I have a feeling that a lot of confusion in this thread will be cleared up after Valve reveals their 2nd and 3rd announcements on Wednesday and (presumably) Friday. Likely including the unveiling of Steambox itself, and their plans for involving the community in shaping Steam's future. By then we should have a clearer picture of Valve's overall vision and direction here. Hopefully Gabe and others at Valve will make themselves available for interviews in the near future to further clarify some things.
I suspect Valve has three main goals here:
- Make further inroads into integrating PC gaming into the living room, obviously...and entice the "living room" audience, while still retaining most of the advantages of PC (mods, openness, etc.) BPM was a clear first step in that direction. However, there is still a belief among many people that PC gaming is some crazy expensive uber-complicated mess that requires editing hex values in some obscure file to get a game to run. There is still a lingering psychological barrier that needs to be overcome.
- Expand the Steam userbase by getting more people onto it. After all this is where Valve makes its money. I don't think Valve really cares how you access Steam...be it SteamOS, Windows, OSX, a monstrous Windows desktop running dual GTX Titans, a mid-range HTPC, a laptop, a Steambox from Valve themselves or one of their HW partners. As long as you're on STEAM. SteamOS (and Steambox) is just another option aimed at certain niche markets. Those of you who are struggling to see the point and the appeal of SteamOS...you're not seeing the forest from the trees. If you already have a Windows living-room HTPC for BPM, then SteamOS isn't for you. If you already have a decent desktop PC in your study and are uninterested in streaming to another room, then SteamOS probably isn't for you. If you're looking to get into PC gaming or upgrade your ancient hardware then a Steambox may be viable option if the specs and price are good.
- Further diversify what OSes Steam can run on, to try and protect itself and the openness of PC gaming from the possibility of Windows someday going fully walled garden (which I doubt will happen even if that's what MS wants). This is more of a long term strategy; it will likely take years to pay dividends. SteamOS may be their primary focus, but Valve isn't going to stop supporting Windows and Mac.
I am eager to learn about Valve's controller and about these "many" AAA game announcements for SteamOS. I've always felt that Valve's biggest challenge will be convincing large devs to natively support SteamOS and OpenGL...it looks like they may have had some success here.
SteamOS isn't for everyone, and for most people it may be lacking in short-term and/or immediate tangible advantages (as opposed to long term), but I think this is a good move by Valve.
I still wish they'd spend some time overhauling and optimizing the current Steam client, though.