The Missing Sync has been released with Windows Mobile 2005 support, and boy this works great. I've been lurking (and sometimes contributing) in their beta mailing list for the past few months, and as a fellow Windows Mobile developer, I've felt their pain during the process.

What most (non-developers) don't realize is the huge expense both in hardware and time in developing for Windows Mobile. This always ends up frustrating me whenever I switch over to, say, updating Conversions In Hand from working on a Mac project. Consider the vast number of devices produced by a seemingly random selection of OEMs, each with their own particular idiosyncrasies and, because OEMs don't like to provide O/S upgrades anymore, different versions of the O/S itself.

We actually have a stack of defunct Windows CE/Pocket PC/Windows Mobile devices, sitting in a box, that we've had to buy over the years. Coupled with the huge chunk of money re-distributors take for each software sale, this ends up making Windows Mobile development, mmmm, less than lucrative.

We've taken to testing on only a small handful of devices, along with the new native code emulator in Visual Studio 2005. Hopefully (for us anyways), this will reduce our costs and frustration, although it's going to leave some of the fringe devices out in the cold.

Interestingly, it looks like Microsoft is going to take the same approach with their "iPod Killer". It's too bad really (for them, anyways), because all they'll end up doing is confusing consumers with the plethora of units produced by various OEMs, some known and some unknown. New units will appear and older ones will disappear on a seemingly daily basis. Said consumers will end up buying an iPod because it's the simplest choice.