Just as one aside on this, it's pretty clear by now that it isn't simply a matter of doing a source control get and the vendor's work is done.
Aside from simply having technical insight into the individual ROMs, you needn't look further than the disaster that has been Sony Erricson and Dell's Android entrants to see the truth in this -- those vendors came out with devices running dated Android versions, turning possible winners into catastrophic losers, eviscerating sales of the Streak and X10.
If they could have just done a merge and they were done, obviously they would have.
So now we're getting to the point where it seems that makers like Motorola and HTC have started to build up a significant Android talent pool, and that bodes well. Despite the constant incantations that Android is free, I wouldn't be surprised if the in-house development costs rivaled or exceeded what something like Windows Mobile cost to license.
I wouldn't be surprised if the in-house development costs rivaled or exceeded what something like Windows Mobile cost to license.
That implies that Windows Mobile doesn't require similar in-house effort, on top of the OEM license fees. Windows phones still require the same hardware driver development effort, and WM7 is already starting to receive the carrier-bastardization treatment. I'd say Android's licensing being free is the only difference.
Android OEMs have to develop their own drivers to get the various components (screen, GPS, WiFi, Bluetooth, 3G radio, accelerometer, etc.) working whereas with WP7, Microsoft will provide all the device drivers. Part of this can be attributed to the fact that Android devices have a much wider range of hardware options, so it's a lot more difficult to have an uniform set of drivers.
All current Windows Phone 7 devices are restricted to the Qualcomm QSD8x50 platform, so there isn't much customization to be done as the hardware is nearly the same for every vendor.
As far as I understand it, Windows Phone 7's API is Silverlight with a few restrictions related to screen size and the like.
I haven't coded anything for Windows CE, but making a Silverlight app for Windows Phone 7 is about a difficult as making a WCF app for Windows, or an ASP.Net app for the web.
I got an X10 for free with upgrade over the holidays and love it. It makes me wonder if all the talk about version upgrades is mere noise. I'm relieved not to be affected by the Android SMS bug in Froyo, so will happily wait until it is fixed. Even if my phone ever sees an upgrade to Eclair, I probably won't notice with AT&T's customizations, which I like quite a bit. It's funny to hear the X10 described as a loser when I like it more than the iPhone. Oh, well, I liked XP more than Vista, too.
Wait, didn't everyone like XP more than Vista? I know I used Windows from version 3.1 right up until my experiences with Vista convinced me that spending an extra $1000 to get a MacBook Pro instead of a Vista laptop was completely worth it...
I am talking purely from a market perspective. The X10 was hotly anticipated and was set to make waves...but then it was announced that it was coming with 1.6 and that balloon deflated almost instantly. It is a gorgeous device, but a simple software misstep seriously hobbled its adoption.
Dell has suffered the same outcome for its entrants into the Android field. Clearly they were trying to enter the market on the cheap and it hurt them.
Aside from simply having technical insight into the individual ROMs, you needn't look further than the disaster that has been Sony Erricson and Dell's Android entrants to see the truth in this -- those vendors came out with devices running dated Android versions, turning possible winners into catastrophic losers, eviscerating sales of the Streak and X10.
If they could have just done a merge and they were done, obviously they would have.
So now we're getting to the point where it seems that makers like Motorola and HTC have started to build up a significant Android talent pool, and that bodes well. Despite the constant incantations that Android is free, I wouldn't be surprised if the in-house development costs rivaled or exceeded what something like Windows Mobile cost to license.