Some time ago I received a private email directly from Elop (just me, nobody else in CC, I am not going to go into details as to why), in which he explained that the biggest problem was the small amount of MeeGo devices in the years immediately ahead.
This is simply not true.
Before explaining why, I’ll quickly say that I actually work on hardware adaptation, so if somebody knows the amount of effort needed to adapt MeeGo to different hardware platforms, it’s us. Plus, I closely follow Linux related mailing lists (linux-arm, linux-omap, linux-media, etc.), and know a lot of people in different companies that work precisely in this area. I have quite a few years of experience doing this, so I know what I’m talking about.
Update: To avoid confusion, I am a mere software engineer. And when I say “us” I’m talking about the bigger team I am part of.
Nobody I know believes what Elop said, and let’s keep in mind that Elop is not an expert in this area, we are. So my guess is that he got his information from some upper management guy who didn’t know what he was talking about
As I explained Elop, if we wanted to ship 10 devices with OMAP 3 (the same platform of the Nokia N9) today, there is absolutely no problem from the software point of view: all the UI software remains the same, and the hardware adaptation would probably require few modifications, if any.
The problem is when porting to an entirely new hardware platform, say Snapdragon. Suppose only 3 devices are planned on the “years immediately ahead”, well, then it makes sense to have 3 different hardware platforms, and each one of those requires work from the hardware adaptation team, not from the upper layers, though. However, that’s not a technical limitation, it could very well be 30 devices instead of 3, it’s basically the same amount of work for us. IOW; what matters is the hardware platform, not the number of different devices.
Note: all these are merely examples, not actual plans
Funnily enough, Windows Phone only supports one hardware platform: Snapdragon (and in fact only certain chips). So MeeGo already has an advantage over Windows Phone; you could ship more devices on more hardware platforms. All we need is the word.
Not to mention the fact that most of the hardware adaptation is already done by hardware vendors. They do it because it’s the easiest way to demo their hardware (it’s Linux). I tried to explain that on an earlier post where I show many examples of people porting MeeGo to a plethora of devices (it’s easy and fun).
Another advantage of course is that MeeGo is already here (Nokia N9).
Not to mention the fact that MeeGo is open source, and Linux is a synergetic endeavor; there’s many more than one company (Microsoft) working on it, in fact, almost everybody else is.
Elop’s answer? “I am simply going to choose to respectfully disagree on multiple fronts”. He didn’t even bothered to mention exactly what was the disagreement.
So there you have it, if there’s a reason for ditching MeeGo, it’s certainly not a technical one, and most likely not a good one either. I hope the people out there like what we did with the Nokia N9 and ask the though question “Why exactly did you leave MeeGo, again?”, specially when there are no signs of any Windows Phone device.
Note: as usual, this is my own personal opinion, and it’s based on publicly available information