An old coworker (or, the shibboleth of an old newsgroup of mine, cow orker), Jason Booth, has blogged about his low opinion of the PS3, and was inundated with a lot of Sony defenders saying all kinds of, well, web-forumish things.
Trouble is, he's right, guys. Face up to it.
Developers all over are not feeling the love from Sony. The PS3 does some things very well, and a lot of other things not well enough. But its real trouble is that the architecture is so weird, so painful to deal with, and so non-portable, that no one really wants to invest in it. I get the impression that, when the Cell processor was introduced and was getting some mainstream press about being the next generation of computer architecture that would blow everything else away, Sony sort of hopped on that bandwagon without thinking things through.
The Cell is a weird beast, no one really knows how to deal with it or get the most out of it, the PS3 is not selling well enough to inspire folks to wrestle with it, and even if you do commit to it, none of that effort is really portable to other platforms: you can't bring over much of your existing tech, if you have any, and you can't really take tech you develop for the PS3 anywhere else, should you need or want to. You sort of have to marry the PS3 and hope you bet on the right horse.
And really, you haven't. There are three SKUs of XBox360, and each of them by themselves outsells the PS3. The PS3 will have Metal Gear Solid, which will make or break the platform: if MGS isn't a must-have, must-play, must-see game, the PS3 will have a very hard time of it.