Why VIA's Open Source Video Driver Is Missing A Few Things

A scenario for you: A company announces that it's going to offer open source drivers for its hardware from now on. Rejoicing ensues. Then the drivers themselves arrive, only to be missing things -- not enough to make them useless, but still frustrating. What happened?</p>

Serdar Yegulalp, Contributor

September 8, 2008

3 Min Read
InformationWeek logo in a gray background | InformationWeek

A scenario for you: A company announces that it's going to offer open source drivers for its hardware from now on. Rejoicing ensues. Then the drivers themselves arrive, only to be missing things -- not enough to make them useless, but still frustrating. What happened?

Here's the real-world version of that scenario. When VIA released an open source edition of the 2D Xorg driver for its integrated graphics chips, a few things weren't included. Among them were accelerated codec playback, 3-D acceleration, and TV-out functionality -- features that most everyone would like to use in some form or other. Naturally, people were concerned about this.

Harald Welte, a software developer who's done a lot of work with Linux and recently went to work for VIA as its official open source liaison, spoke up on his blog about these very questions. The answers he gave are strong reminders of just how difficult it is to open source anything legacy:

Usually there are some parts of the code, particularly for the 3-D driver, which cannot be disclosed due to either

  • parts of the source code are under a proprietary license from a third party

  • parts of the source code refer to technologies (e.g., Macrovision) which are subject to very strong NDA's by the licensor, which in turn prohibit the open documentation or distribution in source code form

In short, the problem here is probably not VIA per se, but a patent licensor (or a whole bunch of them) who would not exactly be thrilled to see trade secrets spilled out for all to dissect. What's ironic is that such worries are typically overblown beyond belief -- something Steve Mosher of OpenMoko pointed out when I spoke to him. There's no secret that can be kept indefinitely, and the whole point of technological innovation is not to sit on the same secrets forever but to keep moving forward.

I would also bet that it's not the loss of the secrets themselves that would be the most upsetting part, but perceptions by stockholders or partners (or even competitors) that the company can't be trusted to keep its mouth shut. Not everyone is a friend to open source, nor can they be expected to be one overnight.

Sadly, this may serve as an example for how future efforts to open source hardware drivers will unfold, especially when the open-sourcing process is spearheaded by the company itself. You're probably not going to get everything you want or need, no thanks to all the things no one ever told you about before the concept of open source hardware drivers came along.

And then there remains the larger question of what to do about such encumbered components. Leave them out entirely? Talk to the patent owners about the possibility of creating an open source implementation of their IP? Me, I'm just idealistic enough to think that a community-driven effort to engage an IP holder in some kind of discussion about open licensing would pay off.

We might as well learn how best to deal with all this on a small, manageable scale first, through companies like VIA and people like Harald Welte, before choking on much bigger fish -- like, say, Nvidia? When and if they ever do such a thing, that is.

About the Author

Serdar Yegulalp

Contributor

Follow Serdar Yegulalp and BYTE on Twitter and Google+:

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights