Continuing on the Valve theme…

Today’s controversial game development news is that Valve have decided to charge $100 to developers who wish to submit games for approval via their Greenlight system. This is not to cover costs, but to ‘cut down the noise in the system’, ie. people submitting other people’s games, unfinished games, or things that aren’t games at all – the problems that have blighted Greenlight‘s release a few days ago.  The fee itself goes to charity, not Valve.

This fee has led to a lot of complaints, and a lot of counter-argument against the complaints. The obvious objection is mainly the claim that $100 is a lot of money for independent developers. This is laughed off by a lot of people – it’s just a few hours’ pay, right? And this is certainly the case in North America and Western Europe. But when you look elsewhere, the situation is different – in Poland, the average disposable income is not even 20% of what it is in the USA, so it feels more like a $500 fee to them. So non-trivial monetary barriers like this risk will deter people from poorer countries from submitting, despite their game being as good as the competition. It’s also a lot of money for students, those out of work, or those working minimum wage jobs to pay the bills while developing games in the evenings. The Rampant Coyote mentions that “a lot of indies out there are practically charity cases themselves, living on Top Ramen and working from their cramped studio apartment.” Of course it is certainly possible to overcome adversities like these and raise $100 to spend on the chance of getting your game onto Steam, but this is still going to be a pretty high threshold for many people. Surely we want to encourage them, not discourage them.

Some argue that you shouldn’t expect to enter business without making an investment, and that you can’t expect the world to hand you things for free so that you get to do what you want. These statements are certainly true, but they also miss (or ignore) the point. If every game was purely made with a market in mind the world of gaming would be significantly diminished. Valve certainly aren’t a charity and nobody expects them to subsidise unpopular games, but unlike a normal publisher they are paying nothing up front and hosting the bits and bytes of a game that hardly anybody buys costs them next to nothing anyway – so why not select based on quality, rather than ability to invest? And nobody is expecting to be given things for free, or below cost price – just to not have an arbitrary fee imposed that doesn’t even buy them a service or product, as is the case with any fee that is immediately taken and given to charity like this. By way of analogy, when not programming games or writing silly blog posts, I’m a musician – and although I have no problem with spending money on buying a guitar for my performances (because that instrument costs money to make, and the guitar maker deserves compensation for his or her efforts), I would take issue with being asked to pay to submit my music to record labels, because it costs them virtually nothing to listen and if they like what they hear, they’ll split the proceeds with me anyway.

Comparisons with PS3 or Xbox360 developer costs are also somewhat flawed. When you enter into a development relationship with the likes of Sony or Microsoft you’re usually paying for an expensive and bespoke piece of hardware, access to internal libraries and tools, as well as getting a degree of documentation and support with all that. Besides, their costs are arguably deliberately high to keep smaller companies out – is this the future we want for PC gaming? Sony and Microsoft have their reasons for operating that way but it’s essential to keep an alternative for the kind of games that don’t suit massive console development budgets.

Still, surely any game that is worth publishing can raise $100 from somewhere, right? Maybe just $1 from each of 100 fans? But this brings us right back to why Greenlight even exists in the first place. Steam has the majority of the PC digital distribution market sewn up – between 50% and 80% of it, depending on who you ask – so developers are desperate to get on it. Lots of gamers are buying all their games from Steam and nowhere else, so if you’re not on it, it’s very hard to reach people, as most of them are simply not looking beyond Valve’s store.

But worse still, even the gamers who do look outside the Steam ecosystem will often demand developers put their games on Steam, often not realising (or caring) that the developer can’t just choose to put it up there.  As Sophie Houlden says, ‘ “I won’t get it if it’s not on steam” is such a common attitude I could spit and it’d hit someone who thinks that.’ There are good reasons why players prefer Steam, but that’s no consolation to the developers. They know that not being on Steam is a massive hindrance to their chance of survival, possibly being a larger factor than the relative quality of their game. Without being able to get on Steam, hardly anybody knows you exist, and that situation is getting worse every day. And if nobody knows you exist, reaching 100 people to get them to pay you $1 each is non-trivial. This is also why simply choosing to use another distributor like Desura is not the whole answer – the real problem is getting your name out to people, not sending data across the wire.

Some angry developers have suggested Valve should not have brought in Greenlight at all, and instead of having the voting system, they should be paying a team to properly check all submitted indie games. That swings the pendulum an unreasonable amount to the other side though – Valve shouldn’t have to incur all the cost and risk of getting good games onto the platform and weeding out the bad ones. We can hope they have good intentions, but again, Valve are not a charity. Ideally all sides would come to a compromise where Valve get to make a good profit from making as many good games available to people as possible, without any side having to take an unnecessary financial risk in the process.

So, what could Valve have done differently?

  1. Deposits, not payments. Here in the UK you need to pay a deposit if you want to stand for election as a Member of Parliament. This keeps the number of candidates down to mostly those who have a serious agenda. (Mostly.) But the important thing is that you get your money back if you do reasonably well, which reduces the monetary risk while still deterring time-wasters. Steam could implement a system like this, returning the deposit to any game that reaches a certain threshold of votes (presumably proportional to other games, or total votes cast over a period of time, etc). It would still be something of a barrier to the poorer developer, but it would be easier for them to get a loan from others in this case if there is a significant chance of them getting the cash back.
  2. Better organisation. There’s no excuse in 2012 to just have a handful of alphabetical lists sorted by crude genre categories, as is currently the case on Greenlight. With everything lumped in like that then of course noise will get in the way of the signal. Where is the “If you voted for This, then you might also like: This, This, and This” system to help you find similar games to the ones you like? Why is there no segregation of games that others are voting positively from those which others clearly think are not ready for publication? Why not have tags for certain aesthetics, such as 2D, 3D, pixel art, cartoon, etc? Why not show me games that are tagged similarly to the Steam games I already own, ie. in line with my interests? You don’t need to curate a collection aggressively if it’s easy for users to find what they want anyway. I don’t care that 99.99999% of the things on Amazon, Reddit, StackOverflow, or eBay are of no interest to me, because they do a good job of keeping that out of the way. (Incidentally, Valve made all the same mistakes with the Steam Workshop.)
  3. Improved moderation tools. YouTube is full of notoriously awful comments. Reddit is better, but not perfect. StackOverflow is almost all high quality, by comparison. All these sites are based around user-supplied content and user comments, yet some work better than others.  I won’t go into this too deeply but it appears that if you want to increase the signal to noise ratio, then it’s not sufficient to vote on something, but the layout of the site and the positioning of the articles and comments needs to change in response to that voting. When put that way, it’s almost too obvious to point out, but it’s still something Greenlight missed. Promote the highly upvoted ones. Demote or even hide the downvoted ones, unless I ask to show them. But at least let the crowdsourcing actually mean something.
  4. A more reasonable fee amount. If the figure had been $10 rather than $100 then probably nobody would care. Pretty much any payment would deter most of the time wasters, but smaller payments would have less of an adverse effect on developers with less cash to spare. A small non-refundable admin fee could be combined with the refundable deposit and few could argue with the fairness of that.

I think that with all that in mind, Greenlight could have avoided most of this negative publicity. But even if Greenlight does improve in future, independent developers and gamers who want a thriving ecosystem of diverse games will need to carefully consider the way the PC market is going and whether we’re just trading the old evils of retail for an equally bad set of problems.

Share this page:
  • email
  • Facebook
  • Twitter
  • StumbleUpon
  • Reddit
  • LinkedIn

Related Posts:

  • No Related Posts