If you had to explain 2006 to somebody in one sentence, the honest version is that a great many companies spent it positioning for fights that start next year.
The console generation finally arrived and immediately got complicated. Microsoft had a year alone and used the back half of it well. Sony launched at a price that startled people, into supply it could not meet, and left Europe waiting until March. Nintendo turned up with underpowered hardware, a strange controller and a number low enough to be an impulse buy, and cannot make them fast enough. I wrote at the start of December that two of these machines are not really competing with each other, and nothing since has changed my mind.
Apple finished moving the entire Mac line to Intel, which is the sort of transition that normally takes a company three painful years and destroys a product cycle. They did it in about twelve months and the machines got faster. That deserved more attention than it got, largely because everybody was busy speculating about a phone that did not appear.
Google bought YouTube for one and a half billion dollars, which was either an enormous overpay for a site with no revenue and a copyright problem, or the cheapest acquisition of the decade. I genuinely do not know which and I am suspicious of anybody who claims to. What is clear is that video moved from something you downloaded to something you clicked, and it happened faster than the people who own the video expected.
AMD bought ATI, which reorganises the entire component industry. There is now one company selling processors and graphics as a package, one company selling graphics alone, and Intel sitting on top of the volume with integrated parts that are terrible and shipped in enormous numbers anyway. Nvidia is the one with the most interesting problem there.
Windows Vista did not ship to consumers, again. Five years after XP, the delay has stopped being a story about a schedule and started being a story about whether a project of that size can be run at all. Businesses will get it first and take years to move. The knock-on effect that matters is DirectX 10, which is tied to Vista and therefore tied to the delay.
The format war started properly, with both Blu-ray and HD DVD on sale and neither winning. Sony has bolted its side of that fight onto a games console and is asking players to fund it, which is the most expensive part of the PlayStation 3 and the part fewest buyers asked for.
In phones almost nothing changed, which is itself the observation. Nokia still ships about a third of everything. Motorola still sells one silhouette in five variations. The interfaces are still nested lists navigated with a directional pad. There is more capability in a midrange handset than anybody can reach, because there is no way to put software on it and no shop to sell that software in.
That is the thread running through the whole year, actually, and I did not notice it until I wrote the list out. The interesting constraints in 2006 were almost never technical. Sony’s problem is diode supply and a pricing decision. Phone gaming’s problem is the carrier deck. Vista’s problem is organisational. The graphics business is being reshaped by an acquisition rather than an architecture. Almost nothing that mattered this year was limited by what engineers could build.
Which makes me think the thing to watch in 2007 is not a specification. It is whoever works out how to get software to people without asking permission first. I have no candidate. I would like one.