My cell phone is no longer my friend. Not that we've had an argument and fallen out. Or even for that matter that we were particularly intimate in the first place. It's just lost its place in my affection as a result of not caring for me enough.
About half a dozen times in the last week I've received a call out of the blue with the caller ID of "unknown". Each time, silence at the other end. And each time the caller drops the call the moment I speak.
A rogue telemarketer system? A war-dialing trojan PC? Who knows.
So, I called my cell phone company's customer service. I won't name them, because I suspect they're all equally useless, and it is unfair to pick on one. The only two options they give me are to change my number or report the incidents to the police. No ability to trace the caller, block the caller, or even screen out calls without a caller ID. I'm at their mercy, and they don't have any mercy in stock.
I hate it so much. The powerlessness of the whole situation. It's my phone dammit, and I should control who can call me.
So the PSTN deserves to die (with prejudice) for having such a feeble identity system built into it.
Next my cell phone network operator deserves a swift blow to the temples for not doing anything to make the situation better, including simple features that could easily be incorporated in their switching network.
Then my phone manufacturer should be sent to the gallows for not including features to handle inbound calls without a caller ID. My network edge is dumb, not smart. Bad choice.
And finally the bastard who keeps calling me needs to be sent through the gates of Hades for an extended vacation. Curses to all of you!
Just a quick plug for David Isenberg's upcoming WTF forum near New York on 2-4 April. David's an original dump pipe dissident, and it's guaranteed to be a hoot. I'll certainly be there.
I'm sorry, I just can't resist, even if I'm a bit late. I have to chip in my two cents on Eli Noam's recent FT article on infopocalypse, as summarized by Om Malik.
Like most people, I also think Noam's one sandwich short of a picnic on this one. For those unaware of the debate, Noam is arguing that information industries are caught in a deflationary death spiral due a market failure. He claims the failure is inherent in information distribution business models. This point of view has caused a lot of debate, particularly because of Noam's proposed macroeconomic interventionist remedies.
My response is: who cares? He's talking about information as a product, not a service. Lumps of ones and zeros, untailored to the user, mass distributed. That market's vanishing, and good riddance. I mean, you're reading this super quality blog, written at unimagined effort by someone educated at great pain and expense -- and all for free! I'm unable to charge you. It even costs me money to host the blog. Why? My voice is just one of thousands clamoring for your precious attention. There's no shortage of quality opinion on the web. No shortage of quality music on P2P networks. No shortage of news and reportage. Just a shortage of time to track down the interesting stuff. Economics is the study of allocation of scarce resources. No scarcity, no market, no problem.
So, have we run out of information businesses that we can charge for? Have we reached the bottom of the value mine? Well, value in communications essentially comes in two forms:
This is what I was referring to in my previous Telecom Theory article by saying:
In product terms, I believe the purpose of a telco is to maximize the number and value of user communications events, whilst simultaneously protecting the user’s attention from unwanted interruption.
(I would now modify that to include "protecting the user's attention, reputation and assets". No matter. We can also debate which activities are or aren't aligned with the connectivity provider another day. I personally believe a telco can enable the above without defining or controlling end user services, which is a subtlety lost on most people.)
I can't begin to imagine the number of possible but unfulfilled connections still to be made between people, places and things. We're only five minutes into the information age and already Noam's declaring it all a failure and he wants to get off! I'm sorry, but my imagination tells me we've a lot of excitement and economic activity still to come. Just not from selling undifferentiated blobs of bits a-la analog information industries.
So, the real market is for customized information products. Information as a service. Noam is surely aware of the mass customization trend -- so why ignore it? Services can't be replicated by KaZaA. Selling proprietary and dynamic data is a sustainable business model. This is not news to the millions of people in those businesses.
For example, I subscribe to Listen.com's Rhapsody streaming music service. To me, the value largely comes from its ability to recommend new music to me, and show what the most popular things other people are listening to. If all I had was a plain Google-style search box, but access to every song ever recorded, I would paradoxically get far less value. I don't want to schedule my own listening at the individual song level. It's the service, not the content, that matters.
To quote Noam directly:
The basic structural reason for this problem is that information products are characterised by high fixed costs and low marginal costs. They are expensive to produce but cheap to reproduce and distribute, and therefore exhibit strong economies of scale with incentives to an over-supply.
Bzzzt! Nul points. Is there an oversupply of credit rating agencies? Auction sites? Operating systems makers? Metals exchanges? They all look suspiciously like information products to me. Each with a small pool of highly rewarded participants, and large barriers to entry. His argument chokes and dies at the first step, because the definition of "information product" he uses is so narrow as to be useless. He's forgotten about the explosion of interactive possibility that cheap, versatile and pervasive connectivity brings. Staring in disbelief at the price collapse of bypassed industries and technologies is classic telco think! The rest of us are moving past that stage -- do come join the party.
Now, for a laugh, let's dismember the rest of the carcass.
The main result of these factors is that prices for content, network distribution and equipment are collapsing across a broad front.
So, abundance is a problem? Perhaps information and connectivity famine is his ideal state, because the scarcity produces such wonderfully clean market effects.
It seems to have become difficult to charge anything for information products and services. The music industry is unable to maintain prices.
Music distribution has yet to transition to a service model, so bye-bye. As Clayton Christensen would have it, music is bought to scratch a specific itch. The itch is "entertain me" or "stimulate me" or "calm me". In a world of multiplying entertainment possibilities, it's no wonder music is declining in price. Would Sir like a facial spa treatment or another listen to some ambient electronica?
Online publishers cannot charge their readers, except for a few premium providers such as the FT.
Sorry, no banana. Adverts are the staple of publishing, not subscriptions. Paid subscription is a filter to ensure only truly interested readers are attracted, and boost the price of adverts -- since the audience is more tightly defined and relevant. Publishing is about connecting eyeballs to adverts, not lofty goals of journalism and the search for truth. Even for the FT. My apologies for despoiling any reader innocence in these matters.
International phone call prices have dropped, and with internet telephony will move to near-zero.
Errr, ending decades of over-charging for a simple duplex audio stream application. This is a feature, not a bug. Superabundance of cheap connectivity adds to human happiness. Even if it reduced GDP (and it doesn't) then it just suggests he's measuring the wrong thing. GDP is a contributor to well being of the species, not an outcome.
Web advertising prices have collapsed.
Which would be very interesting, except for the minor inconvenience of the facts. They haven't collapsed. Have you tried buying an ad on Google? Interactive, presonalized ads are expensive. Static, old style bag-of-bits ones are cheap and ineffective.
Much of world and national news is provided for free.
Much of the world and national news is boring and irrelevant to me. You'd have to pay me to read it. But I'd pay to have it filtered for me, as a service. I'm running out of time to even read The Economist in full. That's serious!
A lot of software is distributed or acquired gratis.
So what? A lot of software engineer careers are being built on those skills and reputations. And if you've ever tried to buy a support licence for an open source product, you'll quickly discover the difference between "free speech" and "free beer". Service, service, service. Get it?
Academic articles are being distributed online for free.
With open publishing they even pay to get an article into print. The scarcity is attention, not words. Who cares if the market for words collapses if there is no scarcity? The market for attention is working like clockwork.
TV and radio have always been free unless taxed.
Crap TV has always been free. You get what you pay for. I distinctly remember receiving and endless stream of invitation from my local cable company to part with significant lumps of my disposable income. Would Mr Noam please tell us where to get free HBO access?
Even cable TV, at 20,000 programme hours a week, is available to viewers at a cost of a 1/10 of 1 cent per hour.
That's because the real money is in the TiVo, not in the TV. Quantity is not value. Indeed, as noted, excessive choice degrades value.
Newspaper prices barely cover the physical cost of paper and delivery; the content is thrown in for free.
The market is for attention, not for articles. Again. How much of the content of the newspaper is there because someone else wants you to believe it, not because you need to know it to have successful and fulfilling life?
OK, so we've had our fun blowing raspberries at the FT. Where is all this value going to come from?
Firstly, by tying the fixed bag-of-bits-or-atoms asset to the dynamic data, you can still charge for the fixed asset. The blobs of bits can be tied into distribution channels that reward the content creators for providing raw material for services or other products. Just like iTunes is a loss leader for iPods.
Secondly, by cranking up the middlemen roles in matching people to other people, places and things. Hence social networking, IM, syndication, etc. You know all that stuff already. Either get someone to pay for the data or idea you've got, or else grab their attention with loss-leader content and use it to sell them some other capability you have.
Thirdly, look at the underbelly. Look for the social problems the connectivity causes, and find a fix. Prevent the unwanted connections. Just look at the recent headlines...
Firewall VPN sales soar ...
Enterprise security spend to hit $6bn ...
Trojans as spam robots: the evidence ...
UK Watchdog bites mobile spam scammers.
It just goes on and on.
History has been full of "end of the world" predictions, and every time it turns out to be the opposite. Our limited vision of how amazing the future will be blinds us to the possibilities that will open up. The information economy is no different. You too can look forward with confidence, as long as you understand information is a service, not a product.
PS -- the article title is a pun, if somewhat British, old and obscure. Kudos to anyone who cracks the code.
Today's best discussion on Slashdot is about how VoIP providers of PSTN interconnect may be regulated by the FCC to mandate emergency service ("911") capability. There an excellent article at Voxilla from last December on 911 regulation that covers a lot of the same ground.
I'm going to argue the unexpected, and say that 911 service should be a mandatory regulated feature. But with a major twist I'll leave until the end.
911 service is not a normal market product. The service that is summoned is provided through a compulsory levy on local businesses and residents. The fire brigade don't ask for a credit card before turning the hoses onto your burning house. And if the caller is the beneficiary, you may be in no state to engage in any form of usual market interaction. The service always requires urgent response; there is no time to discover how the service should be accessed, or evaluate competing actions. When lives are on the line, you shouldn't be reaching for the yellow pages.
The beneficiary of the service is often either not the caller or owner of the telephone line. There is an externality in choosing not to have 911 service; other people suffer without having a choice of exercising the market mechanism. If my babysitter has a heart attack, and I don't have 911 service, some innocent third party suffers. It's not reasonable for babysitters to have to inquire about the 911 service my telephone is attached to, because the heart attack cannot be reasonably foreseen. If my house burns down and takes the rest of the neighborhood with it, other people suffered for my parsimony. If I see the burglar breaking into the house opposite, but can't call the police, more heartache.
Users are also mobile, and are not realistically able to create a personal comunications safety bubble around themselves. You're dependent on other people having the service for when you get in trouble. You can't escape the externality by being richer and buying better service. If you're lying at the side of the road unconscious, the good samaritan passer-by isn't going to frisk you down to see if you carry the thousand dollar Hypersafe Satphone™. At the point of use, there is a market failure. Market failures require regulation.
Externalities like pollution, health injury, or social damage call for regulation. By definition, they are at best addressed through indirect secondary market mechanisms like emissions trading to rebalance the costs more fairly. It's hard to imagine such a system for a personal service like emergency calls. How do you provide financial incentives for everyone to adopt 911 service via market mechanisms? It just doesn't fly.
The fire brigade or police department will effectively ask for a credit card for a business property that has repeated false alarms. So there is a degree of market encroachment. Abuse of service can be punished through quasi-market mechanisms. The capitalistic instinct to provide truly enhanced emergency response should also be encouraged. Burglar alarms and monitoring systems for the elderly and infirm are existing examples built atop the PSTN. Niche markets should develop their own enhanced response systems. Maybe epileptics will carry special equipment that monitors their wellbeing, calls for help, and uses a loudspeaker to advise passers-by what to do if the owner has a seizure. But society needs a baseline; a reasonable expectation of emergency response wherever you are, catering to the needs of the ordinary person.
So I hope you'll agree with me by now, we need regulation. The only question is to what ends and how to achieve them.
The end-to-end principle tries to separate service from connectivity. Normally, that's a laudable aim. In this one case, however, it's the opposite of what you need. I might take my Vonage service on the road, and plug into a high-speed line in a hotel in Paris. If I have a medical emergency, I want to get a Parisian ambulance to come and take me to a Parisian hostpital. Au secours! I don't want to speak to someone in the emergency response center in Overland Park, Kansas, no matter how nice, upstanding and midwestern they are. I want the service to be associated with the connectivity.
So here is what I propose. We don't break the basic end-to-end nature of the Internet. But we regulate the provision of public Internet connectivity. We insist that basic IP voice 911 service is always included. Maybe we reserve a private IP address (168.9.1.1?) as the default route. It isn't hard. If we're feeling fancy, we include a default route to a WSDL directory to discover additional enhanced emergency services on offer. (Yes, I'd like to connect to a parlez-vous Anglais operator in that Parisian hotel room.)
Whether I'm wardriving around, sat in my hotel room, or feet up at home, the service is always there. We don't yet have massive dynamic meshes of roaming users, and probably won't for another decade or two. The end points of the Internet are always physical and fixed, to a reasonable approximation (i.e. at most one hop away to a cell tower). Paid-for Internet connectivity is always a hop away. Payments require a legal framework for commerce. Those contacts are regulatable. Taxing VoIP service providers and forcing 911 provision just doesn't work, because the service will just move offshore. Connectivity can't move offshore. Call it the revenge of geography. I want the ambulance to arrive at my very physical door, not my extremely virtual IP address. IP addresses and end-to-end are just abstractions. A heart attack isn't.
So, today's headline is "Nokia & IBM announce enterprise level mobile solutions". Note that the partner is Nokia, maker of smart handsets. The partner conspicuously isn't a carrier, deployer of smart networks. The smarts keep moving to the suburbs. Dumb pipes rule.
The marketing departments of cellular carriers are about to hit an interesting problem. Up until now the great unwashed of the general public have not really had to understand the difference between the Internet and the Web. Most people probably have an intuitive feeling that they aren't one and the same (after all, we have two different words), but couldn't really explain if asked.
Why the sudden problem? Well, the first generation of data-enabled handsets only supported WAP and its brethren. So there was no difference to explain. Everything looks webby. IP connectivity and Java followed in the next wave on GPRS, EDGE and CDMA 1xRTT networks. But the Java applications were very limited in their multimedia and networking capabilities -- always-off in an always-on world, weak audio I/O processing, feeble CPU power. The applications themselves tended to be only available from carrier-approved sources. Some phones could also be plugged into your laptop using a special USB cable and used as a wireless modem, at the risk of your service being terminated or getting a sudden and unexpectedly large bill for "abuse of service". All in all, the network demands weren't too great. You could give everyone the run of the Internet house at Web prices without too much revenue leakage.
Now suddenly Bluetooth is pushing the issue. It's really easy to get a Bluetooth phone to act as a modem or hub. It's a common out-of-the-box feature you can hardly expect users to ignore. (Plus the native applications on Symbian and MS Smartphones are getting pretty nifty too.) The low-priced all-you-can-eat data plan targeted at the casual wireless web surfer looks unbelievably cheap to the laptop and data-intensive crowd. And even for native handset applications the potential to disintermediate carrier services like MMS becomes very high. Why pay the carrier toll when you can download a client messaging application that does all you want for free? So the open-network Bluetooth phone has to be priced onto a laptop-like plan. Ouch. And Bluetooth (or a future look-alike technology) will increasingly become standard issue.
This means we're approaching a fork in the road. Carriers either have to start walling in the garden and charging a premium for access outside their verdant plantations (to the extent maybe of cutting off such access entirely to ordinary handsets). Or they have to explain why the better Internet-capable phone doesn't work with the cheaper Web-only plan. Total control enabling uniform pricing but a PR nightmare, or variable pricing depending on carrier control and a confusing market message.
There are times when I'm glad I'm not in marketing.
I'd like to draw your attention to this article penned (or syndicated) by our most esteemed and noble friends at The Register. It relays Gartner's suggestion that employers should pay for employee hotspot access to extend the productive working day to commuting time (presumably by bus or train).
It raises the interesting possibility that public transport becomes a loss leader to attract a captive audience to whom you can market service. The most basic service is connectivity, but context-relevant transactional services as well as advertising can be imagined. What if there was free WiFi on the train to work, but every tenth web page was an advert? Would you use it? What's the attention of eight carriages of highly-paid London commuters worth? Is there even a business in connecting the people in the carriages to each other? Why does business have to wait until you get to the office? Why do fast multiplayer games have to wait for you to unfold your laptop at your desk? The train service itself faces a much stiffer competitive situation (from bus, road, telecommuting and simply moving house/jobs) than the services provided on that train.
As Michel Porter suggested in his Strategy Pure & Simple books, telecom is at heart a distribution business. Since we have yet to reach the technology nirvana of inexhaustible universal connectivity, there is a reasonably gap into which you can pitch your business model. Examples like Truckstop.net, which address the road freight vertical market, may well point towards a brighter near future for connectivity providers.
The trick is to bundle the right connectivity and service offering to the right people. This niche/vertical-driven approach goes against the grain of the PSTN and data services like T1 lines: a broad-brush one-size-fits-all approach with a generic retail and sales force. It challenges telcos to turn themselves into business platforms on top of which other people can build end-user businesses. A business platform is more than an application platform. It also opens up things like access to the forward and reverse logistics chains that supply retail stores. (Reverse logistics is the discipline of taking things like broken handsets from the field and returning them to central location.)
Becoming a virtual network operator needs to become easy. The endless months of negotiation, business development extravaganza dinners and reams of lawyers need to be streamlined away.
Incidentally, and completely off topic, the Truckstop.net example highlights another paradox of the high-speed cellular data industry. By providing a wireless WAN connection, you by definition cover a huge geographic area, and have a large potential market. But at the same time, with today's technology you're forced into a pretty uniform pricing policy. You have to price somewhere that's attractive to the marginal, lowest common denominator user. You can't easily price discriminate between different uage sub-markets and geographies. On the other hand, a network sewn together from smaller pools of wireless connectivity, but tailored to a specific need, enables precision matching pricing to value received. Thus the "best" network may not be the biggest, fastest or cheapest. Just the right one for the job.
To those familiar with Goldratt's Theory of Constraints, Cingular's win over Vodafone will come as no surprise. As Om Malik reminds us, Cingular was constrained from growth by spectrum availability. With rising network usage, its ability to service even its existing customers was coming under strain. A key stage in TOC is elevation of the constraint. Since Cingular are (excuse the pun) technologically constrained in their ability to make more efficient use of spectrum and exploit the constraint, acquision was the only way forward -- at whatever price it took. Which, presumably, Vodafone knew as well, and exacted the full price.
The irony is that Vodafone may not come out of this so well after all. Rationalization will help prop up prices (although competition remains fierce, with many regional players in bigger markets). Spectrum shortage will discourage entrants. And the high price Cingular paid will discourage them from competing on price, because they need the cash flow. This raises returns across the industry. Stock values of all the wireless carriers are rising fast. Verizon becomes even more expensive to buy out, others are less attractive to buy up.
Perhaps Vodafone should have been bold a year ago when the market was still weak, rather than waiting for events to overtake them? Easy to throw stones from the blogside, but if you're a CEO being paid an eight-figure salary, the expectation is your foresight exceeds my hindsight.
As reported at CBS Marketwatch -- "Cingular to acquire AT&T Wireless
Paying $15 a share, U.S. carrier outlasts U.K.'s Vodafone". Looks like Vodafone is the winner here -- weaken a competitor to the tune of $11bn without any real effort. The big question is what does a hungry Vodafone do next?
Saw the word "convergence" pop up again on the ITU web site. I love that word. So soothing -- "service convergence". Looks perfect in your vendor presentation. We're clearly remedying some unwanted divergence. Friendly. Reassuring. Wrong.
Now, what we need is someone to write a little browser plug-in. Every time you go to a web site with more than the threshold level of telecom jargon, it is triggered. And all it does is replace "convergence" with "displacement". Voila! You can see the truth. Legacy service providers cease to be able to force bundling of service with access. Telecom and IT get "converged" into access and even more IT. The IT folks are doing just fine, thanks.
So next time you see a car wreck on the freeway with blood and oil oozing under the twisted metal, don't panic. Have no worries. Let your fear subside. The car has simply converged with the concrete barrier. Converged with fate. Converged into history.
PS -- Bear in mind that the ITU is a global industry association mostly populated by traditional voice carriers. It's run under the auspices of the UN, hardly the foremost proponents of wrenching capitalist change. Loss of control of the application space weakens their political importance -- viz their fights with ICANN. They don't want to be converged with the unemployment queue, either.
Just been reading at The Feature a write-up on how mobile users prefer driving directions and payments over media content from their mobile phones. You're in the metered bit-moving business and DRM is a poison to moving bits. Novel thought (not). And some strategist gets paid for this stuff?
It's been said elsewhere before, but since the world doesn't seem to get it, here's what wireless users want, in descending order of value:
Somehow the marketing departments of most wireless carriers seem to have started from the wrong end. But you knew that already, didn't you?
The wireless business is a strange one. It involves a high degree of network operator control over the whole environment, enabling precision price discrimination through handset lock-downs, myriad price plans and less-than-wholly-open networks. On the other hand there is fearsome competition in most of the world. These opposing forces may make wireless the last bastion of the vertically integrated telco for a long time to come. Or they may make the walls come crumbling down faster than anyone thought possible. Why is this and which way will things fall?
At the heart of the matter is a paradox created by the very nature of mobility. The critical moment is the ability to locate and authenticate to a network. If you're on the move, you need to be constantly covered by the network, despite moving between coverage zones. Only a wireless wide-area network (WWAN) suffices. WWANs cover a lot of space, and overlap in a way fixed networks do not. Hence people get a lot of choice of service provider -- typically six independent networks in the US, and three to five in most other developed countries. So in one way celular is a terrible business to be in because the physics of it encourages too much competition.
But you don't just need any network tower to be in your sight. You need to be constantly provisioned to that network as you move between cells; the mere existence of technically compatible signal coverage isn't enough. Furthermore, you want to provision to the same service anywhere in the country when you travel, always with contiguous coverage. This favors the emergence of an industry structured around a few very large and pervasive wireless networks, rather than a hodge-podge of connectivity sewn together via multiple roaming agreement brokers. Just like with the Disney/Comcast deal, the theory is that when your industry has a limited number of large suppliers a vertical approach can be pulled off. Too much competition and it falls apart; the oligopoly of control isn't sustainable.
So in summary, the very act of moving around demands contiguous provisioning, and by induction a nationwide network. The huge expense required limits the number of entrants, who in turn practice the fine art of gouging the customer for every penny. The existence of early supernormal profit encourages competitors. During network build-out, regional competition is muted and many can enter the market. Eventually everyone builds perfectly overlapping networks which erode that profit.
Eventually the question of mergers comes around to restore profitability, such as the sale of AT&T Wireless. New entrants cannot easily come in and compete directly. This is because incumbents can drop prices low enough to make a new nationwide build-out unattractive. The only entrants are distuptive technologies like OFDM, virtual network operators attacking niche markets, or business models built by leveraging some other existing asset like power lines.
WiFi is really a portable network solution, rather than a truly mobile one. The lack of coverage and doppler tolerance means Mohammed's laptop needs to go to the connectivity mountain and stay there, and not the other way round. So provisioning to the network is a sporadic event. In the absence of contiguous coverage needed for mobility, being asked to pay to provision yourself isn't a big deal. The connectivity aggregators are only offering a moderate convenience advantage over a normal credit card payment. Coke can build a business around convenience of distribution. Physical thirst requires immediate satisfaction. Unless you're a sad case, Internet access in the airport lounge can probably wait the 60 seconds it takes to type in a credit card number.
On top of this, true mobility implies a small form factor device. Is that a laptop in your pocket or did you get a blind man to design your misshapen jacket? Small form factor devices don't have the user interface required to do things that need tons of bandwidth. So there's little edge a WiFi network has over a WWAN when mobile. A dual-mode cellular/WiFi phone is not of interest to a road warrior. WiFi won't undermine the economics of truly mobile cellular access. It is probably fatal to the business case for most laptop-oriented technologies like CDMA's EV-DO.
For wired connectivity, you aren't mobile, so you aren't needing to travel between provisioning zones. Setting up your DSL or cable account is a one-time hassle. Clicking once at the start of the evening that you promise to pay the hotel $10 for the access isn't a big deal. Clicking OK on your phone as you drive down the highway just to keep a call alive is a big deal. Interestingly, a dual-mode WiFi/cellular phone is of interest as a fixed-line displacement product -- but only as a replacement for a cordless phone converged into a cellphone handset, not as a means of extending mobility.
So perhaps what we're really seeing is a re-layering of the industry into three tiers: fixed, portable and mobile. The network may be a dumb pipe in each case, but the mobile end of the spectrum (no pun intended) will remain closed and controlled. The "end-to-end" concept will be irrelevant because users may be have to pay the full value they receive for each service rather than the marginal cost of a packet. Every service will still need permission of the lethargic carrier.
User- or municipal-built networks using WiFi or WiMax will fail to achieve universal provisioning, because it involves an economic co-ordination problem that is not worth solving. The number of people both mobile and under-served by cellular is too small. So they will not cater to the truly mobile user for the forseeable future. Some limited portability market may survive a shake-out -- he who has the greatest usability and convenience for provisioning will win out. The only price premium will come from that convenience of provisioning, and no application or device control will exist.
Stationary access (fixed or wireless) will be cheap -- it is getting there already. The user will not only overcome the provisioning problem, but will even build their own network access if necessary. Such access providers will be totally driven out of the services market, including voice. Telepocalypse ho!
Note that it is the stationary nature of the user that determines the industry structure here, not the use of fiber, copper, fixed wireless, satellite or yodelling.
As a last thought, a study of the airline industry I once saw suggested that fares were little affected when a monopoly route became a duopoly. A third entrant resulted in a significant price drop, and a fourth (or more) created something akin to the nirvana of perfect competition. Thus while mobility will come at a premium, the premium is very sensitive to the number of operators. The switching cost for airline customers is lower than that for cellular customers. So cellular is probably only competitive at 4 or more nationwide networks. Maybe something for the competition authorities to keep in mind when approving consolidation of the cellular carriers...?
My recent Vonage billing history:
| August 28, 2003 | Invoice # 304310 | $61.16 |
| September 28, 2003 | Invoice # 369213 | $112.01 |
| October 28, 2003 | Invoice # 443730 | $64.50 |
| November 28, 2003 | Invoice # 528637 | $94.32 |
| December 28, 2003 | Invoice # 626952 | $47.73 |
| January 28, 2004 | Invoice # 744466 | $20.47 |
Hint: we got both sets of parents broadband and fast PCs for Christmas, and moved to the $15/month plan. Bye-bye international call charges, hello Skype!
The news (i.e. PR placement) about Intel's achievements at optical switching using silicon chips reminds me to pen a few words on two competing models of the end-to-end concept.
The basic premise of "end-to-end" is that you should put as little functionality into the network itself as possible, within the constaint of that network remaining useful for a broad array of general-purpose applications. The usual implementation is to packetize the data and pump it through a set of short-range wireless or copper connections into a long-haul fiber network. The fiber network itself is switched between nodes using some converters and silicon. The silicon switches inspect the packet, and sends it onwards down the next piece of glass or copper in vaguely the right direction.
The throughput of this system is limited by the silicon. Every conversion takes time. The amount of switching you can do is bounded by the size and clock speed you can build silicon chips. Indeed, even the economics of it are ultimately limited by the power consumption required to do the light to electricity conversion and back again. (Ever wondered what happens to all the used ones and zeros? No, not the great bit bucket in the sky, but heat. There's a big push to try reversible computing to eliminate some of this thermal cost.) More immediately, the cost varies directly with the number of packets swicthed.
Reading some of George Gilder's futurology recently, there's another approach. What if the whole fixed network were optically switched? What if instead of an IP address, you used a frequency? In principle, we have the multiplexing technology to do this, and the physics allows it. It's a fascinating possibility, although well beyond my expertise to critique it. Of course, the "last hop" is increasingly likely to be wireless. The growth of cellular, WiFi and Bluetooth indicates an ever-greater density of short-range wireless access points back into the fixed network. So some sort of hybrid would still be needed. But the cost would essentially scale with the number of fibers, not the number of packets. And the fibers have a mind-boggling capacity when freed of the silicon swicthing cost.
Just to dangle the possibility in front of you, here's some statistics I put together. What we're showing is that the speed of light is the speed limit that counts, and we want to get as close as possible to it as we can. I did some pings and traceroutes from here in Kansas City to www.open.gov.uk (the British government's citizen portal). I chose this because it's a long way away (to illustrate the point), plus the UK is all pretty much the same distance from here, what with being a small soggy island floating off the coast of Europe. So I don't need to care exactly where the packet landed. For political reasons, the web site is also guaranteed to be hosted in the UK rather than an anonymous warehouse in San Jose. Furthermore, any content delivery network nodes are probably going to be in the UK too, since the UK government is unlikely to be paying good money to speed up the access times of US citizens.
The reverse DNS lookups of all the intermediate routers indicate a straightforward route, from KC to Chicago (on the Level 3 network), then to New York and London courtesy of AT&T. The one-way trip is approximately 60ms. Some consultation with Mapquest and Google tells me that the trip to New York is about 1300 miles, and another 3500 miles across the Atlantic. The speed of light is about 186,000 miles/second. (Apologies for the mixed SI and non-SI units -- bear with me.) So a beam of light takes 4800/186000 second to get to London from here, or about 25ms. So by speed alone, we've lost 60% of our time in the 20 hops to London. But because the throughput of a TCP/IP link is very sensitive to latency, the real throughput loss is much worse. And we haven't even begun to address the costing issue.
So, the billion dollar question is will the dumb pipe be displaced by an even dumber one? Answers on the back of a PhD thesis, please.
Since it's a bit uncomfortably close to home, I won't say too much about these deals between wireless carriers and instant messaging services. But which is easier for a customer:
* switch network provider (can still IM your friends using IP-based client and old credentials), or
* switch IM provider (can't IM your friends and need new credentials)?
You'll know who will be wearing brown underpants when these contracts come up for re-negotiation. He who controls the customer's identity controls the customer.
There's been a lot of comment passed today about the proposed take-over of Disney by Comcast. Some people claim it's a good thing, others a stupidity to exceed even that of AOL/Time Warner.
I just wanted to pick up on this because at first sight there is a telco angle to this: Comcast are in the pipe business, and Disney are in ye olde worlde media content. Aren't pipe providers trying to climb up the value chain here again? Are there lessons for wireless network operators and landline voice services? Well, not really.
The thing is, Disney have three great creative media assets (the others being the theme parks and valuable harvesting brands like Winnie the Pooh): movies (Miramax), Sports (ESPN) and vanilla TV (ABC). Good movies cost tens to hundreds of millions of dollars to make; exceptions only go to prove the rule. A well-timed deal can corner a significant chunk of the supply to the market. Consumers can't easily bypass the system by rolling their own movies, unlike telephony. Likewise with sports, there is a finite supply of major events, and some good exclusive deals can keep you going for ages. Comcast want to focus on video on demand, using Disney content to kick-start the service. The TV channel is probably a useful avenue for a Google cum TiVo service in future with personalized advertising content -- which again fits in with the DVR/broadband/VoD approach.
So you have to remember that telecom is fundamentally a peer-to-peer communications business with a bit of content bolted on the side (ringtones, mobile news clips etc.). Comcast see themselves in the content delivery business, with a bit of communications tacked on to sweeten the deal. To state this another way, telecom is about creating a unique distribution system for user-generated content, and a cable/media service is about creating a unique distribution system for professional media content. Similar, but not necessarily propelled by the same strategic imperatives.
Don't read too much into this merger yet, beyond the usual secondary effects such as strengthening Microsoft in their fight to tithe every moving data bit in the universe.
James Seng has some useful musings on how telcos fight back against disintermediation. They assert control over the value chain to extract the full value of the service from the customer's wallet.
He identifies some points of control the carrier has over the user. Let's review his list, and start to put together our picture of where we are today in terms of carrier control of the value chain.
Control the end-device
Since the new design is a dumb network, they will control it by controlling the device. They will set "standards" and mandate designs of devices before they allowed it to be connected to their network.
...
Controlling the end-device is a stop-gap measure because after the market saturated, equipment manufactures can only competes on features not standards. They will and seek to break the restriction. And they already have more control then the telcos; They just didn't realize their power in controlling the end-devices.
Well, my first though is that it isn't just a matter of carriers vs. handset manufacturers. There a very important ratchet effect pushing towards openness. If all the carriers in a market start off equally closed, there's a way to differentiate yourself from the competition with zero product development costs: relax in the direction of more open.
What are the handset points of control? My inventory looks like this:
The hidden factor here is the collapse of handset subsidies. You might have thought that these were a bad thing for the carrier, because they add to the cost of subscriber acquisition. But they're really an asset, because the offer of a subsidy comes at the cost of accepting a closed, controlled device. Moore's "law" (really "observation") is reducing the cost of handsets to the point where subsidy is being eliminated. Prices in the mass market are too close to zero for the handset price to be a deciding factor in network selection. I've seen some rock-bottom wholesale handset prices for quite sophisticated camera phones that have shocked me -- I couldn't believe that the retail prices actually represented a positive mark-up. Furthermore, we're seeing impending business model change that brings a low-cost model with zero subsidy.
Another wildcard is Microsoft. Widespread adoption of MS operating systems for mobile devices by enterprise users could easily bleed down into the consumer segment over time. (The open platform computing device will be an attractive way of dodging charges for services like MMS by connecting to landline 3rd party gateways or alternatives like packet-based instant messaging.)
OK, on to James's next control points:
Leverage their network to control the higher layer
To keep themselves relevant and "sexy" to their investors, they will move up the chain. They will provide not only their infrastructure services but also Internet access and even new service and content. And don't expect them to give right-of-way to other Internet Access Providers, or Service and Content Providers given a choice.
...
Moving to control the upper layers by leverging their infrastructure is a stronger hold. We see this in action when FCC frees broadband providers from the right-of-way requirements. Unfortunately, they are going to learn the hard way they are infrastructure (or access) provider, not content or service company. Moving up the upper layer is equivalent of a power company decide to go into factories or data center business and restrict power supply to their competitors.
In the short term, this could indeed be a strong means of price discrimination. Outright service blocking is unlikely -- it just doesn't make for good PR. Closed systems have a terrible record of market success when an open alternative becomes available. As I wrote previously, NAT is one way of segmenting your users and forcing them through your (metered) gatways. In the long term, it isn't going to work. Technologies like Opportunistic IPSec combined with low-latency wireless broadband will eliminate price discrimination based on port or IP address. NAT will be unsustainable when users roam onto multiple networks -- WWAN, personal WiFi at home, corporate WiFi network, municipal WiMax, etc. It's just too easy to tunnel your way past control points. Eventually someone will package up the tunnel in a seamless offering, and overcome the technical inconvenience.
Let's keep up with James's list of carrier control points...
Control by Regulation
They will seek set barrier on entries by lobbying for regulation on new services (such as VoIP). With the army of lawyers and experience in regulation, it is a battlefield they are familiar with and have confidences in.
... [hacking some text around]
If we are lucky, they (and the regulator) will learn their mistake and correct this but most likely, this will end up cripple the industry.
Some capitalist natural selection will take care of this in the long run. Economies that regulate the dumb network out of existence to keep incumbents in business will simply fall behind. In the short term, the regulators are being disintermediated too -- they're only just starting to launch policy fora on things like VoIP wiretapping, but the encrypted IP telephony genie is already out of the bottle. Blanket taxes like access charges may slow adoption of open networks, but they don't increase carrier control over those who would disintermediate their services. I can see some big fights coming over ENUM and control of the E164 (phone number) name space when applied outside of PSTN telephony.
I think there are some other control points. The default wireless portal is not really part of James's list. Carriers are really "value chain integrators" -- they deal with market co-ordination failures. This is the message coming loud and clear from Vodafone and NTT DoCoMo from the recent Midem conference in Europe. Retailing of 3rd party service is an honorable thing to do for a carrier, even in a dump pipe world. Users want the devices to come pre-configured ready to use. This isn't likely to change soon. But I can see some weaker carriers eventually aligning with an MVNO like Yahoo!, MSN or AOL and losing control here. In the US, Verizon Wireless (the market leader, but with a weak focus on data) is already half way there with its alliance to MSN.
For wireless operators, there's still a lot of scope for tweaking the network in favor of home-made services: faster network authentication, packet QoS, over-the-air scheduling algorithms, priority access to network resources like SIP proxies, favored access to identity resources like ENUM, knowledge of network topology and current traffic status and preferred routing; there are many ways to give the home team an unfair advantage.
You shouldn't ignore the non-techie side either. Distribution channels matter. Wireless operators have a large own-branded and retail partner network. If your open-API uncontrolled phone is only available on eBay, it won't get market adoption. People buy Coke from vending machines because of the convenience of the distribution network; the availability of cheaper drinks bought in bulk at a wholesaler or supermarket doesn't count. Branding and positioning can create seemingly "irrational" consumer acceptance of the "safety" of the familiar integrated service model, rather than the adoption of the alien 3rd party service. (It may be perfectly rational to reject the cheaper, more functional open alternative if the brand promise of reliability and privacy protection is too weak.)
So while the future of a dumb center and smart edge may be inevitible, it probably isn't immediate. Which is what makes this process so fascinating to watch.
Over at WiFi Networking News they report that iPass are doing very nicely thank you. To remind you, iPass are essentially a provisioning aggregator: they sign deals with multiple WiFi network operators, and retail that unified connectivity onwards to enterprise customers.
What makes this interesting is that it indicates that the whole value chain of a telco is unravelling. Many wireless network operators are based on internal aggregators, through the use of affiliates or part-owned subsidiaries. This function is normally hidden away in a morass of roaming agreements and home agent network elements.
iPass could easily retreat into a wholesale business and let someone else be the retailer. As the broker of connectivity, they would be in a powerful position, with a natural increasing returns to scale effect -- particularly if they signed exclusive distribution contracts for WiFi network operators. It isn't hard to imagine them aligning with a broadband wireless WAN operator to enhance the value of their offering. Indeed, it is possible to imagine a showdown between the major cellular operators for the enterprise customer being won not by having the spiffiest network, but rather through superior distribution deals with 3rd parties like iPass (or even systems integrators like IBM and Accenture). In other words, adapt more quickly to a delaminated business model, focus on connectivity, time your unbundling of your business well, and you'll survive. Hang on to the "squeeze the customer" approach too long, and you won't.
The iPass approach also unties the user equipment from the connectivity service provider. They just sell you connectivity. This is important, because the network operator who issues devices usually tries to lock them down to extract maximum revenue from the customer, forcing the user through the telco toll booth. The lack of a vertically integrated business model makes it ever harder to enforce such an approach. Another seam of the old telco business model starts to rip.
Today comes the well-trailed news that Nextel are deploying a high-speed wireless WAN network, presumably based on Flarion's FLASH-OFDM technology. (Nods to /. and Techdirt. More analysis also at Reiter's weblog.)
As I said last week, 802.20 will be really important. Why? Well, as one well-informed colleage reminds me, 802.16 (aka WiMax) doesn't have doppler tolerance built in from the start. That means it stops working when you start moving. It is essentially impossible to build this capability in later without starting again from scratch on new coding, silicon, FPGAs etc. Some limited hacks may enable a bit of motion tolerance. Flarion's technology is converging with the 802.20 standard, which does have doppler tolerance from the beginning.
I've had a demo of a Flarion network. I sat in the back of a van driving at 70mph down the I435 watching someone play networked Quake. I've never felt such intense motion sickness so quickly! But the demo didn't just make me want to reach for the barf-bag -- it blew my socks off too. They built a low-latency push-to-talk application in ... 2 days. Most carriers have taken years to achieve this. Nextel's 2G Direct Connect technology, which is wildly successful in the US, took years to perfect. New competitors still haven't reached that bar. With Flarion, it's a few hours playing with your SIP toolkit in your favorite IDE.
This announcement is interesting for other reasons. The acronym gives it away: FLASH-OFDM stands for Fast Low-latency Access with Seamless Handoff Orthogonal Frequency Division Multiplexing. Let's digest that piece at a time, and not eat the whole cow at once.
The "fast" bit is the easy one. Some modest knowledge of information theory and physics tell you how much you can pump through a slice of spectrum between two points. FLASH-OFDM is efficient in its encoding and handling of multi-path issues, so it comes out well on any speed test.
Traditional 3G cellular networks establish a virtual circuit to each handset. To communicate, a signalling channel is used to bring the circuit up and tear it down. This can take a long time -- 10 seconds or more under load. The channel is also typically torn down after a timeout following a period of inactivity. So there is a lot of latency in establishing and re-establishing a channel. Then even once you have a channel, a queueing algorithm determines who gets to speak when. The distance you lie from the tower and amount of interference can affect how often you get a chance to speak. If you're unlucky, it might be too long for you to win at Quake against wired competitors. FLASH-OFDM is pretty smart about queueing and also enables QoS for IP communications. (I'll write about QoS another day and when and why it matters.)
As the Flarion marketing department puts it:
Third generation (3G) mobile networks, on the other hand, retain a circuit-switched, hierarchical architecture. Consequently there is tension between the design objectives and the current environment of the wired Internet and mobile voice networks. The resulting design compromises of circuit-switched networks, which are optimized for voice, impair their ability to deliver high-speed, low-latency data cost effectively. The resulting high cost-per-megabyte of data delivery over circuit-switched based networks will prevent the emergence of mass-market wireless Internet access. An alternative approach, focusing directly on high speed, low cost and low latency wireless data delivery is required. Flarion, through its innovative FLASH-OFDM airlink, addresses the challenge of delivering affordable mobile broadband.
The worst-case latency for a packet transmission on a Flarion network is about 5ms. In other words, about 200 times better than a CDMA 1xRTT network, typically found in the US, Korea or Japan.
The usual figures for data throughput you see for a CDMA or GSM/EDGE/UMTS network are maximum burst or sustained throughput. But most people don't spend their lives doing FTPs. Instead they are using chatty, bursty applications. So the apparent spectral efficiency of 3G networks is a mirage. Your true throughput is terrible, because you spend all your life setting up and tearing down channels, and waiting for your one time-critical ACK packet to be sent. For instance, see this paper on page 5 for how throughput of a TCP/IP links varies with latency.
It's the throughput of the complete system loaded with users that counts, not an artificial network test with a single laptop under the tower at 3am. Peak throughput is irrelevant.
You don't have to be a genius to see that low-latency plus high-bandwidth looks like a tasty recipe for next-generation IP-based voice apps. Voice is still the "killer app". But it is evolving once freed from the clammy dead hand of circuit telephony. Should enterprises start demanding end-to-end encypted voice, then Flarion can deliver it using off-the-shelf technology. Integrate presence, IM and voice a-la Skype? No problem. Anyone left with a faux-circuit network will be left spluttering.
Seamless hand-off matters for voice calls. WiFi is weak in this respect (today) because authentication and hand-off is slow. Plenty of people are working on fixing this. Dropped calls (or the perception of a dropped call) are the #1 driver of customer (dis)satisfaction with wireless networks. This is a "must-have" non-trivial feature.
So that was low latency and seamless hand-off. There was a bit more to the acronym soup, though. That OFDM stuff -- what does it mean? Well, CDMA (the basis for 3G) has some unfortunate disadvantages. It needs complex and expensive power modulation circuitry to keep everyone shouting at the same volume. The cell site coverage shrinks as the system becomes loaded (called "cell breathing"). The coding isn't built for IP communications -- it was all done for voice, with packet as an afterthought.
ODFM is generally a more tolerant technology, even if it is a bit heavy on the signal processing (lots of matrix algebra, folks!). For instance, those orthogonal frequencies make power management of adjacent multi-path signals less of an issue. To quote the Flarion marketroids again:
Since all conventional cellular wireless systems, including 3G, were fundamentally designed for circuit switched voice, they were designed and optimized primarily at the physical layer. The choice of CDMA as the physical layer multiple access technology was also dictated by voice requirements. FLASH-OFDM, on the other hand, is a packet switched designed for data and is optimized across the physical, MAC, link and network layers. The choice of OFDM as the multiple access technology is based not just on physical layer considerations but also on MAC, link and network layer requirements.
So we have here a co-ordinated set of changes at layers 1 and 2 of the protocol stack. What's fascinating is that this destroys the last refuge of the circuit carriers from business model change. Changes at the lowest technology levels ripple all the way up to the business model for the application layer. Wireless is not a safe harbor.
Thus far we have seen gradual but unstoppable substitution of wireless minutes for wired calling. Within a decade, Flarion and its competitors could potentially drive carriers out of the voice service business entirely. Cable companies and wireless ISPs will be able to give away voice to get you to buy into their pipe. Why? Because an ever-larger proportion of calls become VoIP, and the only significant direct cost is a remote directory look-up for a SIP routing function. Off-net PSTN interconnect shrinks to zero.
If Flarion tweaked their technology to lower the peak throughput and instead give a predictable low-latency 100-200kbps to more users at a lower unit cost, you would have the killer transport for a wireless successor to Skype. Now that would be interesting. Any VCs out there wanting to fund some market disruption?
So little time, so much to read. I've already got too many books sat around the house that have done nothing but enrich Jeff Bezos. So it was good to read this concise summary of Good to Great. The book is a seemingly excellent guide to success based on rigorous analysis of what separates mediocre from exceptional business performance.
One particular part stuck in my mind:
I learned a lot from Collin's discussion about the second circle of influence, your economic engine denominator. For some companies the focus is profit per employee. For others it is per geographic region, per brand, per local population, per ton of finished good, or per customer visit. What is your economic engine denominator? Dr. Deming preached that a manufacturing company should focus on producing the next unit of profit at the lowest possible cost.
What does this mean for you? If you are an employee or a consultant it may be how much money you can possibly earn in a 40-50 hour workweek. For a company like mine, that licenses information, it could be profit per partner.
Now, the problem that faces telcos runs something like this. The business model is delaminating into three strands. The denominator depends on which strands you follow.
The first is connectivity. It's hard to differentiate yourself at this layer, but not impossible. Higher speeds through new technology, greater coverage, more roaming, etc. can help you capture supernormal profits. But most telcos don't have a cost structure remotely appropriate to such a world. Maybe a few long-haul operators like Level 3 or Williams that have always focused on the pipe business can do it. The usual strategy to turn a dime is to hoover up bit-pipe capacity for a few cents on the dollar -- the previous two owners (the constructor and the rescuer) will have given their investors a cold bath. The pipe business is capacity-driven. Until you've filled the pipe, your whole business strategy should be focused on driving utilization.
The second strand is the application platform. This is a real mess today. The whole thing is just too fragmented for any application developer to be able to rely on a telco to provide an application platform. OMA, J2ME, Parlay, JAIN, 3GPP, SPDE etc. -- the list of API definers is endless. Until Microsoft or IBM come along and clean it up properly, telcos aren't going anywhere in this space. Which is sad, because they could be participating in slices of other people's value chains a lot more (identity management, billing, retail distribution, etc.). I've no idea what the denominator is for a platform business. Maybe adoption among developers, driving a network effect.
Now for the interesting one. The services layer. Telcos have a lingering desire to remain in this space. But what's the driver? Transactions. If you aren't engaging in metered transactions, then you're the pipe provider -- and give up on trying to extract the true application value from the customer. It's transactions that count. Transactions. Transactions. I'll say that again. Transactions. Get it?
Of course, absolutely no telcos are currently using transactions as a key operating metric. Which is why you're probably best keeping your retirement dollars out of the retail voice calling sector of telecom, unless you've some really juicy insider information. Telcos are slowly being forced out of the transaction business.
Unfortunately, transactions alone aren't good enough. We've already got cheap all-you-can-eat on wireline and VoIP. As we head ever nearer to unmetered usage for wireless voice calling, the number of transactions (i.e. calls) rocket upwards. But that alone isn't a measure of success. The moment you flip on a cheap all-you-can-eat minute bucket, your transactions go up but your ability to segment and price discriminate goes down. Usage just drives cost.
So instead the key metric is profit per transaction. All those ringtones and games sold at dollar-per-transaction levels are a Good Thing. Premium instant messaging services. Voice recognition dialling usage. Whatever makes the customer buzz.
A profit-per-transaction metric might even work for a platform-based business. You spend billions issuing phone nubers, network logons and passwords. The ability to federate those identities you created at great expense with third parties (for a fee) helps. What can't you log on to your eBay account with your phone number? Mobile wallet transactions are part of an application platform business. Premium 3rd party content billing is an application platform service.
Lots of big transactions equals success. Lots and lots of extremely low-profit transactions equals danger, because extremely small can vanish into zero. And even a big number multiplied by zero is still zero. People do often pay over the odds for a predictable all-you-can-eat plan. But that's not the recipe for telecom success, where you aim to bundle haulage with user product. It doesn't drive you to seek new value, but rather towards reducing cost. If you want to stay in the services layer, there's much unfulfilled possibility in the communications business. There isn't a rest stop on the route to service innovation called "low cost".