In 2014 though, I think the biggest impact on their success will come down to marketing. Will people know what this device is? Will hardware manufacturers be able to get these units selling? Will people be raving about playing Steam games from their couch and telling all their friends to pick one up?
They're going to be a small player in the console world. I can't picture them stealing Xbox and Playstation users. Those Battlefield, Call of Duty, NBA, Madden, etc., fans make up a huge part of the console market, and I think they're incredibly loyal to their console, while being very unfamiliar with Steam. The Steam Machines might have a market with PC users looking to casually game in the living room, and they might have a market if they can offer cheaper hardware and lower game prices than consoles. Offering better visuals would be another huge plus. If someone visits their friend's house, and says, 'wow, why does XYZ game look so much better on your television?', that's going to drive some sales, since you always have a group looking for the best of the best.
It's a long road, and I think everyone is watching to see how this plays out. Valve is probably the only company that can make this happen at the moment, so it's exciting to see them taking a chance, and I applaud them for that.
They have a chance to make Steam Machines a huge success. If they pigeon hole themselves as strictly being a gaming device, and not a media device that competes with consoles, they should change the name of their UI, because they're missing the big picture.
However, I think you're wrong. Yes, they want a backup plan if they get kicked to the curb by Microsoft. However, if that was the only case, they'd be focusing on making Steam OS an actual operating system. Why spend the resources on a gaming controller, and Steam Machines? Why not go hard on a Linux OS that's a viable alternative for gamers? Their core audience is PC gamers, so why not make an OS for PC gamers, and skip the living room hardware all together?
To me, they are still targeting PC gamers (and Gabe clearly said so, too) and they want to expand that segment from the top, not from the bottom. Maybe they will attract some PC gamers who did the switch to consoles a couple of years back, by providing something better than consoles while still "living-room" centric.
I think their strategy is however the following: - a Steam gamer currently is usually a guy in front of the PC in his room. - By putting a Steam machine in the living room, it offers the opportunity to open up Steam to family members. - Oh, by the way the family games management feature was introduced about the same time as the SteamOS release this year... what a coincidence huh ? - So they plan to expand through immediate social environment first, not necessarily targeting hardcore console users. That way they will create awareness progressively. That's why I am saying they are in for the long run.
Just like Steam was not made to replace from Day1 the box retail business, SteamOS's strategy will not replace consoles from Day1 either. Maybe it will be a big player in 10 years from now, but I don't expect any significant market share any time soon.
As for solving your problem with hardware confusion, they could have manufacturers ship Steam Machines with some kind of rating system. Remember how Windows had those performance ratings for memory, processor, etc? They could do something very similar. You look at a Steam Machine, it says processor 5.7, memory 4.5, graphics, 5.1. Then you look in the Steam game library and see a certain game recommends graphics 4.5 or greater. Or this indie title plays great on anything 3.4 or higher. You use those numbers to guide your purchase.
Changing hardware will come to their advantage though. Xbox and Playstation will ramp up over the next few years, and we'll see some improvements and major titles heading their way. However, what's happening 5-6 years from now? Xbox and Playstation are dated, Steam OS has the bugs worked out, and Steam Machines are available with the latest advances in hardware. Where are you going to buy Call of Duty now, for your 6 year old console, or pickup a Steam Machine that can run it twice as fast, at Ultra HD resolution? That's going to be a big plus for Steam, and if people are buying in 5 or 6 years from now, they're less likely to jump on the next console launch, since they already have powerful hardware in their living room.
Another huge plus for Steam in the living room is Oculus Rift. If it's released below $300, and exclusive to the PC and Steam Machines, they have a unique experience the consoles can't offer.
Nothing is short term, I think Steam Machines will make a crawl the first few years, but I think they're going to be pushing to replace consoles, and I think they might succeed. Cheap games, big sales, Oculus Rift, low cost or high performance hardware to choose from, and eventually an open store available for anyone to publish towards.
Edit: Oh, one other huge advantage. They're backed by every hardware manufacturer they get involved. Dell is going to be dropping flyers in every mailbox around the world, with a couple of pages showing off their Steam Machine line-up. Dell wants money, and if Steam Machines succeed, they get rich as well. PC sales are down, they'd love another piece of hardware to sell. Get every other hardware company doing the same, and that's a lot of free marketing.
As for the ratings system, sure, it's possible but it does not mean it's not confusing for consumers, and it adds another barrier to adoption - that's basically NOT mainstream. And you mentioned Windows performance ratings as a reference, but that's a failed example since nobody ever used these numbers anywhere.
If they go for categories/benchmarking, then they will need to have a kind of authority in place to attribute such numbers, and Valve's message so far is very against centralization. So I don't expect them to lead such efforts. If they ever did, I believe they would make it simple: SteamBox Level A to play these games only and do streaming, SteamBox Level B to play all GPU intensive games... etc...
EDIT; I don't think Dell will get rich, and neither most of the hardware manufacturers involved for that matter. SteamOS is no more no less than an Android OS-like for the PC, and that will push prices down for manufacturers which will fight hard to make small margins on large volumes. It will only be profitable for very few players.
WHen you want to do virtual reality, you need constant 60 fps (or more, ideally) in order to give a good impression of movement in all directions. XBone and PS4 are already struggling to display a SINGLE 1080p screen at 60 fps already, let's not dream too much. They are massively underpowered, and the AMD APU is notorious for being mid-range in terms of performance at best in 2013. Give it one more year and even the cheapest gaming PCs will be more powerful than them.
What? I run Steam as a non-Administrator user on Windows, and a non-root user on Linux. Would you elaborate on your objections?
These installers would be an issue whether or not one was running Steam. If you were installing from a disc (heh) rather than using Steam, all that installation stuff would have happened while you were running the installer on the disc (with Admin privs).
So yes, installers that potentially install stuff under %PROGRAMFILES% and %SYSTEMROOT% will require admin privs. Unless you change permissions on the associated directories and registry trees, there's no way around that.
AFAICT, Steam itself (and, indeed, many games sold through Steam) does not require admin privs. Maybe I'm wrong about this. That's why I asked the OP for clarification.