bandwidth glut to a bandwidth famine in one giant leap.

I'm thinking that we are at a cusp where in you can imagine the following process. Say you and I are running our respective Telcos and we look at each other and admit: " you know we've had these by DSS switches now for some 15 years, it's time to think about replacing them. Let's go to the manufacturer and see what it says The Next generation switch will look like and cost.

I can tell you that the people who want to invest in the next generation of manufacturing telephone switching systems are getting few and far between. The reason is simple. Everyone is getting terrified of IP telephony. The last thing the manufacturers want to do is set up a production line to build the follow-on to the 5 ESS and find that there is no market for the new product. All these forces are coming to bear at the same moment. Consequently I look at the picture and say yes there is a fibre glut because we have no Broadband. And the stuff that they are calling broadband certainly is not. You and I know that ADSL compared to 100 megabit per second Ethernet is pretty pathetic.

COOK Report: What I think I hear you saying is that if you put together the right people and the empower them would be knowledge of where the fibre and the technology is, along with the doors to knock on, you can bring together a community of people that will manage to get valuable things done. And one of the reasons for this achievement will be that a group can aggregate its buying power in a way that an individual cannot. Finally that, if such a group knowing what it wants, approaches the fibre providers or other sources of access to telecom networks and the Internet, it may well be surprised at what kind of a good deal it can get.

In fact, talking with you reminds me of what I heard from another source. Namely that one of the major infrastructure owners in London has just offered a group 100 megabit per second fibre-based connectivity for several of its sites around London. So far, a very ordinary story. But there was a most extraordinary provision attached. They would be granted the bandwidth on the sole condition that there been no IP traffic placed on the fibre. The application would be one that would shift huge files of data. These files were primarily huge audiovisual files belonging to the entertainment industry; huge files for the graphics arts industry; and huge files for architectural designers. They would go point to point to around London only as long as the customers pledged to keep all IP traffic off the fibre. I could hardly believe this because the only way to enforce the IP ban would be to put sniffers on the network.

Cochrane: It speaks eloquently for the phone company's paranoia. Someone is running darn scared.

COOK Report: It seems that to enforce something like this, you would have to make very intrusive attachments to the physical network itself?

Cochrane: Yes. It's like having the highway and saying that you will only allow trucks on it and not cars.

COOK Report: In other words there is some entity with a lot of fibre that's not getting any income for that fibre. Action like this must speak to its desperation to find a use for it just so long and only so long as if it doesn't risk that use taking away any part of it's voice business.

Cochrane: I would say that is a reasonable assessment. As technology has evolved, we have been at this point of apparent collapse several times before. I always think of the transition from canals to railroads. Coal for heating the City of London used to travel by barge at 3mph. When the railroads came along no one could understand why that coal would suddenly prefer to travel on iron tracks at 60mph when it just as easily could go along a placid canal and in a barge at 6mph. What they didn't understand is that people wanted to travel.

COOK Report: Well, as these companies continue to crash and people like yourself and myself and Dave Hughes are out on their own as individual entrepreneurs, one of the problems for us is that without good broadband connectivity, we have one hand tied behind our back in launching our own new businesses.

Cochrane: Quite so.

Next Stage Education?
COOK Report: Consequently, as I watch the strains and stresses way down ever more heavily on the creaky old system and talk to folks like you and Dave Hughes and watch the postman brings books to my door like Deborah Cameron's Optical Networking, I become convinced that the next part of the process will be more one of education than the development of new technology, we need to increase the understanding necessary to speed the restructuring of the larger older companies.

We also need to enable a learning or building process that will assist in the creation of more community owned telecommunications infrastructure that is free to be able to grow up without encumbrances of the old infrastructure base.

Cochrane: Agreed. Last thing the Telco needs right now is the encumbrance of the local loop. Local loop is a real pain in the neck; the cost of about 80% of the Telco's operating budget to maintain. Consider what happened with the advent of fibre when it was installed into the long lines network. In a single stroke the operational costs of the Telco went from fifty-fifty for long lines and local loop to 10% long lines and 90% local loop.

It should be absolutely obvious that ADSL is a dumb idea, and all other DSLs forms are very limited for the simple reason that clock speed, RAM, and hard drive capabilities of terminal machines is advancing far more rapidly than any technologies in the copper local loop. Optical fibre is the only thing we have that can keep up with the speed of chip technology. Just this week IBM announced a chip technology with a clock speed of over 200GHz. They are going to release this technology running at 100GHz. Now what possible use and future can copper wire have when your PC will have a 10 - 50GHz clock and a 1Tbyte HD within a decade?

Think what might happen to the Telco were it to do something really sensible and divest itself local loop by whatever means. That newly divested Telco would have a business model using 10% of its cost and long lines and 10% of its own share of its use of the divested local loop. It could actually make money by renting space in their now vacant cable ducts. They could offload all the nasty cost of this terminal equipment to people like you and I. After all we manage our own PCs. We manage our own television sets and hi-fis. Why not our telephony equipment?

COOK Report: What you are describing is what St Arnaud calls asset-based telecommunications. In this model you push the divested assets out to the edge of the network and into the hands of the individual users. The users become the new owners.

Cochrane: Yes. And with the release of capital from the divestiture, we can either start extending the existing fibre or exploiting that fibre and moving to a market, which in the first instance may mean 802.11. As an alternative, we might put in an even better wireless LAN standard. Or it could mean that we put a fibre Ethernet into the local loop running at 100Mbit/s. All this would be a substantially better world than this limited copper business.

COOK Report: But what you are saying is that if you break the logjam at the local loop, you may enable the rumps of the operating companies that are left to operate long haul loops, in other words, the intercity fibre that we do need. And possibly to operate them successfully.

Cochrane: Yes, we need the long haul. We need the capacity absolutely everywhere. And let's make this absolutely clear. I do not believe that we can have wireless broadband world that we desire without running fibre into absolutely every building.

Why? Because we need cellular systems for broadband access. That means we need them in this room, in this building, in this home, on this lot so that I can access the wireless network at 10 megabits per second. This cell is replicated 10 houses down the street. It is done in such a way that the power does not spill over 10 houses down the street. Now the guy next door has a different allocation of codes within the cellular system spread spectrum and his house is also fed with fibre. That house next door to him broadcasts on a third pattern and so on down until you get to the 10th House, which is back on the pattern of the first. Having a network of these micro cells is the only solution that is going to cope with a twofold problem. One, the provisioning of wide band to mobile devices, and two, the clustering of those mobile devices because human beings meet with each other and so will their devices. Right now I'm the only person in this house. I have a computer running and a mobile phone. Tomorrow there may be six people here each with a laptop and mobile phone and 10 years from now those devices could be using colossal amounts of bandwidth. You cannot support that with copper or any other kind of narrow pipe into this home.

We have this chaos in the network already and we often see it when we attend conferences. I often use this example when I am speaking at conferences. I say you know coffee will bring down this local cell site. Everyone looks up at me, astounded. What happens is simple. When the coffee break occurs 300 people may march into the lobby and 200 of them will reach for their cell phones and immediately initiate phone calls. This unleashes demand on the local cell that the system often cannot handle. The only way that you can cope with such sudden surges in demand is to put fibre into every building in other words into every cell site. It is in a situation like this that the idea of micro and pico cells begins to pop out.

COOK Report: Those are intriguing ideas about how to fill up that fibre a bit down the road. But in the near term one has to start somewhere. And I wonder what you think of Hughes' idea after his return from Wales where the community leaders there talk in terms of establishing upwards of as many as 600 small scale community owned and operated 802.11b Networks known as "e-fro"s. The "e-fro" term is derived from the Welsh word for Council, in this case electronic Council. The idea is that each community owned and operated "e-fro" be responsible for establishing its own technology and for its own user support.

Cochrane: And with each responsible for the training of its own users, everyone understands the importance of doing it right. For if they goof up, they hurt only themselves.

COOK Report: As these community owned networks develop and mature, then your next steps are building out of this greater kind of density that you were describing.

Cochrane: I go back to the American model for cable-TV, this is exactly what happened. It became so big that people saw money in it and decided they had to give a professional care and attention. I watch what is going on with telephony and I think " déjà vu." With the phone companies we are asking Turkeys to vote for Christmas because if they blindly rollout broadband, they could shoot their bread and butter business which is voice telephony.

If somehow they became smart, they would roll this out and change their own business while they do so. Unfortunately, what has happened is that they have squandered the opportunity to do something presented to them by the last 10 years. They are now being dragged into broadband kicking and screaming at a point where they have no money for the investment required and have huge debts that are likely to render getting the capital needed impossible.

COOK Report: And in the United States and they are currently trying to by Congress in order to be granted an unregulated monopoly which they would then use to extend the industry's misery by staving off the inevitable.

I've just done a great deal of writing and thinking about Level 3. I'm still in the opinion of that structured as they presently are they don't have a chance. But they do have a great many intelligent people. That being the case I would think that they ought to be developing a contingency plan. If the local phone companies continue not to get it and manage to hang on and on, then it would seem that a company like a Level 3 ought to have a plan in its pocket that it could take out and explain to the authorities that it had assets that could be used to stitch cities together, given the creation of a stable environment.

Cochrane: That is an interesting idea, but I think there is another issue at play, which may complicate factors somewhat. Many telephone companies in the USA have totally divested themselves of all their technical capability. Their argument has been this is not about technology but about "service." We are the telephone "service" and this is what we do.

I think that they are then looking at the worst of all possible worlds. Why? Because they no longer have the intelligence to be a smart customer for equipment for their own networks and understand what it is they need to do to work themselves out of the fix that they have gotten themselves into. They simply don't know what to buy. Having become overly dependent on contractors, they have no technology radar and they also have no worthwhile market radar because their idea of market analysis is to ask the customer what they want when the customer doesn't know enough to have a clue. They won't see any danger coming until it is too late.

COOK Report: Do you see any particular telephone companies in the United States that fit this mould more than any others?

Cochrane: Not really. I think what you have is the total change of the first hundred years of the Industrial Revolution being compressed into a telecommunications revolution of perhaps 15 years.

COOK Report: I would suggest that the one phone company that fits you description more so than any other is Global crossing. That outfit was run on the one hand light Beverly Hills-based financial traders many of whom like Gary Winnick were pupils of Michael Milken and on the other hand by some technical people who have observed that the people in their front office don't know the first thing about telecommunications.

Cochrane: True. I think it is a little bit dire out there. In the large telecom companies I have watched the corporate memories dying. They are dying with each succeeding round of layoffs of those 50 years old and above.

COOK Report: I am hearing similar stories about the research arms of Sprint, MCI, and AT&T.

But what about IBM and Cisco and Intel and even Nortel? What about the companies that make the equipment that enables the phone company networks? Where do they stand in all this? You have been attacked and one other coalition of Silicon Valley companies that are trying to reinvigorate the American economy by encouraging broadband. But when you look at what they say they are a mincing around the edge of the problem and not meeting it head on. They're talking about Band-Aids for an industry with gangrene.

My only hypothesis is that perhaps someone like John Chambers of Cisco can cannot stand up in public and speak truthfully about the situation for perhaps two reasons. One: fear of alienating some of their sick customers and two: fear of liability from stockholders for saying something that will depress the share price. Can someone like Chambers afford to admit to his stockholders that it may be at least two years before things turn significantly around?

Cochrane: I believe that we have a bit of an incestuous relationship operating in this case. To some extent these companies are in the pockets of the network of providers.

COOK Report: But as the capital expenditure of these companies crashes, doesn't their influence over them get much less?

Cochrane: You would hope so. But I would think that unless these companies see replacement customers with pockets full of money, but they are not going to be a rush to abandon the devil they know for the devil they don't. You can see them asking: if we help kill the Telcos who will be left to buy our equipment? You also face the impediment of human inertia involved and the reluctance of people to have to go out and do new things.

You have to have some cadre of leaders with their minds pointing in the same general direction. Until you achieve this and you could have the best technology in the world and still not escape from your present dilemma.

On the other hand you get these little disasters like the telephone companies overselling of WAP, people believe the TV advertisements. They go out and buy it and then find out that it's useless. It spreads like wildfire and then suddenly dies out. People try it and then catch a cold. You have this fluctuation between overselling and underselling and between over investment and under investment with no easy way out.

New Technologies
COOK Report: While we are waiting then for all of this to sort itself out, what are some of the interesting companies that you are watching and working with?

Cochrane: You have a large amount of standard on the shelf stuff with which we can do an awful lot. For example there is a company in Maryland called etenna . These are the people who developed the antenna for the stealth aircraft. They made a kind of wallpaper that functions as an antenna. They make antennas as architectural features. You can make a multi frequency, adaptive, steerable antenna as well. Their antennas are strictly planar you can even make the license plate and your car into an antenna. You can even make the roof of your car into an antenna that can communicate with a satellite.

(In the interest of full disclosure let me say that I am an advisor to that company and I actually own some shares in it.) Now a company that actually asked me to be an adviser and wanted to give me shares by one in which I refused because it would of been a conflict of interest is a UK company called Antenova. Antenova has some very nice three-dimensional antennas that can do fairly rapid switching on a sectorised basis for mobile devices that are hand held, or worn, or carried. These are key technologies that are currently in beta test stage.

There are devices being made to bypass the fibre network. There are companies like fSONA in Vancouver Canada that have point-to-point high-speed, short haul optical links. There is a company, SOMA networks, in Silicon Valley that has 60GHz 802.11 equipment working. They have a terminal that acts as a multimedia intermediary in that it allows your fax, your phones, your laptops, and a whole bunch of other stuff to operate over a wireless IP network.

COOK Report: What about power supplies for mobile devices?

Cochrane: Here I think fuel cells will play a major role. We have people playing with fuel cells that will give us something like 20 to 25 times the energy density of a conventional battery. I think this is kind of an important thing to have around. On the chip front there is a company worth looking at called Quicksilver. Right now Quicksilver is focusing its attention on chip sets for 3G phones. Instead of having a microprocessor-based approach to building a mobile phone, they have said that they will only use a part of the chip when it is required. What is more, they will reprogram that part of the chip to use it over and over. In doing this they have managed to drop the power consumption of the mobile phone by a factor of six.

So you take a conventional mobile device and use a chip set that requires one-sixth the power and a fuel cell with 20 times the energy density of a normal battery. That is 120 times the capability of the normal device. Next let's attach to that device an antenna from etenna or antenova that is four times as he efficient as a normal antenna. Now you have a device 480 times more powerful than any of our mobile devices today. I get excited about that.

COOK Report: In the sense that you can use it longer without recharging?

Cochrane: We're talking about a blank envelope here. You could either consume six times the amount of power that you're using today or you can get the same RF performance with only a quarter of the RF power and so on.

COOK Report: Do you have an opinion about ultra wide band?

Cochrane: You just triple the capacity with the same functions and I can't get all that excited about that. There are also lots of people getting excited about OFDN orthogonal frequency Division multiplexing. This is almost identical to spread spectrum except you don't need the training sequence and it's easier to implement. I can't get that excited about this either. All you're doing is saying we have a frequency space here into which we are going to cram signals.

COOK Report: What is your opinion of those people who will warn that the spread of spread spectrum will ultimately cause interference?

Cochrane: If you want to hear the ultimate in interference just pick up a GSM phone. This was the worst possible specification for mobile phone people to come up with. You don't want a packetized mobile phone system because you get instant interference with any device. The sound of the 21st century is that "yaga dag daga" as your mobile phone logs on and interferes with every possible device. Spread spectrum, simply and fundamentally, does not interfere on anything like the same scale.

If I were to pick something that may be an Achilles' heel for all of this, something that everyone is ignoring and no one is doing anything about, it is troubleshooting. You build the whole system and then somebody stands up and says: "Wait a minute. How do we find a fault?" And then everyone looks at each other and says, "Oh damn."

And at that point you have a huge amount of retrospective effort to put into the design to get it to tell you where it is failing. This kind of problem in these chaotic and networks that are going to be set up my individuals needs thinking about very, very early on. For example, supposing and in 802.11 network you get one faulty rogue transmitter that starts to drift in its transmission up and down the boundaries. You will then find that you cannot hop onto it. You cannot even find it and you have a serious problem because it is wiping everyone else out.

Or consider a different situation. You have an 802.11b network that is working. Then it's not working. Then all of a sudden it's working again. You have no way of knowing that the problem is caused by someone with an illegal imported cordless telephone that is pumping out about 5 watts of transmission power. You can't find the thing. Now it is all sorts of those little issues that do worry me a bit.