Ahonen is a smart cookie in a lot of respects, but he can't let go of the idea that Symbian and MeeGo were the future for Nokia. While I think in retrospect Elop made a few bad calls, when he got there MeeGo was not in production-quality shape. Nokia's own assessment was that they would have one MeeGo unit ready for shipping in 2011, and that particular unit--what eventually came out quietly as the N9--was an intriguing but crashably unusable demo unit even in late 2010. This isn't speculation on my part, either: I was at Nokia in 2010 and used one of them briefly. Or tried to. There was a lot of stuff that just didn't work. In November 2010. On the only device that was far enough along to have hardware that could be shown to other groups.

And Symbian? Jesus. To implement a "search as you go" feature for Symbian^3, something that would narrow down search results both on the phone and off as you typed (like iOS and Android devices do, so this was just achieving parity), we had to write our own keyboard because the Symbian keyboard took over the whole screen. You selected a text field and got the keyboard, then hit the keyboard dismiss button to see the screen again. But look on the bright side: in addition to the "Universal Search" team in San Francisco working on that, there were four other groups around Nokia, both employees and contractors, working on what was, essentially, the exact. Same. Product.

And this was de rigueur for Nokia. This was the problem that Elop had to fix beyond anything technical: massive duplicated effort on what, at any other company, were already solved problems, because Nokia had become a bunch of fiefdoms all fiercely protecting their own turf. I'm pretty sure this is why Symbian stuck around in essentially the same form for so long. I'm sure Tomi Ahonen loved it as a user, but the number of engineers working on Symbian who liked working on it that I talked to was, to the best of my recollection, zero. (Just getting the developer toolchain running on my machine made me feel like I'd time traveled back to the days of SLS Linux.)

If I ran the zoo, I'd have probably gone with Android rather than Windows Phone; Elop went the other way not because of his past connections with Microsoft, but because he wanted to keep using lucrative Nokia services (rather than Google counterparts, most notably Maps) and be allowed to use the Android branding, and Google said no. I wouldn't have cared about the Android branding--instead I'd have cared about porting Qt to Android and not burning bridges with Nokia's existing development community. They'd just spent a lot of time, money and effort convincing everyone that Qt was the Way Of The Future: instead of writing to Symbian or MeeGo APIs, you'd write to Qt APIs, and then porting would be simple. Losing that was a bad call.

But I'd say the most serious mistake Elop made had nothing to do with technology. Love Apple or hate them, one of the things they're really good at is not talking about products before they're ready. The rumor mill may be in high gear for months, sure, but when Apple makes their official announcement, it is nearly always in the form of "Here is our new Superlative Magical Thing. Here is a video of Jony Ive talking earnestly about it. It will be available for this price (two weeks from now|tomorrow|as soon as the band stops playing)." If Elop had been running Apple during the PowerPC-Intel CPU switch, he'd have leaked an internal memo describing the PowerPC in some colorful apocalypse-invoking metaphor, then three months later announce the switch, then six months after that introduce one laptop with an Intel chip that wasn't the most current generation and say they'd have more coming next year.

The lesson to draw from Nokia under Elop isn't that they should have stuck with MeeGo; it's that they should have kept their mouth shut, kept shipping Symbian phones through 2011, and started 2012 with their new devices ready to ship when they announced them. I wouldn't be surprised if Ahonen gets his wish within a year and Elop is canned by Nokia's board -- but I'd be very surprised if doing so improves their situation.