Posts Tagged‘lars gustavsson’

The Work Required For Linux’s “One Killer App”.

It shouldn’t be a secret that I’m something of a Windows guy as I’ve essentially made my career in IT out of their products as well as it being my preferred gaming platform. It’s not that I have anything against the alternatives per se, more that there really isn’t another platform capable of doing all the things that Windows can do currently. If I was to stay on the PC platform my only alternatives are OSX and Linux and the former requires an exorbitant investment in hardware which I, as someone who builds his own PCs, am quite adverse to. Whilst the merits of Linux are vast it’s still got a long way to go before I can consider it on par with Windows, even if there’s been significant progress of late.

Battlefield 4Indeed it’s gotten to the point where some industry veterans, like DICE creative director¬†Lars Gustavsson, have gone on record saying that Linux is only one killer app away from seeing explosive growth. There’s definitely been an escalating amount of investment in the platform over the past couple years, mostly in the indie space thanks to things like the Humble Bundles, however Linux gamers are still only make up a tiny minority, on the order of 2% (even on the current champion of the platform, Steam). With that in mind whilst I agree with Gustavsson’s point that Linux is only one killer application away from seeing a lot of growth that statement hides the significant amount of work required to make that happen.

For starters hidden within that 2% of users is an incredible amount of diversity in terms of which distribution they’re using. This is less of a problem than it used to be since a couple base distributions now power the majority of the Linux world (Debian, Red Hat, etc.) however it still presents a challenge that needs to be overcome. I think (and feel free to correct me on this) that the majority of this stems from a driver level where there’s a huge amount of fragmentation thanks to either a philosophical standpoint, I.E. no non-free software so binary blobs are out, or simply because manufacturers aren’t willing to provide that level of support to Linux users.

There also needs to be a critical mass of users in order for it to become attractive for bigger developers to want to support Linux as a platform. Now there’s some potential for this to happen with SteamOS and SteamMachines¬†although it will still take some time for that to permeate. It will be interesting to see if SteamOS users will translate into Linux users over time or if they’ll remain as users of the platform, just like current console gamers are. There will need to be significant traction for this critical mass to be reached however as even OSX, which commands around 6% of the PC gaming market, still hasn’t managed to reach that level where big developers and publishers see it as a priority platform to support. What that critical mass is however I am not sure of but it’s definitely far above the current level which Linux reaches to currently.

I’m not saying that any of this isn’t possible, it most certainly is thanks to the mountains of work done by dozens of companies, just that “One Killer App” is so much harder to achieve than what the soundbite makes it sound like. Personally if it happened I’d be pretty excited about it as more competition means better products for the end consumer, even if I don’t completely agree with some of the motivations that are driving it. It’s for that reason that I signed up for the Steam hardware beta as I’d love to see the PC platform make a resurgence as the king of gaming regardless of the software platform it runs.