I get a bit tired of hearing about Steam needing more quality control. Not because I think it would be a bad thing, but because the kind of control people want them to exercise is not only bad for PC gaming, but probably impossible. Having at least some minimum level of quality control on consoles isn't so hard. The hardware is set, it's impossible to release a game on that hardware without the manufacturers okay, and you can thereby make sure that games meet a certain level of quality by insuring they're using your dev kits, meeting certain programming standards, etc. Of course the thing people tend to forget is this is all really just bullshit and these licensing requirements and "quality control," have never stopped shitty games that barely run from being released and frequently only serves to support the failing publishing paradigm in effect right now and bleed small companies dry if they try to navigate them on their own. You can look at Xbox Live Arcade as a fairly obvious case of this model utterly failing to bring in smaller companies. It launched with everyone heralding it as a new way for people to get their games on consoles, and it fizzled and died because the requirements were frequently unclear, certification was incredibly expensive, and they couldn't compete with other distribution platforms that were easier to get one and significantly cheaper.
And all of this without ever even having to point out that there is no hardware standard on the PC, and now Valve is dealing with three different operating systems that games can release on. And christ knows there's enough Linux variants alone to make it impossible for them to quality check every single game. Sorry people, but it will never happen. But for some reason, a small handful of examples of shitty games slipping through the cracks as the system exists right now is enough to have people running for the hills saying the sky is falling. It's not. And it won't.
Because here's where the system needs to go, and where it probably will: user control of content quality. Valve can't police everything. They'll never be able to check out every game that's unreleased, or properly quality test everything out there when it's released, but the community reacts a hell of a lot faster to these things. Give users the power to bring claims against companies. Limit it to a certain number of trusted users and punish those who falsely report if you must. Take the time to investigate pre-release claims, but suspend purchases of released titles until seemingly legitimate claims can be addressed, and if someone does release a game that's a scam, early access or not, cut them off. They can no longer release a title on Steam, and no company which they work with in some sort of management role can do so either. Come down hard on the people who scam the system, but let the users police and gather evidence for this stuff. They will do it faster and better than Valve ever could. Will some get burned by making bad purchases? Probably. But they'll have a recourse. And if you're buying a game sight unseen that you've never heard of before and no one has tried yet you need to be prepared to get burned no matter who made it. That's not being an apologist for a lack of quality control, that's just good old fashioned common sense because no matter how much control is in place some people are going to buy things they don't like. But if they can take action against the people that actually scam others, that counts for a lot. And it goes without saying that anyone who's game is pulled for being a scam should have refunds given to their customers and not be given a dime from the sales.
People who keep going on about a need for quality control on Steam are only half way there. Steam shouldn't be more closed off, it should be as open as possible, but that openness needs to cut both ways. If anyone can get on, then customers need ways to get the bad ones removed quickly and decisively to protect themselves and others. The open system can't just go one way, but just because there are problems now does not mean that we should go the other way in trying to solve them. That's the wrong view to take. We need to be open to trying new pathways to make this stuff happen and recognize that there will be some growing pains along the way. Too many people are too quick to jump straight to "this isn't working so let's go back to doing what consoles do even though that model has been failing them and they're moving away from it now too." Sometimes when given the choice between going forward or turning back, you just have to say fuck it and go left instead.