The ongoing Philips webcam driver saga [LWN.net]

Linus has just merged a patch from Alan Cox removing some of the new decompression code from the Philips webcam driver. "The original pwc author raised some questions about the reverse engineering of the decompressor algorithms used in the pwc driver. Having done some detailed investigation it appears those concerns that clean room policy was not followed are reasonable." The hope, at this point, is to merge an improved version of the driver in 2.6.13 which will support (properly reverse-engineered) decompression modules in user space.

to post comments

The ongoing Philips webcam driver saga

Posted May 27, 2005 21:12 UTC (Fri) by clump (subscriber, #27801) [Link] (1 responses)

This issue is very sad to me. I do not want assign blame to any party, however I really can't see anything wrong with Greg KH removing a proprietary, single-use hook. That, of course, was the reason Nemosoft was so unhappy.

From: http://lwn.net/Articles/134703/ regarding PWCX:

Anyway, I'll inform my contacts at Philips tomorrow. I don't know how they will react; maybe they'll go nuts, maybe they'll let it pass quitely; it's hard to tell.

This attitude is what concerns me. The original author need not have behaved this way, and by his own reasoning regarding his NDA could have worked with Philips to make the hook unnecessary. If what he later said about not being able to disclose his work is true, he still likely would have been the most influential person to speak and free the code from Philips. I am glad that Alan was able to talk to Philips. Hopefully something good will come of it.

The ongoing Philips webcam driver saga

Posted Jun 2, 2005 13:17 UTC (Thu) by coolian (guest, #14818) [Link]

From the get-go, it has seemed that the guy in this story has pushed to extend his ill-gotten 15 minutes of fame. He likes this role.

The ongoing Philips webcam driver saga

Posted Jun 2, 2005 5:47 UTC (Thu) by eru (subscriber, #2753) [Link] (2 responses)

The removed decompressor appeared to mostly consists of tables of arbitrary-looking numbers. I suspect that it going to be impossible to reverse-engineer in a clean-room fashion. Even if someone somehow succeeded, the result would be tables of similar numbers, open to unfounded accusations of copying. It seems to me Linux will never properly support this hardware unless Philips itself provides the code with GPL-compatible license, or at least a detailed description of the algorithm.

The ongoing Philips webcam driver saga

Posted Jun 2, 2005 21:41 UTC (Thu) by giraffedata (guest, #1954) [Link] (1 responses)

Are unfounded accusations of copying a problem?

In today's story, we see that Alan Cox thought the accusations were reasonable, and I don't think he would find unfounded accusations reasonable.

Assuming it's technically possible to do the reverse engineering, I'll bet it's possible to arrive at the same set of numbers in such a way as to convince Alan and his ilk that no copying took place. And that's enough for the code to survive.

The ongoing Philips webcam driver saga

Posted Jun 3, 2005 5:50 UTC (Fri) by eru (subscriber, #2753) [Link]

Are unfounded accusations of copying a problem?

They can be. It's called "FUD".

I hope you are right. But I think for that to happen more is needed than the finished code: The author of the clean-room version must be able to document how he independently arrived at the numbers.