And what I'm saying is that if the kernel is Linux (really who cares about pedantic details like whether it is a kernel or an OS... I mean, really?), that it can't run all the existing Steam Windows games natively and hence you are limited. Unless all those games magically get ported to Linux.
Or care about. Which was one of my previous points. The only places Linux gets some success in the consumer market is when the OS is nicely hidden away and end users don't need to interact with it directly. No-one goes to Frys and buys a router because it runs Linux. No-one cares.
Yup, and that's the point that Bass was making: there is no reason why the user will need to be exposed to the internals of Linux (that was the second part of his comment which you skipped over), in much the same way that users don't have to interact with Linux when they're using their set-top boxes
I even develop and debug on both Linux and Windows at work (although by far I prefer to develop on/for Windows).
And this proves what exactly?
But there are cases where it just doesn't make sense and trying to create essentially what is a gaming PC using Linux just isn't going to be a success.
The success or failure of the box has nothing to do with the operating system it is running; it will live or die based on the marketing ability of the companies behind it. Linux is just the free core they've used as a operating system for the box. I suspect what we really need to look out for is the gaming engine and the UI that they've built on top of that. That's the important part.
And no-one has answered my question as to what makes a Linux based "SteamBox" better than say any one of the available pre-built, living-room friendly gaming PCs running Windows and Steam? Hmmm?
Yeah, the problem is that you're focussed on the Linux part; try to see a little further. The detail we don't yet know much about is how the games are going to be built? How will they talk to each other'? That's this multi-room, multi-screen thing all about. Linux is unimportant in this regard.
A lot of folk have also pointed out, quite rightly, that they have no chance of taking on the XBox360. I think they know that. What they're looking for is to build a profitable infrastructure around a networked game engine (not a Linux PC – we already have enough of those, thanks). They don't actually need to outsell the PS3 and the XBox to do that. The first thing they should do is avoid a strategy that results in their losses mounting as their sales increase.