YUP!

I think his ego is getting bigger and bigger now that Orange Box is out. I mean why would you go out of your way to show your strong disapproval toward other companies' products unless 1) it is in the interest of your company 2) you are flexing your mass media muscle.

While I think PPU may not have been a good idea, it was something new and I think its developers should be applauded for that. Just because Valve decided they don't want to use it, they don't have to go around insulting its future while it has the media spotlight. Similarly for the PS3, just because Valve decided not to develop for it and let EA UK do it instead, doesn't give them the right to say it is useless. Unless Mr. Newell and his team has done the port themselves, seen the future, and came back on a time machine to warn us all, this is nothing more than an elaborate excuse of why they didn't choose to do it. Most companies would just say they themselves didn't want to use their limited resources on the pursuit, but he has to try and speak for the entire industry.

The reason for the PS3 comments was the difficulty in programming them effectively (long learning curve) as the various parts of the PS3 are quite specialized.
Various articles have talked about that over a year ago as has Mr Newell being quite vocal about it.

He is right, PS3 sucks, and physics cards won't be needed with multi CPU\GPU.

Speaking as a CS double major, I just find it odd for a programmer to be so against doing development on something new and (r)evolutionary. I don't know any hardcore CS undergrads and Meng who are against a long learning curve when it is interesting.

perhaps, instead of (r)evolutionary he vioews the cell processor as a dead end technology.

Judging historically, the x86 architecture has vanquished most other architectures.

New Sun boxes all ship x86 solaris with AMD chips, apple uses intel in place of their old RISC architecture.

so far x86 has been very resilient.

He may not see the point in learning to use a difficult new architecture if he believes it will die in a couple of years anyway (which, jusging from his statements, is exactly what he believes)

I think he's spot on.
I don't think Valve make the best games or write the best code but I think that Gabe has thought this through.

Think about it!
The more complex the games become, the more dependant each part of the game engine becomes on another part.
PS3 is the only gaming machine in the whole world that uses Cell, why learn how to program for it if you'll just put the code in the trashcan later? Computer architecture will look different when Sony's next console hit the streets.

But, you can still make software for PS3 and still make money from it. It's fun to create fun games and make money from doing so but I think Gabe is thinking further than that.