Telepocalypse

Vote for VotI!

My apologies. Mea culpa. I sincerely regret any confusion that may have been caused.

VoIP is a complete waste of time.

OK, I exaggerate slightly for effect. But in fact, VoIP is indeed not very important. Why? Because it mistakes technical efficiency for economic efficiency. What really matters is Voice Over The Internet (VotI) -- the network or networks, not the internet protocol itself. Yes, by definition that means VotI is VoIP, but the converse isn't true. There is a difference.

You can run VoIP on a private network: VoIP without VotI. If that network is one you own and control, then having a stupid IP network isn't a massive advance over an inflexible smart network. If a smart middle node gets in the way of progress, just change it. No service provider gets in the way, no third party permission required. Yes, there have been "revolutions from the edge" in corporations (e.g. the deployment of PCs was initially driven by departmental budgets to but spreadsheets for accountants). But that's the exception, not the rule. Most IT and communications change is centrally driven, because that's where the budget lies. Apart from anything, you're probably going to get fired for breaking the corporate security policies: don't try to mount an insurrection armed with IM clients and Linux boxes hosted in cubicles.

We're all familiar with public smart networks and their comprehensive failings. Conversely, the stupid Internet crushes all rivals, one by one. Which is why the end-to-end principle is possibly misstated. It isn't that functionality is best placed at the edges. That's just a corollary of control being best placed in the hands of the user of that functionality, be that an individual or a corporation. And since we don't (yet) operate and control our own networks, that in turn pushes the functionality to the edges.

So VotI is important to society because it represents a shift in the control of economic assets. VoIP is just a technology with negligible stand-alone economic significance in the absence of the Internet. It's like a wave-particle duality theorem for neworking. You get a much better understanding if you see the Internet as being as much an economic phenomenon as a technical one. The network of networks has formed because of the low cost on internetworking. Kind of obvious, but needs to be said.

One consequence is that the much-vaunted ability of companies like AT&T to offer consumer VoIP solutions that run over their own private backbones is irrelevant over the long run. Private IP networks are only needed to the extent that (i) there are technical dificiencies in the Internet [hard to find] or that (ii) the digital identity and governance infrastructure is missing and you are swamped by abuse [still some deficiencies, but improving].

What does this all mean for the end-to-end principle? Imagine a local ad-hoc mesh network. These networks are effectively little communities, and the "user" is the whole network. Intermediaries only transit packets on the basis that they are willing tolerate your through traffic. There's no money flow, no contracts. Everyone's a stakeholder. I wouldn't be surprised to see some regression of functionality into the "core" of such a network. This challenges the "technical" end-to-end principle, but perfectly in line with the "economic" version. (The end-to-end principle only says the default place is the edge, so it doesn't contradict it. But the end-to-end principle doesn't say anything about when or why that default is broken.)

To push the point home in a more concrete manner, just consider the degenerate case of an open WiFi neighbourhood access point. The "edge" is my laptop borrowing your connectivity. But it would be perfectly reasonable for you to apply some form of QoS to make my packets second-class citizens compared to yours. After all, you're paying for the upstream broadband connection. Some "smarts" have crept one hop upstream from the edge. Again, the "economic" end-to-end principle tells us where the functionality goes, whereas the "technical" one doesn't.

He who pays the piper calls the tune. It's as simple as that.

PS Notice how BT changed their logo from a piper to an Internetish connected globe. Anyone else see the irony in the piper's escape?

A terrible toll

This is a tale of deceit, greed, and intrigue. Well before the broadbandits burst onto the scene, the copper highwaymen have been perfecting their tricks for decades. Their sly plans continue to this day.

How I found out about the scam

One of my many tasks on return to the UK from the USA has been to select a landline telephone provider. Like many people in the UK, I have a choice of exactly one access network provider, namely BT. Whilst cable passes the front door, it somehow failed to make it up into the apartment block. And wireless broadband has only been deployed on the other side of town. (There's a hill I can see from my living room about 10 miles away across the Forth river -- anyone looking for a good site for a tower, just drop me a line, I'll tell you where to build it...)

I can't buy naked DSL, so I might as well make the most of the circuit-switched connection I'm forced have. As in the US, it has been possible for a long time to dial a prefix for an alternative carrier, or to permanently switch via carrier pre-select. Both cases have required a separate bill from the basic BT connection charge, although that's about to change; the third party provider can soon opt to remit the BT line charges too. There's been an explosion of service providers, and a profusion of different rates and prefixes.

One very useful website has been set up, uswitch.com, for enabling comparisons between utility providers. Incidentally, there's a business opportunity lurking here. I want a box that does intelligent carrier preselect and least-cost routing for me. When I dial a call, it uses my broadband connection to look up the best provider from an on-line database. It would also optimise the provider accounts I hold over time, based on my call patterns. (Some accounts require a monthly fee for lower call rates.) Take the uswitch.com idea, and take it down to the individual call level. Any takers?

Anyhow, I haven't yet signed up for any alternative carriers. Whilst I procrastinate I've been using an alternative means of bypassing BT's outrageous rates for the many international calls I make. (This is all in addition the usual free calls I make over Skype, of course.) A whole raft of countries can be called by dialling a special number of an alternative access provider. You don't need an account with them, it just gets billed as a normal BT call. In their own words:

Call USA for just the price of a national call!
- 1. Just-Dial 0870 794 0000

- 2. Then dial your international number in full

- 3. That's it - you're connected - try it now!

NO accounts to open NO credit cards needed NO bills from us

So how does this work? How are they making money? How much am I payng to call this number? Well, this is where we discover a dirty great big scam being pulled on the British public. The villains aren't these toll-bypass carriers, though. Rather, the whole numbering system has been poisoned to favour the telco incumbents. Oh my, this is a real humdinger of a story. Hold on tight.

Numbering in the UK

A bit of dull background first. Following years of confusion and scandal, the regulator set up a national numbering plan in the mid 1990s. In the UK, normal geographic phone numbers begin 01xxx or 02xxx. Mobile numbers (plus some other stuff - another scandal for another day) were given 07xxx. Premium rate numbers have 09xxx, and are regulated by a separate body. Finally there's the 08xxx range for "non-geographic numbers". Keep this last one in mind.

The transition to the new numbering system was so expertly managed that people in London only had to change their number, oh, four or five times in a decade. Gives you a clue as to how ineffective regulation can be. Clueless and toothless regulation is a bit of a theme here, actually.

Lies, damned lies, and BT price lists

Everything in this scam hinges on the phrase "just the price of a national call". You, and about 98% of the UK public, probably think that's the same as calling a geographic UK number, right? Nice try, no banana. The national call rate was, indeed, the price of a UK national call on BT's standard call plan. It's expensive, at 7.5p/minute peak times (about $0.14). But BT scrapped standard calling plans this summer, and forced everyone onto the more expensive BT Together plans. These (well established) plans have lower call charges, but higher monthly fees. So feel free to desert BT for metered calls -- they've just unilaterally decided to charge you more regardless. Only a few obscure social-obligation plans remain on the old call charge scale. You and I don't qualify.

And the "price of a national call" isn't yet regulated, either. So if you're not calling from BT, it could be anything your provider feels like charging. Maybe a bit more than BT, or a lot more than BT from a mobile or less scrupulous alternative carrier. (Orange, for example, charges post-paid customers 10p/minute. Good luck finding out the charges -- they aren't posted on the mobile carriers' web sites. If you're pre-paid, they'll even charge you for calling customer services to find out!)

To put this in perspective, the cheapest UK toll-bypass carrier charges a 1p flat connection fee for a (geographic) national call, and that's it. So if you spent 10 minutes calling an 0870 number, you're paying eighty times the going rate. Yes, you read that right, no need to rush to Specsavers for some new glasses.

You scratch my back

Now for the really juicy bit. It isn't just the carrier raking in the cash here. These numbers offer revenue sharing to the callee. This can be up to about 4p/minute. The volumes of calls can be huge, and business plans commonly run into the millions of minutes. That means you're paying the callee for them to answer your call. And you're paying to stay in their queue. And paying to wait. And paying more to wait more. And paying to talk to Mandy the customer service rep who'll no doubt transfer you to Dave her supervisor for an extra chat. Pay and pay and pay some more.

So you can see why British businesses have flocked to using 0870 numbers. Forget the US approach of offering freephone 0800 numbers. We good old Brits know how to give our customers a good rogering.

(I'm going to book some airline tickets using flyer miles later today. But I'm not calling the British Airways 0870 number in the UK, no siree. I'm using Free World Dialup to call their US 0800 number at no charge. So take that and stick it, etc.!)

And it's not just businesses using 0870. Want to call your police station? Your school? Driver licensing? It goes on and on. University students in campus accommodation can only be reached via 0870 numbers, with the university taking a cut from every inbound call. Credit card swipe machines autodial 0870 "maintenance" numbers in the middle of the night.

It wasn't meant to be this way. The 09xxx range was supposed to be for premium rate calls. They're pretty tightly regulated. For example, you are not allowed to have a queueing system on a premium rate number. Either the call is properly answered immediately by the service provider, or there is no charge. The cost must always be prominently displayed. And they were supposed to be the only place for revenue sharing between carrier and callee. Whoops. A triple regulatory failure when it comes to the covert premium rate 0870 numbers and their cousins.

Billion-dollar bank heist, no arrests made

You might be thinking what's the big deal really? Charging 8p a minute isn't much of a premium rate, is it? Err, actually it's more expensive than some "offical" premium rate numbers. But here's the bit that really surprised and shocked me:

... BT wasn’t wishing to jeopardise the future of the UK [Number Translation Services] market as this accounted for between a quarter and one third of switched revenues for BT. [my emphasis]

Wow. And you can bet this figure is increasing rapidly too. We're talking billions of pounds here.

As some of you will have noticed by following the links, there's a small consumer backlash forming. The site SayNoTo0870.com lists geographic alternatives to 0870 numbers. Remember all those times on the back of your credit card when it says "If calling from outside [your country] dial +44 123 4567890"? Well that's the number you really want to dial, regardless of where you are.

Oh, and could someone please update my little personal least-cost call routing box to do these translations for me too, I'd be grateful. Cheers.

Take-aways

There are several phenomena here of note. The first is that this is microeconomics in action. Who gets more value out of the call? If I need to change my airline ticket, then my airline has temporary monopoly power over me. Need to tell your pension provider you've changed address? You're hardly going to shift your pension funds about to avoid a 50p call charge. It's a beautiful example of price discrimination and market economics in action. (Of course, it's also underpinned by fraud and obtaining financial advantage by deception since the telephony transaction is neither based on mutually clear terms nor freely entered into -- two essential elements of a valid contract.)

The second thing of interest is the size and complexity of the billing tables. You practically need a broadband connection just to download and view the BT price list! I counted 58 screenfuls of data. Obfuscation of real prices has long been an essential part of operating a telephony service. The art is now progressinging into a scientific discipline. How can you evaluate the relative attractiveness of a VoIP provider to an incumbent if an increasing majority of your call costs are to mobile, non-geographic and premium numbers? It would take you a week to re-rate you bill by hand.

Finally, there just the sheer cunning of it all. If the smart network loses against the stupid network on an architectural level, then just change your locus of competition to the financial and political spheres. Tens of thousands of businesses and government agencies now have a financial interest in perpetuating the old system. The media aren't exactly going to be up in arms when they and their advertisers have a finger in the pie. Who cares about free person-to-person VoIP calls when you need to pay protection money to a telco in order to engage in any commercial or public activity?

PS If you have any comments or complaints about this article, please contact me on my 0870 number listed here. (Only kidding!)

One to watch

According to The Feature, Siemens are working with Flarion to deploy FLASH-OFDM technology in the 450Mhz range. I've written previously about Flarion and why it's a significant technology.

This announcement is interesting because it very much follows a "disruptive innovation" trendline. The frequency band in question isn't universally available around the world, so isn't attractive to high-end business customers who travel internationally. But the lower frequency means much better in-building signal penetration, as well as lower cell site density (as the article notes). The spectrum is also pretty cheap. This means more attractive economics than traditional 3G services in achieving acceptable coverage. The average user couldn't care less about megabit throughputs on a mobile handset, so the lower frequency and peak throughput isn't a big deal. Plus Flarion's FLASH-OFDM is highly spectrally efficient when measured by total system throughput (rather than peak throughput for any one handset).

Expect to see VoIP being deployed in this frequency, and innovative services that attract sub-prime market segments (such as teen users). If I was to launch a Skype phone, what Siemens are doing is exactly what I'd be doing to get there. Expect future surprises.

Supernumary

According to Voxilla, Vonage are charging $4.99 a month for a London phone number, and Lingo are charging $9.95 a month.

So how much is a telephone number worth, given they're not fully tradeable commodities?

Let's make some assumptions and create a model:

(i) the median retail price of a number is $7.50 a month (i.e. $90 a year)
(ii) the "raw material" cost of a phone number to the value chain as a whole is zero. (Vonage et al may be paying others for the privilege of allocation and resale of numbers, but there isn't an number mine in deepest Africa being excavated at great expense for more precious integers. They came along for free somewhere between the Big Bang and Fermat.) There's no cost of inventory.

(iii) the SG&A expenses are 10% of revenue

(iv) the mean tenure of a customer is 2 years. This is very conservative; people move home on average every 7 years.

(v) each business has a minimum rate of return of 15% per annum on investments.

Remember, we're only after order of magnitude here, so there's no need to be pedantic about the numbers. We get $180 in revenue and gross profit, less $18 in expenses, making a net profit of $162 spread over 2 years. How much would you pay for a financial instrument that required you to pay out $18 on day 0 and then gave you $7.50 a month for two years?

To avoid any fiddly net present value calculations (I'm a mathematician by training, thus extremely lazy) just say this is like someone promising to give you $162 one year from now, risk-free. At the corporate rate of return a phone number asset is worth roughly $162 less 15% = $137.70. That means you would be willing to hand over around $140 to obtain that revenue stream.

Compared to a $10/year domain name, phone numbers are very expensive. If you though Verisign was a rip-off, you're only dealing with the amateur side of the address space sport.

Whilst we're on the economics of phone numbers, it is worth noting that not all numbers are equally valuable. Obviously, there are sequential, alliterative or lucky numbers (123-4567, 777-0000, 888-8888) that are sometimes sold for a premium. Then there are desirable area codes, such as Manhattan's 212 prefix, or 0207 in central London. Pedants writing in to say that the London prefix is really 020 with will be ignored. Then there are desirable and undesirable country codes. Not many multinational businesses are gagging to get hold of Nigerian numbers, I suspect.

Given the increasingly easy ability to have virtual numbers in any geography, I wonder if area and country codes will become mini brands in an of themselves. In part the brand will reflect the geographical origin of the code, but may increasingly have a life of its own. The limits by convention on number length might render 212 as a sign of yuppie urban aspirations. A French number might say more about your political outlook than your place of residence. Could some libertarian-leaning billionaire buy out the remaining address space of New Hampshire numbers and offer them in conjunction political party memberships?

This may sounds like loony ramblings today, but in a decade from now geographical pricing for calls may have collapsed and vanity and community may be the drivers of how you number yourself. Unless of course, we're all on Skype instead...

UPDATE: Before you all get too excited about the end of telephone numbers, bear in mind you almost certainly can't type my wife's Lithuanian name on your keyboard. How would you feel if your keyboard was shorn of a few letters? You had to drop the vowels from your name? Now how keen are you on non-numeric identifiers? Ah, the joys of English alphabetic imperialism.

TV trespass

So, if the TV stations are spectrum hogs then here's a way to deal with it.

Remember when, in the US, the cablecos started up? They got their content by, ahem, redistributing broadcast TV stations.

Here's my subversive little proposition. Let the TV stations keep their spectrum. Just make it so that anyone can redistribute their content on a not-for-profit basis. What's good for the cablecos is good for you and me.

Suddently, every man and his dog will be legitimately uploading and downloading TV progams. BitTorrent and torrentocracy will become as mainstream as VCR and TiVo. And you've just sown the seeds of the replacement content distribution businesses, divorced it from connectivity provision, and given symmetric network connections a big boost.

If we're blessed with another kid, and it's a boy, I think I'll have to call him Machiavelli Geddes. Although Mrs G, looking over my shoulder, doesn't seem so keen. Perhaps she thinks we'll have another girl. Machiavella Geddes, then, I guess.

Citibanked

Off-topic but amusing: on my most recent credit card statement (i.e. not from April)...

For details visit www.automatic.billpayment.aadvantage.citicards.com to have your bills paid automatically with recurring bill payment.

Seriously. Someone must have won a bet in their marketing department to come up with something that long.

Abe Lincoln speaks

I'm listening to Brough Turner give an excellent presentation on why we need wireless spectrum deregulation and a re-framed debate on right-of-way access for optical fibre to the premises. You're probably familiar with a lot of the arguments already. He neatly frames the access network debate as "layer zero competition". What is this? Before you lay down a physical network, you need access to rights of way. Competition policy needs to focus on enabling multiple access to those rights of way. Or, in other words, fibre isn't a natural monopoly when you make it easy for newcomers to lay fiber without having to ask incumbents. Anyway, it prompts me to note something else small but important.

Many assume that PSTN interconnect is vital. But for youth, it won't be such a big deal. The IM generation is seceeding from the PSTN. Yes, cell phones are an extension of the distribution model of the PSTN, and are essential teen social props. But imagine if TV spectrum was opened and there was a panoply of excellent cheap IP-friendly wireless networks. They'd all be calling each other using Skype and Yahoo! IM. Our family already spends a lot more talk time on Skype than the PSTN. All they need to be able to access the PSTN is a gateway somewhere on the Internet.

The PSTN is the "network for strangers". It's the network you use to call people who aren't on your buddy list. Which means we're probably lacking in terms of buddy list management. It needs to be easier to get on (and off) someone's buddy list, even if only temporarily or as a second-class member with limited view of the corresponent's presence. Why can't my bank's call center get itself onto my buddy list? Why can't they know I'm a Skyper and when I log in to their website there's a "Skype Us!" logo? I'm already Internet banking, so use the Internet to finish the job when I need personal help!

Of course, secession tends to spark conflict. But this one is being done one person at a time. PSTN slaves slip through the underground packet networks heading north to IP land. There's no simple target for incumbents to attack. The PSTN dies a death through a billion service cuts. It doesn't go away. It just becomes increasingly irrelevant.

We've only just begun this journey. When Michael Powell was preaching a VoIP revolution that sweeps away the past, I suspect that most of the audience didn't understand "and that means you!". There's still a world of issues and opportunity beyond replicating an audio stream on IP and making it hop over gateways and network boundaries.

UPDATE: Joe Rinde is doing a great job of listing dozens of features we don't have today and can look forward to. It's been obvious for a long time that we need to move from selling VoIP on price to features. His is the same message as mine, just more eloquently put.

Very naughty

I'm sat in the back of a VON keynote session. In my right ear is an in-ear headset, which means the guy sitting next to me isn't disturbed. Poking out of my laptop bag is a microphone. On my knees is my laptop. And 4000 miles away in Britain is my brother listening in via Skype. I'm using IM to talk to him as I can't physically say anything.

This is evil. This is what VoIP is supposed to be about. Re-invent telephony.

My brother jokes that in future nobody will ever go to lectures, they'll just send a tape recorder. The room will just have a microphone on each chair. And nobody will go to present a talk: they'll just send a tape to be played at the front.

Listening to predictions by the speaker on the death of big iron telco gear, Geddes Jr quotes: "The world of IT is going to take over the world of telecom because it has better APIs.".

How true.

PS There is free WiFi at this conference, they just mis-configured it on the first day to ask for money. Still incompetent.

A ton of VON

It's about half way through the Fall 2004 VON conference. I just couldn't take one more dull-but-worthy keynote, so decided to duck out and have a tete-a-tete with my invisible (and possibly imaginary) audience instead.

It's annoying to pay over $2000 for conference entry, be lavished great food and facilities when you get there, and then be dinged $100 by the conference center for WiFi. Particularly when the conference is all about the connected IP age. So I'm not paying. I walked 15 minutes back to my guesthouse in protest, to suck up some free bandwidth from a generous Mrs NETGEAR somewhere over the road. (The B&B has WiFi of its own, but the owner didn't realise he needs to tell you the WPA access key, and assumed a WiFi card was all his guests needed... doh!)

The restricted connectivity at VON is highlighted by the hordes of people lined up at the breaks to access a row of public access PCs. The demand for connectivity exceeds that for micturition, judging by the length of the queues. And artificial scarcity of access isn't the only frustrating Bellhead legacy at VON. This is an industry in a deep state of denial at the cataclysm facing the legacy smart network model. Sir Terry Mathews gave a speech which really laid things bare. He talked about the death of the (networked) canal industry in the face of orders-of-magnitude improvements offered by railways -- and reminded people that cost improvements of the same magnitude had occured in fiber, silicon and storage in the space of half a decade. Christopher Fine of Goldman Sachs spent 27 minutes on some interestin but well-known stats and projections on VoIP, and then 3 minutes on why most operators and vendors in the room were (by implication) totally ******. That last 3 minutes was the meat. Even FCC chariman Michael Powell engaged in some revolutionary oratory.

Yet I've sat through several talks and seen many booths where the old world is still alive and well, dressed up in ill-fitting new IP clothes. My goodness -- a hockey-stick graph showing an explosive growth of IP PBXs! If this fails to happen, it's sad for the shareholders and employees of those companies. But if it comes true, it's an even sadder outlook for the end-to-end Internet and user-driven innovation. I don't want to have to ask an IT department or network operator every time I want to try a new device, service or feature. No, I demand not to have to ask. I'm a responsible adult (don't laugh). Let me provision my own stuff. Let the vision of distributed computing blossom.

There's just too many people at the conference with session border gateways, deep packet inspection probes, proxies, IP PBXs and IP Centrex. It gives me the shivvers. These aren't evil technologies in and of themselves. A web server is a smart element attached to a dumb network, and is pretty harmless. It's just when the only route between you and I is via one of these intermediaries, we're busy re-creating the failed smart network of the PSTN, just with a new and richer feature set. These aren't network services, they're network disservices. Somehow we need to get a better grip on the good stuff -- targetted firewalling, virtual network segmentation, intrusion detection -- and ditch the application-specific stuff being rewired into the network.

Many speakers have talked about the "IM generation", kids who have grown up with IM as their primary means of keeping in touch with their friends. There have been repeated references also to the blurring between home and work, and the diffusion of the work environment away from the office. A world of geographically dispersed teams, and collaboration between suppliers, customers and partners. Doesn't anyone see the contradiction here with smart networks of gateways and proxies? A closed enterprise messaging system doesn't let me IM with my customer. My personal professional support often comes from other bloggers and former colleagues who are outside the traditional employee domain. Yahoo IM is a business tool, not a kid's toy. But at Sprint I could only access Yahoo IM via a secret and illegal SOCKS proxy someone had installed in the IT department. Doing real work at work is verboten; your job is to be present, not use presence. The only way to get quick advice from your personal knowledge network was to break corporate security rules. This is horribly broken.

You should all harbour a great fear that the mistakes of the past are being repeated, driven by short-sighted network designs and corporate security paranoia. Yet a glowing light in the darkness exists in a small booth in the exhibition hall. Popular Telephony are changing the world, and making most of the other exhibitors obsolete. Their Peerio product is putting a SIP and H323 server into the silicon of phones on people's desks. Married to this is an encrypted, peer-to-peer content management network. Your voicemails, emails, and even directories can be smeared around, in duplicate, at the network edge. No servers, PBXs, or centrexes. Want a telephone network? Available at Costco, $50 a phone, buy in bulk. Oh, and they're quietly hijacking and subverting the PSTN numbering space, too. Just don't say it too loud, in case someone hears.

Maybe they should rename themselves the Popular Telephony Liberation Front. Or was that the People's Telephony Liberation Front? Never mind. This is peer-to-peer as it was meant to be. It's horribly subversive. I'm in love.

So if you're putting your pension savings into the VoIP business, tread very, very carefully. This boom going to hurt as bad as the .com explosion. One way or another, there's going to be a lot of roadkill, and a lot of get-rich-quick money from inflated promises.

There have been some other nuggets at VON I'd like to share with you. Sorry for the disconnected random notes. Must try harder.

Firstly, there's no doubt that there's a lot of excitement about. VoIP is either making lots of money, or destroying other people's ability to do rent seeking and make money. There are a lot of suits at what could have been a rather geeky conference.

At the Telecom Policy Summit, someone astutely commented that a key aspect of IP was not just decoupling of the applications from the service. It is also the decoupling of the financial relationship between the connectivity and service provider. This difference is subtle but vital. In principle, we could re-impose some form of access charge on services in order to fund socially or economically desirable connectivity. This separation of connectivity and service at the financial layer, above the application layer, is a vital battleground. How far should end-to-end be pushed? At an extreme, you could even try to have separate regulators for connectivity and services, parting them at the polticial sphere.

I liked the parallels drawn between telecom and car dealers. Both have intermodal competition (e.g. cars vs. public transport; cable vs. copper vs. wireless). Both have lots of lobbyists. Both have protected wholesale markets. Cars and connectivity are necessities. There is regulation of car dealers, and bad ones get punished. The regulation is post facto, nor a priori: wait until you have a problem, but then act realy quicly. Please, no posthumous validation of complaints aginst abuse by incumbents. The government doesn't set prices for cars. There are some minimum social requirements, such as emissions and fuel economy. These are public goods, and ones which the owner of the vehicle derives little benefit for that isn't reflected in the price. But there is no minimum rate of accelleration, or allowable number of defects. It's vital to avoid creating "regulatory goods": features that are acquired by petitioning public bodies rather than through market pricing.

There's a lot of confusion at the conference when people say "VoIP". You have to work hard to understand whether they mean slot-in PSTN replacement using phone numbers, or just any real-time duplex audio stream over IP. One of the best things the regulators could do is introduce some new terms and people will, by necessity, start to use them. It was noted that adopting different terminology for economic and social regulation has sometimes worked well. This avoids "regulatory bleed", where case law refines the definition in one area, but this changes the use in unrelated spheres. For example, a court decision saying new service X that bundles aspects of connectivity and application isn't subject to state sales tax shouldn't automatically kick it out of the scope of wiretap as a by-product.

A big fight is brewing over whether to attach fees and regulation to phone numbers, or to connectivity. If the former, it may kill off the PSTN, or drive people to out-of-territory numbers with lower fees and burdens. If the latter, it contradicts the aim of promoting broadband adoption by keeping it cheap and cheerful.

Being a US-centric conference, there has been a lot of talk about a re-write of the infamous 1996 Telecom Act. It would be nice if that discussion wsa framed differently. Instead of looking backwards at the 1996 act and fixing the problems, why not look forwards? Adopt an explicit policy of rapid decomissioning of the PSTN. Allow the telephone numbers to stand as an independent resource, just like DNS. Why not have a desirable destinatin in mind, instead of yanking the rudder one way and 'tother?

A take-away of the conference is that the US is, sadly, relatively backwards when it comes to broadband and IP. Yes, a lot of the technology, capital and excitement is American. But the political system and legacy of the Bell system hang over everything as a dark cloud. Much as the decline of the railways and naval power heralded the end of the British Empire, the days of American communications technology hegemony may be numbered too. Which is sad, because at heart I like vibrant American capitalism and innovation.

Several people are concerned that the incumbent operators will discriminate against VoIP service providers, for example via port blocking or inducing jitter to competing VoIP services. Chairman Powell explicitly said that the "four freedoms" speech (which he had re-iterated today) was a shot across the bows of the industry. Discriminate, and the regulations will arrive. Others had previously noted that there were few if any examples of this.

My take is that the incumbents are far, far, smarter than this. They won't add blocking to existing services. They'll just create a DSL-lite introductory product which lets you roam within the walled garden. In the most restrictive case, they'll sell DSL tied to VoIP alone (without Internet access), and try to bypass fees and regulations. The price of full-access DSL can then be jacked up. The telcos will get kickbacks from the application service providers and partners based on incremental revenue.

Jurisdiction is going to be a massive regulatory food fight. The FCC is minded to grab control for itself and only delegate certain implementation details to the US states. But the elephant in the corner is the global nature of the Internet. Services can flee the US entirely. Which is why the telcos need to presenve the geographically-based numbering system, as it support a geographically-based lobbying footprint.

I saw some regulatory capture in action, with a representative of the New York public utilities comission. When large financial institutions in Manhattan wanted to achieve multi-path redundancy, he ruled to force Verizon to hand over network maps (for a regulated price). Much better would have been to question why competitors find it hard to gain rights of way and access to build competing redundant lines.

Getting bored with regulation discussions, I slipped into the enticingly named session "It's the Smart Device on the Stupid Network", part of the 2004 Landline Replacement Summit. I watched a crigeworthy presentation from Motorola's CTO of broadband demanding the re-imposition of centralised QoS and circuits. Hmm, wonder why they've been struggling and laying people off. Somehow Moto things that the network should not be stupid, but should still be application agnostic. I'll write an article on what QoS is good for another day, but this is just a design for a car with square wheels. Hopelessly wrong.

Debby Hindus of Rapport Inc. actually said something new, which was out of character for VON. She highlighted the possibility of denial of service attacks being targetted against individual user devices. This would be a new phenomenon. She also predicted the need for smart devices with lots of processing power to, for example, screen out background domestic noise when working from home and calling into work. Similar image processing would be applied to video calls; the customer wouldn't see the unmade bed behind you. Search functionality is progressing from server to desktop, but the next step is into mobile devices. Some day your "always on" will come from a continual murmur in your ear from your digital assitant.

Kevin Kealy of AT&T's security labs gave an attention-grabbing presentation of VoIP security. Summary: there's isn't a whole lot of it. One great quote was that 70% of the people calling an automated 411 directory enquiries system never realised they were talking to a computer, the voice synthesis and recognition is so good.

I keep hearing the term "VoIP subscribers". This is nuts if you look at Skype. We need to track "VoIP desubscribers". People who no longer have to pay a service provider for the trivial service of VoIP.

Last, but not least, was Microsoft. No surprise, but they announced the beta release of their new desktop communications client. Think of this as the "real time" equivalent of Outlook, incorporating presence, IM, and telephony. It looked good. But note the wording: Microsoft Office Real-Time Communications Server. They know they have two monopolies: Windows and Office. Everything must be tied into one of these two. So Word and Excel are now presence and communications enabled. Microsoft never crow about killing the competition openly any more. But for a lot of people in that room, they were announcing their businesses had been scheduled for execution. The Beast has spoken. They want to own the corporate desktop and office telephony. Be scared.

Playing forfeits

I take it all back. Icelandair is oficially fab. Yeah, the flight attendant on the first flight out was half troll and half beserker. "Sit there, do as I say, weakling customer!" But when I checked in at Glasgow for the ride back to the USA, they asked if I wanted a trip to the Blue Lagoon while in transit! Free transport and entrance. I mean, do dogs have fleas? You bet! So I found myself, mid-Atlantic, bobbing about in a nice geothermally heathed swimming pool in the sunshine. Now that's the way to travel. Exceed the customer's expectations.

Frustrating the customer's expectations were Vodafone, whose allegedly free WiFi access points at the Blue Lagoon simply didn't allow any connections. I don't have a cell phone, calling card or Icelandic coins to call with. Just a laptop and VoIP. Money can't buy back the chance to call your wife from Iceland and tell her you love her.

So a swimming pool from Icelandair and a slimy poo to Vodafone. Anyway, moving on to business, on the flight over I was staring out of the window and pondering...

One of the big governance problems of the Internet today is misuse of copyright legislation. Laws like the DMCA and EUCD are easily abused. Content is too easily torn down by complainants who may have either no rights over the content or attempt to undermine fair use of that content.

This is because of an asymmetry between content owners and members of the public. Content owners are often large and rich. The more famous the work, and therefore quotable and parodied, the richer the owner is likely to be. You and I, dear reader, aren't so lucky to have a bevvy of lawyers to hand. The power of a takedown notice, and the cataclysmic effects it can have on free speech, need to be balanced by more appropriate responsibilities.

So here's my proposal. We take a leaf out of demanation law. In the UK (IANAL, YMMV) the defendant in a libel case can offer to settle. If the plaintiff refuses to settle, and the final award of damages is less than the offer, the plaintiff is totally and utterly buggered. So there's an incentive to settle early and often for reasonable amounts. It's dangerous to go fishing for a large libel award.

Applying this sort of reasoning to copyright, we need some changes because the problem isn't the same. The subject of libel is looking for large damages, whereas the initiator of a DMCA-style takedown notice is simply looking for the erasure of part of a web server's disk drive. The public interest argument is different too. Libel law has a narrower effect on free speech (only affecting comments on individuals), whereas copyright affects a much broader set of expressive means. We need a game where the stakes are finely balanced.

Here's what I would do. In asserting copyright abuse, the complainant would have to put their balls in the guillotine for a while. The rights to the work against which the abuse is being alleged would have to be put at risk. The complainant would have to lodge in their complaint not only the exact details of the work being infringed, but also the market value of the rights to that work.

If they win their takedown notice, end of story.

If they don't, things get interesting. Firstly, say they were found to have valid claim to the copyright to the work, but the use was non-infringing. Then the defendant would have a choice. They could either receive as damages the cash value of the work as declared by the plaintiff. Or they could choose to purchase the right for the declared value.

If the plaintiff pitches the cash value too low, then they potentially lose a valuable revenue stream from the work. Too high, and they're going to have to remortgage the house to pay you in hard cash.

If they don't own the rights to the work (e.g. because it's public domain), then you get the cash. Any future complaint they file must be for at least double that declared cash value. So you can file a spurious complaing for a dollar's stake, but soon you're going to find it an expensive hobby.

Lawyers paid via contingency fees would no doubt be attracted to this scheme. Initial abuses would be less likely because a community web site illegitimately attacked by a multinational corporation could simply call in their friends in the law. Without big payouts for abuse, access to the law becomes available only to the rich and powerful.

Instead of transfer of the copyright to the winning defendant, you might instead assign the revenue and lciensing rights. If the plaintiff stops selling the work to spite you, tough. Their prerogative and loss nonetheless.

A variant on the scheme would be for the defendant, and not the plaintiff, to declare a cash value for the work. The losing plaintiff would then get the choice of handing over the cash or the rights to the work. Similar logic as before applies to keeping the pricing honest.

Yet another variant would be that the work automatically becomes public domain if the takedown fails. No cash value declarations needed.

A limitation of these schemes is it isn't easily transferable from copyright to trademark infringement. Whilst the revenue streams from Snow White and the Seven Dwarves might fund a comfortable retirement, the rights to Mickey Mouse are less fungible. The whole point of a trademark is that it associates a symbolic form with a branding entity in the public's mind. Once outside the Disney domain, transfer of Mickey to another private owner is likely to destroy the value of the trademark and have collateral damage on other trademarks owned by Disney. Goofy and Mickey have to stick together. Copyrights are regularly traded and re-assigned, whereas trademarks are much less portable. I don't have an answer to this one.

And in the absence of a witty recursive sign-off, that's all from me for today folks. Time for an afternoon walk around Boston and a spot of lunch.

OPINION://Number of the beast

Last week I wrote a brief article noting how Korea is issuing phone numbers for VoIP applications, but encumbered with a number of restrictions on their use. David Isenberg has posted a comment that raises a billion dollar question: from a long-term view, is it a good thing that many VoIP applications are trying to re-use the numbering space of the PSTN? (I hope that's a correct precis of your concern, David!)

This issue is so important I've decided to take a break from the endless task of unpacking our sea container and write a separate post about it.

So, is there a case to be made for re-using the PSTN address space for VoIP? I think the answer is a qualified "yes". Nonetheless, a little voice inside me wants a cleaner break from the past.

Clearly VoIP technically doesn't need PSTN phone numbers, as any user of Skype or Yahoo! chat will attest. But there's a problem with trying to break free. VoIP doesn't scale well socially and politically without a strong identity infrastructure.

The previous tying of the application identifier (phone numbers) to the connectivity provides a rough-and-ready means of providing accountability. For example, caller ID helps prevent and trace abuse. I wrote more about this in the summer. In brief, you can receive your teleconnectivity in four ways:

1. From a central authority to a fixed line into your own home.
2. From a central authority to a mobile device which is post-paid and for which you have passed a credit check.

3. From a central authority to a mobile device that is pre-paid, anonymous, but tied to one provisioned device.

4. From somewhere at the network edge (e.g. ad-hoc networks, meshes, open WiFi access points) that is anonymous and not tied to any device.

In the first three cases, there is what you might call "identity collateral" built into the supply of the connection. That means if you misbehave while using your connection, there's the possibility of retribution by society. Case #1: Your end point has a unique network address. Someone can come to your house and arrest you; Case #2: You are provisioned to the network, and that's tied to an account name. Someone knows who you are and can tap into your government IDs or bank account details and track you; Case #3, someone can de-provision your pre-paid device as punishment. Varying degrees of suffering, but a much greater degree that, say, having to set up yet another Hotmail account to run your scams and access MSN Messenger voice chat.

Think of it this way. If you dial 911, 999 or 112 and there isn't an emergency, you should be punished. You've risked lives for no purpose. Any VoIP system that can't hold you accountable is useless for this service and cannot be allowed to connect to emergency services. Likewise for a thousand other abusable services, from the local pizza delivery outfit to dating chat lines, in varying degrees.

Now, the following is so vital, I'm going to put it in large red bold. It's the only large red bold ever posted on Telepocalypse, and I promise not to do it again unless something even larger, redder and bolder is being said. Which isn't likely.

What's the core purpose of a telco? A telco joins the physical world to the virtual world.

They do this in two ways. Firstly, they map physical points on the globe to network destination addresses (e.g. other people's IP addresses or phone numbers). Secondly, and often forgotten, is that they give you a virtual address of your own and tie it to a physical place, person or device. [Sorry, I'll go easy on the red and bold from now on.]

My hunch is that the future of end-user telecom is in the latter function, not the former. The retailing of bit passing is a loss-leader to other functions that map the physical to the virtual. Indeed, much of the cost of running a telco comes from the care and sales functions that are aligned with creating and maintaining identities. The cost of building networks is dropping fast, while the cost of managing the customer relationship is not.

Consider a typical large mobile carrier with 20m customers. Say the network costs about $15bn to build, and generates a similar amount of revenue a year. The network asset turns over once a year. The "cost per gross add", i.e. the cost of customer acquisition, is about $300-400 per customer, making a total of $6-8bn. This customer data itself generates less than $100m of revenue from caller name exchange, syndication into directories, etc. So the data asset turns
over (order of magnitude) once a century. Hmmm. I know where I'd be putting my energy in growing my telco: expanding the number of transactions the data assets can take part in.

Incidentally, the arrival of m-commerce and proximity payments is a natural extension of this model. Make that #3 on the list -- linking a device to a payment instrument. I believe telcos eventually will be very successful at this function.

Obviously, telcos are not the only ones in a position to provide digital identities backed by physical collateral. You can see employers, banks, goverments, etc. doing this too. But telcos are uniquely well positioned to do so for the base layer of general-purpose public identities. This is because the overwhelming majority of Internet connections are of types 1-3 above, and the telco automatically is custodian of your identity collateral. Furthermore, the phone number is already the de-facto standard identifier, even for many "data only" devices.

Without PSTN numbers, VoIP systems tend towards "calling cliques": buddy lists of people whom you are happy to receive calls from. The social and regulatory infrastructure of PSTN numbers makes that scale to the level of the whole of society. David and Martin being happy to publish their Skype addresses isn't the same as everyone being happy to do so.

I've an amusing T-shirt from a recent conference which adds "layer 8" and "layer 9" to the OSI stack, and labels them "financial" and "political". We're talking scalability at "layer 9" (political), not layers 1-7.

In a nutshell, the case for retaining PSTN numbering for VoIP rests on the ubiquitous existence of identity collateral in connectivity provision, and the pervasive use of PSTN numbering to represent that identity relationship. Whilst this is an accidental by-product of tightly coupled connectivity and service, and new identity system that fails to embody identity collateral is inferior and will not succeed in mass adoption.

The qualifiers? Firstly, the system of PSTN numbers, as we have both noted, may break down within a decade or less. Caller ID spoofing, number exhaust, geographical toll bypass, VoIP spam, regulatory by-pass to avoid associated universal service fees etc., the list goes on. Secondly, you can argue that the need for a single, monolithic real-time voice application called POTS has gone away, and there isn't a need for a monolithic VoIP replacement with a single identity infrastructure. A balkanized voice world of varying degrees of central control and interoperability is the primodial soup of innovation we have to live with. Passing chaos is the price of change.

PSTN numbers also come with some unwelcome legacy. As my brother dryly put it today:

I see Skype switched your id across even though you not only moved house but also IP address provider.

So number portability is a crap imitation of what we need. Also the geo-centric nature of numbers and the consequent outrageous charges for crossing the boundaries is a barrier to keeping the system alive.

We could also see other systems emerging, where the focus on PSTN numbers is a distraction. For instance, let's look at this entry in the RIPE database for IP addresses in Europe:

% This is the RIPE Whois secondary server.
% The objects are in RPSL format.
%
% Rights restricted by copyright.
% See http://www.ripe.net/db/copyright.html

inetnum:      82.70.155.144 - 82.70.155.151
netname:      ZEN000021806
descr:        Mr Martin Geddes
descr:        ADSL
country:      GB
[...and so on...]

Because I have a static IP block from my ISP, I'm in the database. But as a basis for an identity infrastructure it's positively paleolithic compared to the PSTN's numbering system. Too sparsely populated, too little asserted about the content, too fragmented governance, too opaque to the everyday user. But I'd love to be involved in fixing those problem. I just suspect it's a generation's work.

Another objection is that new forms of identity collateral may be required for new types of service. For instance, using your eBay login as the identifier and and your reputation score as collateral. Maybe by 2015 you'll need to lodge some eGold in an escrow account to make a phone call and assuage fears you're a VoIP spammer. (Only Warren Buffet will be able to affort to call Bill Gates!)

Make no mistake, identity issuance is fundamentally centralized. Why? Because we issue identities not only to distinguish between individuals, but also to regulate the behaviour of those individuals in a social and economic context. When you give details for a credit checked identity, lying can result in imprisonment. That coercive power is ultimately vested in governments, who in principle have a monopoly on the legitimate exercise of violence to enforce good behaviour. And unless you're an anarchist, government authority is central in execution. (Please, don't write in about the sovereignty of the masses. Please. Not here. Also, for those who believe corporations are usurping the power of government, I suggest you try being imprisoned, have the baliffs retrieve your goods or having your home suffer compulsory purchase before you go much further.)

So, finally back to the Koreans. The two steps forward are having tons of ultramegabroadband, and getting VoIP-PSTN numbering sorted. One step back for assuming the importance of PSTN numbers comes from mandatory technical rather than social characteristics.

Now, I wonder what's in that mysterious unpacked box in the corner of the study...

UPDATE: Here's another way of putting it, maybe a little more succinctly and coherently. Connectivity provision creates de-facto identities via assigned network addresses tied to account numbers. These are backed by identity collateral, by the nature of bridging the physical and virtual worlds. The application layer is wholly virtual. VoIP is just a software application. But to socially scale it also needs to access some form of identity infrastructure that ulimately links to physical entities. The most natural one to adopt is the one we already have in connectivity provision. And the telephone number is an acceptable identifier already in global use by normal people. The issuance, use and governance of IP addresses is even more screwed up. The question is how to extract the best of PSTN numbering and translate that into an IP world. If it proves impossible to retain the (identity collateral-backed) phone number baby while ditching the (subsidy-ridden anti-competitive telco) bathwater, then we'll have to commit infanticide. But not yet.

UPDATE: Maybe this is the central undecided issue of VoIP: incremental vs. revolutionary change. I'll be bearing this in mind at VON next week and will try to gauge the wind direction better.

Oprah's iPod

Following a link on a BBC News diabetes story today turns up the following interesting facts:

Diabetes websites too complicated

Online health advice for people with diabetes is often too complex to understand, analysis suggests.

... He found people would need a reading ability of an educated 11 to 17-year-old to understand the sites.

However, he said the average reading age of people in the UK was equivalent to an educated nine-year-old.

The Register documents it rather less subtly:

People still thick despite internet

The lamentable truth about the mind-expanding claims for the internet has finally been revealed - people do not use the bottomless well of knowledge to advance themselves, preferring instead to indulge in casual surfing related to hobbies and music.

This is the conclusion of a Cardiff University study which shows that the concept of online "lifelong learning" has largely fallen on deaf ears. The number of adults in education has hardly risen since the advent of the net and less than half of people over 18 make use of the wibbly wobby web - despite its widespread deployment in public buildings.

... Sadly, web utopians will just have to accept that the internet is ultimately no different to that previously-hailed great leveller: the printed word. As the old saying goes - you can lead a horse to the complete works of Shakespeare but you can't make it read.

One of the limits of current broadband penetration is the use of PCs in the home. Broadband is associated in the minds of the public with PC use. But just as Andrew Odlyzko has observed, when electricity was introduced, people would buy a motor which could then be attached to various devices. Nowadays there are motors in dozens of applicances in a typical home. The same will happen to connected applicances.

Expect to see two parallel trends emerge. One, also widely noted, is the emergence of the application-specific device. The iPod or TiVo are the standard examples. It does one thing extremely well. I have several highly educated friends who don't have PCs at home, or don't have them plugged into the Internet. There is a market to sell people broadband on the basis that a whole raft of information-centric applicances will hook in easily. Today broadband is marketed on features of no meaning or use to the end customer. Megabits? Asymmetric whatsits? Fugeddaboudit. Time to sell the benefits, not the features.

WiFi is likely to be the delivery mechanism for the "last plasterboard" scale. But WiFi today is horrendous as a consumer offering. How can any mass market technology succeed when you're expected to make decisions between WEP and WPA encryption, and dream up 64-digit hexadecimal security keys? If I was running Apple, I'd be looking to out-flank Microsoft by moving into home automation and bypass the PC roadblock. The Apple co-branded heating thermostat, Apple door security system, Apple smart clocks and pervasive computing devices. All seamlessly interoperating.

The second trend is a move away from textual displays. VoIP enables everyone to become their own radio talk-show host. Personalised call-in programs will be the RSS feed for the masses. Edward Tufte notwithstanding, for the semi-literate plebfolk, talk is how they get their daily data fix. Voice may be a terrible way of passing information (and a great way of interacting) compared to the written word, but that's no consolation if you can't read well. For the textual and graphic interfaces that remain, the most powerful brands may turn out to be those of tabloid newspapers. They understand how to interact with sub-literate but otherwise intelligent adults. So I hope Apple and Sony don't mind burying their upmarket positioning when they co-market with the Sun and Daily Mirror.

One irony of this shift is that Microsoft are spending billions of R&D dollars on their next-generation Windows, codenamed Longhorn. Having been the greatest creator of corporate wealth the world has ever seen, they are now engaged in a progam of capital destruction and waste of incomparable proportions. If Microsoft had already handed all its cash pile back to shareholders, do you thing they would find it easy to raise the money to build Longhorn? I suspect not. Longhorn does nothing to deepen the spread of connectivity to the masses. It will be an expensive disappointment.

The PC era is over. Time to reallocate your tech stock portfolio back to the consumer electronics companies.

OPINION://Dirty old town

The area in which I live in Edinburgh is at the edge of what's called the New Town. The distinguishing feature of New Town -- certainly compared to suburban Overland Park, Kansas from which we've fled -- is its considerable oldness. The trees surrounding our apartment complex were originally planted as part of an arboretum, which relocated about 200 years ago further outside the city centre. Yet the pavements beside the cobbled streets have ubiquitous "CATV" hole covers, and coax sneaks up the building facades.

If we can get cable TV, we should be able to get optical fibre. But I'm not holding my breath. A 1Mbit/sec DSL connection is all I'm getting for now.

One reason for such slow change is that the nature of the immovable incumbents is changing. People assume that the enemy of fibre is the old telco running copper lines into your home. But it isn't necessarily so. In fact, there's a whole digital ecosystem based around the ekeing out of artificial scarcity.

Take Microsoft's foray into DVD video formats. Clearly, they are trying to capture the interface between machine and digital instructions to operate that machine (again). No doubt they have some splendid wavelet-cum-fractal patented recipe incorporated into the MPEG standards. But if everyone has gigabit fibre to their homes, I don't need to care about your stinkin' compression patents. If I have to choose some public domain and inefficient alternative, or no compression at all, who cares?

Similarly we're seeing an end to storage scarcity, which opens up new services like Google's GMail. This threatens Microsoft's Outlook franchise.

Then the scarcity of CPU cycles in mobile devices starts to erode. Your ASIC that sipped a few milliwatts of power gets replaced with a general-purpose CPU that can run a multitude of VoIP applications. What was once co-dependent on a bunch of smart base stations and switches at the carrier gets liberated.

So there's a whole digital economy based on the rationing of scarce resources that are rapidly ceasing to suffer from contention. Before you berate and bash a telco, consider who else has an interest in keeping you underfed and hungry for more connectivity, storage and CPU cycles. There are many out there who stand to lose their cosy lifestyle.

In the face of such overwhelming inertia, it's no wonder few entrepreneurs risk breaking the copper legacy. What we need is public policy that creates new public land and ends the digital enclosures.

There's a slightly stretched historical analogy to this. One reason, I believe, for the success of the USA is the historical difficulty of imposing tyrrany and sustaining repression. The abundance of land has encouraged the fleeing from opression and poverty: blacks from south to north and whites from east to west. Likewise (and an intellectual hyperspace jump later) for virtual real estate. Once you have a fiber connection to your home, you effectively have your own digital homestead. You no longer need to rely on someone else to host your pictures, archive your email or give you permission to put up a web page.

There's a catch, however. New Town in Edinburgh has gorgeous architecture, with buildings complementing each other in great circles, broad streets and quaint back-alleys. It took planning and central control to get there. An Act of Parliament, no less. It was created anew because the Old Town had become a squalid and overcrowded cess-pit. The days of malnourished serfs pandering to the needs of a monarch were over, and enlightenment had taken hold.

What we now need is the same radical insistence on building for the future. Any fixed telecom facility that traverses public property must be fiber-based. No exceptions. It should be impossible to obtain permission to dig up a road to access a copper cable except to replace it with glass. It isn't a question of money and budgets, but one of permissions and property rights.

Of course, the Old Town didn't go away. The route up to the castle is lined with shops and tourists. The area is clean and pleasant. But the lifeblood of the city has moved on. Time to wave goodbye to the feudal communications system, too.

Three is the magic number

Via SIPthat I notice that the Koreans are issuing phone numbers for IP telephony from a reserved range. What makes this interesting is that they are only being issued to operators whose calls meet certain minimum quality criteria.

Does anyone else see the irony of this? Firstly we've seen an explosion of cellular usage despite the grossly inferior voice quality. Nobody blinked at giving out phone numbers to mobile phones. We now even give out numbers to data cards merely so they can be provisioned on networks where the phone number is a required identifier -- even if they can't ever make or take a call. Expats around the world happily make free VoIP calls of questionable quality to keep in touch.

To me it's a classic way of excluding competitors, particularly P2P networks where there is no operator to certify. It also descends into meaninglessness when you consider that the key determinan of voice quality is the connection's bandwith, jitter, latency, and application contention, none of which are under the control of the service operator.

I look forward to this Korean innovation being copied by incumbents around the world. Yo Korea! Leading us on into the broadband revolution... two steps forward and one step back.

Nothing really matters

It's quite a surprise, but the number zero -- the abstract concept of "yes, we have no bananas" -- had to be invented. What might be obvious to you and me, along with round wheels and stupid networks, is actually the result of millennia of cultural and scientific development. The Romans struggled to develop their mathematical skills without positional digits and the decimal system, for which zero was a prerequisite. Forget calculus; arithmetic was enough of a challenge. Funnily enough, "zero" had been invented (discovered?) by by the Sumerians yonks before the Romans arrived on the scene. But hey, what have the Sumerians ever done for us? It took another 20 generations after the fall of Rome before zero made its way into Western culture via India and the Arabs.

I've been reading an interesting book, Who Really Matters? by Art Kleiner. It has a simple theme: what are organisations for, and who directs their resources? But the genius inside is the business equivalent of inventing the number zero. The purpose of an organisation isn't to service it's shareholders, customers, employees, beneficiaries or suppliers. It is to service a Core Group of people who may be drawn from any or all of these.

Well, "duh!", you say. None of us truly believed that executive salaries were allocated in the best interests of shareholders, customers where #1 or the front-line employees were a corporation's prime asset. But at last someone has stated the intuitive, and given a name the those who matter: the Core Group.

At first read, you might take this to be circular reasoning. Who really matters? The people who matter, of course! But it isn't so. Who matters are those whose perceived needs are taken into consideration when decisions are made at even the lowest levels in the organization. And that's it. Forget the formal structure and written statements of stakeholding. An informal social structure is what underpins the true power and legitimacy of those who matter. These Core Group members hold "equity", i.e. other people believe their needs have to be taken into account and met.

This has some important consequences for anyone planning (or in my case, failing) to be a change agent in a major corporation such as a telco. Using this book as a lens, I have a much clearer view of why the skunkworks I was part of did not meet its goals, and how I would approach this differently in future.

Mr Kleiner gives some useful advice at the ends of the chapters on how to identify core groups, interact with them and influence them. Thus the book contains a useful mix of analysis of the Core Group phenomenon, and synthesis of personal procedures and public policies to deal with them.

Overall, I found this an engaging and revealing read. If it has any weakness, it is that it requires further refinement. To elevate itself beyond management pseudo-science it needs to more firmly state the objective criteria for Core Group membership, to make testable hypotheses and predictions. Now that would really matter.

UPDATE: This all offers an interesting twist on principal-agent theory, one of the great innovations of 20th century economics. Grossly and irresponsibly simplifying, this is about how to incentivise other people (agents) to do stuff on your behalf (where you are the principal) while keeping the costs of policing them in check. Normally we would consider things like "how to make the board accountable to shareholder", "how to make the executives accountable to the board", or "how to make employees accountable to the company". But if the principal that matters is the Core Group, and everyone else is an agent, then we've got a new pattern. So I'd like to see the mathematical rigor of principal-agent theory applied to Core Group corporate sociology. No doubt a PhD in Economics ready for someone's taking...

A friend in need

Some of you will have seen yesterday's announcement that Tony Blair has a dicky ticker. One of the BBC correspondents had the same condition, and recounts his ailments thus:

I was woken up in the middle of the night by a racing heartbeat, sweating heavily and feeling slightly faint. [...] My symptoms had stabilised, but as I read about shooting pains in the arm and shortness of breath I started feeling worse again. My heart was racing at about 200 beats per minute, and I was definitely feeling faint.

Of interest is the following vignette:

Bemused and slightly shaken, I went to my computer and immediately asked Google what were the symptoms of a heart attack.

(This is all very British, not wanting to disturb anyone by calling an ambulance. By now, any self-respecting American would have called two ambulances, a cardiologist, their insurance comany and a lawyer. And probably put up a web site to tell their fiends and solicit sympathy. But I digress, enough of stereotypes...)

Clearly something pretty profound has happened here. The Internet has become an emergency response service without being invited.

Any human disaster can be defined along two axes: the scope, and the severity. Putting my consultant hat on and whipping together the inevitable 2×2 matrix, it looks something like this.

Now let's look at the patterns of communication that occur behind each of these types of disruption.

Suppose you have a minor ailment, miss a plane, break something at home. It's up to you to do something about it. You go to the doctor, book a new flight, order a new plate. The type of communication is a one-to-one and is initiated by the user. The system is totally decentralised. There is no grand authority to call that says "if you have a mishap, call 0800-4BIGBRO and we'll tell you what to do next". Although it's probably a good basis for an amusing sci-fi book or movie.

When a mass inconvenience occurs, we see a different pattern. Consider a blocked motorway, an ailment sweeping the country, a train strike. You generally learn about these through broadcast media (even if relayed through interactive means such as the web). You tailor your response accordingly. So the information dissemination is centralised, and the reaction is decentralised.

The familiar emergency response system has a different pattern again. When you call for an ambulance, you call a central switchboard and are dispatched a centrally-controlled emergency vehicle. The conversation is one-on-one. (Don't confuse geographic dispersion with decentralised control. Not the same thing.)

But as we scale this scenario, we end up in the four horsemen of the telepolcalypse upper-right quadrant. Big disasters don't work well today, if you'll excuse the turn of phrase. One traffic accident or plume of smoke can cause a torrent of identical calls from mobile phone users, gumming the system up. If you have a chemical leak or approaching hurricane, you need a different soft of pattern. You need to target populations of people, instruct them, and receive feedback on what's happening on the ground. Chaos and confusion may be natural consequences of the unexpected, but are not inevitably so.

An assumption in many discussions about VoIP is that emergency service calling must be transferred over from the PSTN. You have to ask yourself, is this the right thing to do? If people are turning to Google when they have symptoms they believe could presage a heart-attack, something's up. People are assembling their own ad-hoc response systems. IP services and VoIP give you flexibility to do something new.

Will anyone take on the challenge of a 21st century public safety system that takes account of this new communications reality? How do we combine the power and flexibility of decentralised mobile and pervasive communications with the power of central intelligence and control?

Perhaps you should ask Tony Blair. After all, he has to worry about all four quadrants these days.

Today's lightbulb

Here's a free idea, and a bit of public prior art to ruin someone else's future patent application.

Some e-commerce these days is based on opaque purchases. This is where you know what you're buying, and the price, but you don't know who from. Examples tend to come from the travel industry, and sites like Priceline, Hotwire and Autoeurope. You know you're flying from Boston to London, and what days, but you don't know the airline or times before you commit yourself. It's very useful for sellers, because they can offload inventory, and the opacity prevents people by-passing the premium that would normally be charged for differentiating features like extra legroom or frequent flyer points.

Here's an alternative version. Say you're looking for a rental car on the Orbitz travel site. Their summary matrix shows the car companies, the cars and prices. You choose the cheapest company at the car size you want, say Hertz. Enter your payment details. And click "buy".

Now here's where things get interesting. Instead of confirming your purchase, you may get a counter-offer from a named competitor, such as Avis. You can't rely on getting such an offer, so you have to be ready to accept the first offer. Avis can keep their advertised prices high, and selectively pick off the customers whom they believe to be price-sensitive. The travel site takes a payment for presenting the counter-offer. Everyone, apart from Hertz, wins.

After all, if markets are conversations, why shouldn't the process of purchasing involve a bit more two-way haggling?