Not that kind of steam machine. (Photo by Sir_Iwan )

Last week was a big week for Valve, creators of the popular Steam gaming platform. The firm announced the Steam Machine, a line of Linux-based gaming consoles slated to be released in 2014. Our own Hayley Tsukayama warned Friday that fragmentation could discourage developers from adopting the platform, making it hard for Valve to gain traction. But beerdovezeus is more optimistic:

They are essentially porting over the existing model of direct sales of computer games, a market they dominate, to a different screen (the TV).

While it's true that being Linux based will require developers to produce a version of their games that run on Linux (which I presume some do since Steam is already available on Linux) this hurdle seems small. Just because some nerd with a lot of time on his hands could theoretically build their own Steam powered robot doesn't mean they can expect the same support they'd get on a standard box. Besides, users on both PC and Linux can currently build their own hardware out of a myriad of compatible parts -- and it *already works* fine with Steam.

So, while it's true that current consoles are walled off systems that prevent users from accessing their proprietary hardware and OS, I don't see how they can hold off the Steam console for long. Once users can buy standard Steam hardware with a huge catalog of cheap and innovative games at launch, Steam OS will blow them out of the water.

jwilker2 adds:

While I agree with the majority of your comment, I would point out that game development on Linux is no trivial task although it would be significantly aided by the SteamBox. Linux has always lacked the unifying proprietary underpinnings Windows and closed console manufacturers have made the 'accepted' standards which allow those platforms to enforce a more formal stricture on their developers. Game console manufacturers have another notable advantage in 'freezing' their hardware in each release.

On Wednesday, I argued that "Patent thickets owned by IBM, Microsoft and other incumbent technology companies act as a tax on innovation, transferring wealth from today's innovators to the innovators of the past." bsr811 responded:

This is a criticism, not of patent thickets, but of the underlying principle of the patent system. A single patent (as well as a thicket) will transfer "wealth from today's innovators to the innovators of the past." The argument about whether that incentive is a good one is well-documented, and necessarily includes a discussion of the counter-incentive of disclosure, i.e., you can't have a patent if you refuse to disclose your discovery to the public. The legislative proposals you mention will have no effect on that debate, which rears its head more often in academia than in Congress.

The patent thicket problem is not so much about the fact that a patent grants the holder the right to exclude others from use, but about the fact that a company with a thousand patents, can force an unfair licensing negotiation by reserving for itself superior information about the risks and rewards. If both the patent holder and the accused infringer had perfect information about the patents being asserted, they could make a rational decision about the liability risk and come to an appropriate license cost. This is in fact what the Sun attorneys did with the 7 known patents IBM presented in the above anecdote. The problem arose when IBM then alluded to an immense collection of patents of which Sun had no information, but which IBM did have information. As a result of the disparate information available, IBM was able to strong-arm a license fee.

Thanks to all our readers for their thoughtful comments this week.