(This isn't a product review, it's a big-picture overview brought to you from the universe of "Halting State".)
It shouldn't be news to anyone that smartphones — as a category — really took off in the second half of the noughties. Before 2005, few people bothered with PDAs, and fewer still with phones that had keyboards and could browse the web or send email. Current projections, however, show 25% of all phones sold in 2010 being smartphones — and today's smartphone is a somewhat more powerful computer than 2002's laptop.
At the same time, the winners in 2005's smartphone market (Palm, Windows Mobile, Symbian Series 60, 80, and UIQ) are losing ground rapidly (PalmOS is already dead, modulo the Hail Mary pass that is WebOS on the Pré) while strange new mutants slouch towards market dominance — Android, Mac OS X, and maybe Maemo.
What's happening?
Here's my hypothesis ...
Pre-2005, digital mobile phones typically ran on GSM, with GPRS data limited to 56kbssec, or Verizon's CDMA. This badly choked their ability to do anything useful and internet-worthy. By 2005, the first 3G networks based on WCDMA (aka UMTS) began to open up. By 2009, 3G HSDPA networks can carry up to 7.2mbps. The modem-grade data throughput of the mid-noughties smartphone experience has been replaced by late-noughties broadband grade thorughput, at least in the densely networked cities where most of us live. (I am not including the rural boondocks in this analysis. Different rules apply.)
To the mobile phone companies, 3G presented a headache. They typically offered each government billions for the right to run services over the frequencies freed up by the demise of old analog mobile phone services and early TV and other broadcast systems; how were they to monetise this investment?
They couldn't do it by charging extra for the handsets or access, because they'd trained their customers to think of mobile telephony as, well, telephony. But you can do voice or SMS perfectly well over a GSM/GPRS network. What can you do over 3G that justifies the extra cost?
Version 1 of their attempt to monetise 3G consisted of walled gardens of carefully cultivated multimedia content — downloadable movies and music, MMS photo-messaging, and so on. The cellcos set themselves up as gatekeepers; for a modest monthly fee, the customers could be admitted to their garden of multimedia delights. But Version 1 is in competition with the internet, and the roll-out of 3G services coincided (and competed) with the roll-out of wifi hotspots, both free and for-money. It turns out that what consumers want of a 3G connection is not what a mobile company sees fit to sell them, but one thing: bandwidth. Call it Version 2.
Becoming a pure bandwidth provider is every cellco's nightmare: it levels the playing field and puts them in direct competition with their peers, a competition that can only be won by throwing huge amounts of capital infrastructure at their backbone network. So for the past five years or more, they've been doing their best not to get dragged into a game of beggar-my-neighbour, by expedients such as exclusive handset deals (ever wondered why AT&T in America or O2 in the UK allowed Apple to tie their hands and retain control over the iPhone's look and feel?) and lengthening contractual lock-in periods for customers (why are 18-month contracts cheaper than 12-month contracts?). And the situation with international data roaming is dismal. It doesn't hit Americans so much, but here in the UK, if I travel over an hour by air, the odds are good that I'll be paying £6 per megabyte for bandwidth. It's as if my iPhone's IQ drops by 80 points whenever I leave home.
Enter: Apple and Google.
Apple are an experience company. They're a high-end marque; if they were in the automobile business, they'd be BMW, Mercedes, and Porsche rolled into one. They own about 12% of the PC market in the USA ... but 91% of the high end of the PC market (laptops over $999, desktops over $699). How they got into the mobile phone market is an odd and convoluted story, but it's best to view it as a vertical upwardly-mobile extension of the MP3 player market (from their point of view), which has taken on a lucrative life of its own. Apple's unique angle is the user experience. Without OS X to differentiate them from the rest of the market, their computers would just be overpriced PCs. So it should be no surprise that Apple's runaway hit iPhone business team have a single overriding goal: maintain control of the platform and keep it different (and aspirational).
Apple don't want to destroy the telcos; they just want to use them as a conduit to sell their user experience. Google, however, are another matter.
Google is an advertising corporation. Their whole business model is predicated on breaking down barriers to access — barriers which stop the public from accessing rich internet content plastered with Google's ads. Google want the mobile communications industry to switch to Version 2, pure bandwidth competition. In fact, they'd be happiest if the mobile networks would go away, get out of the users' faces and hand out free data terminals with unlimited free bandwidth. More bandwidth, more web browsing, more adverts served, more revenue for Google. Simple!
This is where the Nexus One announced last week may be significant. If the rumours are true — that they're pushing it at a low or subsidized price, and have strong-armed T-Mobile (the weakest of the US cellcos) into providing a cheap data-only mobile tariff for it, and more significantly access to VoIP and cheap international data roaming — then they've got a Trojan horse into the mobile telephony industry.
I think Google are pursuing a grand strategic vision of destroying the cellco's entire business model — of positioning themselves as value-added gatekeepers providing metered access to content — and their second-string model of locking users in by selling them premium handsets (such as the iPhone) on a rolling contract.
They intend to turn 3G data service (and subsequently, LTE) into a commodity, like wifi hotspot service only more widespread and cheaper to get at. They want to get consumers to buy unlocked SIM-free handsets and pick cheap data SIMs. They'd love to move everyone to cheap data SIMs rather than the hideously convoluted legacy voice stacks maintained by the telcos; then they could piggyback Google Voice on it, and ultimately do the Google thing to all your voice messages as well as your email and web access.
(This is, needless to say, going to bring them into conflict with Apple. Hitherto, Apple's iPhone has been good for Google: iPhone users do far more web surfing — and Google ad-eyeballing — than regular phone users. But Apple want to maintain the high quality Apple-centric user experience and sell stuff to their users through the walled garden of the App Store and the iTunes music/video store. Apple are an implicit threat to Google because Google can't slap their ads all over those media. So it's going to end in handbags at dawn ... eventually.)
The real message here is that if Google succeeds, the economic basis of your mobile telephony service in 2019 is going to be unrecognizably different from that of 2009. Some of the biggest names in phone service (T-mobile? Orange? Vodafone? AT&T? Verizon?) are going to go the way of Pan Am and Ma Bell by then; the ones left standing will be the ones with the best infrastructure (hint: that doesn't look like AT&T right now — by some analyses, AT&T mis-understand TCP/IP so badly that their network trouble is self-inflicted) and best interoperability (goodbye Verizon), selling bits at the lowest price to punters who buy their cheap-to-disposable (phones are part of the perpetually deflating consumer elecronics sector; today's $350 BoM should be down to under $100 by 2019, for something a good deal more powerful) unlocked in WalMart and take ditchwater-cheap international roaming service for granted.
Probably around the time VoIP takes over from the current model, we'll see something not unlike DNS emerge for mapping OpenID or other internet identities onto the phone number address space. (God, I hate phone numbers. Running a phone service that forces everyone to use seven to twelve digit numbers is like running an internet that forces everyone to use raw IP addresses.) Then the process will be complete, and things will have come full-circle, and the internet will have eaten the phone system.
What's good for the internet is good for Google. Right now, the phone companies are not good for the internet. If I'm right about the grand strategy, the Googlephone will change that.
