Now there's a moving target. Given the amount of testing that is required just to get modern games to work across most hardware without problems (on Windows), Linux (which distros?) as a tier-1 platform is wishful thinking. The mass gaming market exists on stable, more or less predictable platforms.
He didn't say anything about Linux as a "tier-1 platform". I don't think there are many of us fooling ourselves about the worthiness of that kind of investment; there's just not enough money in Linux gaming to hold up a release on Mac/Win for Linux issues.
The guy said OpenGL-based games might be good for Linux, which is indeed true; it's much easier to go from an OpenGL Mac game to an OpenGL Linux game than it is to go from DX Windows to OpenGL Linux.
Games on Mac are good news for Linux because if the game is made sanely, it shouldn't be much work to get it to at least compile on Linux, as both systems are more or less POSIX compatible. This is good! Even if there are problems with the Linux build after a successful compilation, we're still much further ahead than we are with a DirectX Windows game as the target.
Furthermore, making the OpenGL renderers available under Windows may greatly improve the experience in WINE, especially for non-flagship games. Again, there's no guarantee that this renderer will be exposed in Windows, but even if not, there's at least a tangible possibility that it might happen sometime.
The bottom line is that the investment required to port a Mac game to Linux is almost always going to be much less than the investment required to port a [DirectX] Windows game to Linux. Therefore, previously Windows-only games coming to Mac is great news because at the very least it makes the possibility of a Linux release that much closer.
