Let’s Argue: Full Steam Ahead

Greenlight, the Marketplace, the Workshop, Trading Cards… Valve’s digital distribution platform Steam has been changing a lot, and changing fast. As the unspoken leader of the digital future, what Valve decides to do with Steam is going to have some pretty massive effects on this hobby of ours. So what should they do with all that power? Ben and I talked about it, and here’s what we thought:

What do you think should be Steam’s short-term priorities (next six months)?


As Steam grows, its servers have inevitably grown in size with it. It takes a lot of bandwidth and muscle to handle millions of players at once, and it’s admirable that they manage to keep the servers running for a sizable crowd. However, events like the Steam summer sale, releases of hotly-anticipated titles and even the odd Team Fortress 2 update bring the servers to a screeching halt and disrupt the service for everyone.

It’s undoubtedly a road filled with substantial fees, but Steam’s somewhat-frequent server downtimes have to go. Many users have undoubtedly poured thousands into their Steam libraries, and they expect to access their games 24/7 (or as close as possible). Valve can afford to be closer to that golden standard by buying servers at a faster rate. With enough servers, Steam could even eliminate the possibilities of downtime by shifting users from one server to another whenever maintenance is required.

Reliable servers build trust in the service. For a service that’s often treated with more respect than the entire Internet, adding a few notches to Steam’s credibility would further cement their position as the dominant empire in the PC world.


It’s worth mentioning, I think, how far Steam’s servers had come. I’m not even talking about those first couple of years where everything was terrible forever- even last year the chat servers were painfully unreliable, going up and down at a moment’s notice. It’s true that big events work the store over these days, but they’ve gotten so much better, and I fully expect that improvement to continue. There isn’t much cause to worry.

I think one of the things that Steam really needs to take care of, in the short term, is some kind of introduction to… well, everything. I know how Steam cards work because I heard a podcast talk about it. I know that I can sell TF2 items on the marketplace because of forums. Steam itself has told me basically nothing about the features it has added. They are just strange options on drop-down menus. For power users, that works out okay- we tend to do our homework. But what do you call a feature that the user doesn’t know about?

Uhm. Not a feature? That’s not very pithy. But it’s a dumb thing to do.


What do you think Steam’s long term goals should be?


Long term, Greenlight needs to go, and a better system for approving games needs to be implemented. Valve cried before about how they only had one person handling approvals, and it was too much, so they had to create the miserably ill-considered Greenlight. Uh, Valve? I don’t know if anyone told you, but you have tons of money. Hire people to approve games if you’ve got too much work. Goodness, you hired an economist just because. This hiring will actually make good financial sense.

The Steam Store needs a redesign in my opinion as well- it’s way too cluttered and chaotic. Push more developers to take advantage of controller support, Steam Cloud, and other standardizations that come with your service. Oh, and give me a way to add to my Steam wallet while I’m buying games, not making them separate transactions. Small stuff, I know, but it’s irritating.

It’s not that Steam has any glaring issues long-term beyond Greenlight, just a bunch of little things that are just not quite right.


All the issues you mentioned are indeed problematic, but they strike me as challenges that should be dealt with in the immediate future instead of putting them off for a while. Redefining ownership of Steam games is truly a long-term goal worth pursuing. The Steam library is one of the most convenient services in the world, yet it’s difficult to feel any sense of ownership when we can’t even trade games with our buddies or sell them to other users. This may be a controversial issue in the disc-based console world, but Steam is already poised to handle it in a way that leaves consumers, publishers and developers thoroughly satisfied.

Steam’s brand-new trading card system allows players to buy, sell and swap cards with one another. These cards act as free marketing for developers, but they also get paid when users sell, trade and craft the cards to create badges. The formula would need tweaking before it’s capable of managing games, but in addition to giving users the freedom to curate their libraries, it eliminates The GameStop Problem: instead of losing potential sales, developers and publishers get a fair cut for every user.

Sites like Green Man Gaming have attempted digital trade-ins before, but Steam has the knowledge and clout to make it work. If successful, they might even craft a model that solves the used games crisis on consoles!


How much of Valve’s efforts should be on improving Steam as a service, versus making and improving their games as a developer?


For Valve, game development and Steam management are one and the same. Their games serve as testbeds for new forms of monetization, community engagement and client features (cloud saves, achievements, etc.). This seems to negate the danger of stretching themselves thin, or focusing on one side of their empire to the detriment of the other.

Whether Valve should mess with their current game/platform balance depends on the games they want to create. Their madcap experimentation lends itself well to the ever-shifting world of multiplayer gaming, but there’s only so much they can try with single-player experiences like Portal 2 before they would have to move onto another project. Too much tinkering would run the risk of significantly changing or destroying the original experience, and that wouldn’t sit well with players.

Though we’d all love to see a new Half-Life, I’d rather Valve make the games they’re best suited to building. They pioneered the Hollywood-esque AAA campaign with two breathtaking Half-Life games, but we all keep coming back for Team Fortress 2 and Counter-Strike: Global Offensive, and those games are the best fit for their modern “mad scientist” attitude toward development.


I don’t know if I agree with Ben here. Certainly, working on new content for Dota 2 and TF2 provides new ideas for Steam, but it also takes resources and time. Your developers can only do one thing at once- and if their work on games gives you ideas for the Steam platform, wonderful, but you’d get more work done on the platform if you weren’t making those games. That’s just logic. And that goes double for making a whole new game- sure, you can innovate on the features, but you’re not gonna come up with new Steam features by making textures for grass.

So what should the balance be? I would say mostly work on Steam, do enough work on Dota 2 and TF2 to keep the community interested- mostly just approving and adding community content, which is simple enough (but still more involved than a lot of people think)- and maybe work on one new game to be released in a couple of years. I wouldn’t mind a Half-Life 3, but I also don’t see a huge need for it. As little developer work as you can do for people to stay engaged. Valve games are great games, but I just think your work as a digital distributor is more valuable.


Is there any value in the idea of a Steambox, and if so in what form?


The Steambox as a piece of hardware sold by Valve that only does Steam is a really stupid idea. Given that it would still have to be a PC to be compatible with the games and programs on Steam, you would gain absolutely nothing. It’d be a console in terms of single-minded focus, and a PC in terms of stability. The worst of both worlds. Garbage.

But there’s another way to make the Steambox work. Quite simply, make it a sticker. Like the nVidia bumper that they put on some games (“nVidia- The Way It’s Meant to Be Played”), Valve could license some PC manufacturers to be Steamboxes. They pay Valve a small fee, they get to put a sticker on a certain model of computer that says “Powered by Steam” or something. Not as exciting as a new platform, perhaps, but a much more practical move financially.

To add incentive for users, these licensed Steamboxes could have an easy rating to let you know at a glance what games they could run- this Dell M57334 is a Steambox Level 4, so it can play any games up to hardware level 4. Kind of like the Windows Experience Index, only actually used by people. You could also work with hardware manufacturers to ensure that any box licensed as a Steambox is guaranteed a level of stability with Steam. Small quality of life improvements. That’s the value in the idea of a Steambox.


I agree with Colin’s assessment of the Steambox as a standalone product. We’re seeing a host of hardware manufacturers attempting to bulldoze their way through the Android console market (a market whose existence is still up in the air) by making dedicated devices for their dedicated stores. It’s nothing more than a wild goose chase that will lead to many financial woes, and companies like Razer, Alienware and Sony would do well to stay out of it and stick with what they know. Steam as hardware is setting itself up to be an outdated box of wires discarded when a newer, shinier box arrives years later. Steam as software is immortal; if handled properly, existing as a package of bits and bytes leaves the platform flexible, malleable and future-proofed.

Consumers upgrade their PCs as time goes on, but they never have to upgrade or buy their way back into Steam. Their lack of dependency on physical hardware lets them skip the delicate dance of convincing fans that they need to buy their latest expensive, magical box. Owning Steam is as simple as owning a computer, and their range of free-to-play titles further lowers the barrier for entry. Entering the hardware business would put them in a state of unnecessary vulnerability

That said, a diagnostic system capable of examining a Steam user’s hardware and spitting out a compatibility report would go a long way toward assuaging lingering fears. One of the biggest problems with the PC gaming market is the uncertainty; is my two year old tower capable of playing Metro: Last Light smoothly, or do I have to invest in a new video card? Giving users a simple “Yes/No” and expanding on what might need improvement would build confidence in making purchases through Steam.

Colin’s “Level” system is a convenient way for consumers to decide which prebuilt PC suits them, though they will have to pay close attention to their numbering system as hardware continues to improve and hardware requirements soar. An NVIDIA situation where a 660 video card is worse than a 580 would only lead to widespread confusion and panic.