The Looming Risks of Stores Fragmentation

By

While on the whole things are pretty much getting better in many places, sometimes there are two steps back instead of one step forward. The looming disaster for PC gamers in general is store fragmentation. Up until not too long ago, PC gamers had it good: you had a few main poles to get your games, with fairly separate roles: the main one being Steam by far. Then, GOG for DRM-free, older titles and a few more recent indie titles, and The Witcher series. Add to that the Humble Store which mainly resells Steam keys while distributing a few DRM-free games as well. And if you wanted to play any Blizzard game you had to get on Battlenet, their exclusive store.

Right now there’s an abundance of stores to buy games from:

  • Steam (Valve games exclusives + everyone else)
  • GOG (No exclusives)
  • Humble Store (No exclusives)
  • Battlenet (exclusives from Blizzard)
  • Uplay for Ubi Soft Games
  • Epic Games Store (driven by Fortnite, with exclusives on their platform)
  • Discord (timed-exclusives and Nitro store)
  • Rockstar Social Club (no exclusive games but exclusive content)
  • Itch.io (free games and indies mostly)
  • A bunch of third party stores reselling Steam keys

Steam has been, so far, the dominant player and one of the most virtuous. They have not imposed any restriction on who can sell stuff on their platform. You are free to sell on Steam and on any other game store at the same time. They do not enforce any kind of exclusivity, apart from their own titles.

Now, Discord and Epic Games are trying to attract developers with better margins than Steam (providing around 90% instead of Steam’s estimated 70%) and gamers with exclusive titles that they cannot get anywhere else.

While I am all for competition (and Steam needs some to avoid becoming complacent and corrupted), this kind of behavior is reminiscent of the consoles world, and I loathe it with passion.

Just like I do not like when Sony “buys” a game to make it exclusive to the Playstation platform, I find it despicable to reproduce the same model for PC games distribution.

Of course, the PC model is different since you can technically install 10 different stores on your machine to play all the games you need. But… this is inconvenient as it requires the end user to keep x different applications that do the exact same thing, have x different accounts instead of a single one, and expose one’s payment credentials across x stores (thereby increasing the attack surface attached to one’s accounts). It makes it difficult to track what games one currently owns since it’s not centralized anymore. It makes it unclear what are the licenses of ownership attached to each store: does “owning” a game mean the exact same thing on every market? For games that have multiplayer, will you be able to play cross-stores? Or do you and your friends need to buy the game in the same place? The only upside is for users to get games a little cheaper - that’s a pretty bad deal across the board.

For developers, while there is the promise of additional revenue, it increases their workload as they have to post every new update to every single store, and support each store’s API - this is already an issue as we know GOG is quite often late to get the latest patches and contents compared to Steam - ironically this happens for Witcher 3 as well!

For Linux users the effects will be even worse, since the support of Linux on these stores will be either nonexistent or sub-par (GOG anyone?), so any more away from Steam is a potential loss for our library of games. It will increase the need for other solutions like Lutris.

This is not the first time I talk about it, but a much better model would be to dissociate the storage of game binaries from game stores. Let every dev/publisher store their own binaries, expose them to a shop API so that ANY store can decide to list them if they wish, with a unique content-source. This would have the following benefits:

  • Stores/launchers would compete on being user-friendly and providing additional features instead of focusing on growing their own libraries of games.
  • Users could move from one store to another and still find the same games that they purchased (since the user credentials would be recognized by the content-source) making it frictionless.
  • The work to draw updates from each content-source would be on the store’s part, not the developers’.

This would however mean that the cost of bandwidth would have to be dealt with by developers/publishers, but we can certainly imagine that such an ecosystem would lead to companies specializing in such hosting services for a monthly fee and based on the volumes of transfer, creating further competition in this area.

I don’t think that this model will emerge in the current situation, but this is something that we can keep in mind as a potential solution that does away with centralization while enhancing competition and freedom across the board.