Users of the Philips webcam driver known as 'pwc' will be relieved to the learn that the driver will continue to be part of the Linux kernel. In an earlier dispute over the removal of a special hook only used to load its binary-only counterpart the author requested that the entire driver be removed, a request agreed to by Linux creator Linus Torvalds [story].

Alan Cox [interview], evidently returned from his one year sabbatical [story], disagreed with this decision, and to make a point suggested that his own extensive contributions should also be removed from the kernel, "point made ? We can't go around throwing out drivers because the author had a tantrum." Regarding this specific instance, Alan explained, "he is not sole author. Large parts of the code are based on other authors work and simply copied from the standard framework. Please put back the version without the hooks. It is useful to all sorts of people in that form."

Linus quickly replied, "I'm disgusted by how many people have been complaining, yet when I ask people to step up and actually _do_ something about it, people suddenly become very quiet, or continue complaining about it ignoring the fundamental issue." He went on to ask Alan if perhaps he was willing to step up and become the new maintainer, to which Alan agreed. Shortly following this exchange, Greg KH noted that someone else was also already working with the original author to get a patch together to restore the driver back into the kernel, minus the problematic hook. Thus, in one way or another, the pwc driver is alive and kicking.

From: Alan Cox [email blocked]
To: Linus Torvalds [email blocked]
Subject: Re: Termination of the Philips Webcam Driver (pwc)
Date: 	Sun, 29 Aug 2004 15:36:22 +0100

On Gwe, 2004-08-27 at 01:03, Linus Torvalds wrote:
> Yes and no. From a legal standpoint you're right. However, we should also 
> be polite. If he's the sole author, and he asks for it, I think it's 
> reasonable to honor his wishes.

He is not sole author. Large parts of the code are based on other
authors work and simply copied from the standard framework. Please put
back the version without the hooks. It is useful to all sorts of people
in that form.

When the author GPL'd it he gave up his rights to remove it. Expecting
people to clean-room reverse engineer GPL source is a joke.

Alan



From: Linus Torvalds [email blocked]
Subject: Re: Termination of the Philips Webcam Driver (pwc)
Date: 	Sun, 29 Aug 2004 11:09:22 -0700 (PDT)



On Sun, 29 Aug 2004, Alan Cox wrote:
> 
> He is not sole author. Large parts of the code are based on other
> authors work and simply copied from the standard framework. Please put
> back the version without the hooks. It is useful to all sorts of people
> in that form.

Are you willing to stand up for that and be the maintainer for it?

I'm disgusted by how many people have been complaining, yet when I ask 
people to step up and actually _do_ something about it, people suddenly 
become very quiet, or continue complaining about it ignoring the 
fundamental issue.

Everybody (including you, Alan, so don't go hoity-toity on us) has
apparently totally ignored my calls for a new maintainer, and asking the 
people involved who wrote parts of the driver for their input. I quote an 
email from me:

	Date: Fri, 27 Aug 2004 11:13:09 -0700 (PDT)
	From: Linus Torvalds [email blocked]
	Cc: Christoph Hellwig [email blocked],
	    Craig Milo Rogers [email blocked],
	    Kernel Mailing List [email blocked],
 [email blocked]
	Subject: Re: Termination of the Philips Webcam Driver (pwc)

	On Fri, 27 Aug 2004, Xavier Bestel wrote:
	> 
	> What if someone steps up and want to maintain and extend this piece of
	> code ? Will you forbid him (as in "not in my tree") ?

	I'd suggest you contact the people who have worked on that driver (there's 
	certainly people outside of nemosoft, at least according to the 
	changelogs) and see what they feel like and try to gauge how much they 
	were part of driver development. 

	...

I've got _lots_ of emails in my mailbox complaining.

I don't have a _single_ one actually responding for my calls to actually 
_do_ something about the driver.

Until people turn from whiners to doers, nothing will happen.

			Linus



From: Alan Cox [email blocked]
Subject: Re: Termination of the Philips Webcam Driver (pwc)
Date: 	Sun, 29 Aug 2004 19:57:14 +0100

On Sul, 2004-08-29 at 19:09, Linus Torvalds wrote:
> Are you willing to stand up for that and be the maintainer for it?

Yes.



From: Greg KH [email blocked]
Subject: Re: Termination of the Philips Webcam Driver (pwc)
Date: 	Sun, 29 Aug 2004 12:40:55 -0700

On Sun, Aug 29, 2004 at 11:09:22AM -0700, Linus Torvalds wrote:
> 
> 
> On Sun, 29 Aug 2004, Alan Cox wrote:
> > 
> > He is not sole author. Large parts of the code are based on other
> > authors work and simply copied from the standard framework. Please put
> > back the version without the hooks. It is useful to all sorts of people
> > in that form.
> 
> Are you willing to stand up for that and be the maintainer for it?

Someone has contacted me and Nemosoft and is willing to be the
maintainer, so it looks like this will happen.  That person is currently
working with Nemosoft to hash out a few issues and should soon be
sending a patch in to add the driver back, minus the hook.

At least that's what they say they will do :)

thanks,

greg k-h



From: Alan Cox [email blocked]
Subject: Re: [linux-usb-devel] Re: Summarizing the PWC driver questions/answers
Date: 	Sun, 29 Aug 2004 14:40:00 +0100

On Sad, 2004-08-28 at 00:13, Oliver Neukum wrote:
> Keeping drivers against the wishes of the authors in the tree would
> be very troubling for the future. I can assure you that no maintainer
> will lightly pull a driver in this way.

Then the kernel community is no longer fit to use my code. So you should
remove everything I've written from Linus kernel too. I'll maintain my
own kernel.

Oh gosh, look I've just crippled Linus tree and stolen his project.
Thats *WHY* you can't just rip drivers out. A license was granted, for
ever. You can certainly remove him from maintainers, and if he insists
from the author credits.

Alan


From: Alan Cox [email blocked]
Subject: Re: pwc+pwcx is not illegal
Date: 	Sun, 29 Aug 2004 15:00:19 +0100

On Gwe, 2004-08-27 at 20:29, Linus Torvalds wrote:
> So stop whining about it. The driver got removed because the author asked 
> for it. 

Please put it back, minus the hooks so the rest of the world can use it.
If not please remove every line of code I've even written because I
don't like the new attitude .. so ner..

Point made ? We can't go around throwing out drivers because the author
had a tantrum. Its also trivial to move the decompressor to user space
where it should be anyway. Similarly the driver is useful without the
binary stuff.

Or do we need a -ac tree again where this time -ac is "added camera" ;)

Alan



From: Nemosoft Unv. [email blocked]
Subject: Re: pwc+pwcx is not illegal
Date: 	Sun, 29 Aug 2004 18:33:37 +0200

Hello,

On Sunday 29 August 2004 16:00, Alan Cox wrote:
> On Gwe, 2004-08-27 at 20:29, Linus Torvalds wrote:
> > So stop whining about it. The driver got removed because the author
> > asked for it.
>
> Please put it back, minus the hooks so the rest of the world can use it.

No, don't! There is one very practial reason for that: the utter confusion 
it will cause when suddenly PWCX cannot be loaded anymore, because users 
will assume that since PWC is in the kernel, PWCX will work too. I really 
would not like to be at the receiving end of the support mailbox when 2.6.9 
comes out with such a crippled version of PWC.

That's one of the reasons I requested PWC to be removed. For me, it's also a 
matter of quality: what good is a half-baked driver in the kernel when you 
need to patch it first to get it working fully again? I don't want my name 
attached to that.

> If not please remove every line of code I've even written because I
> don't like the new attitude .. so ner..
>
> Point made ? We can't go around throwing out drivers because the author
> had a tantrum. 

I'm not having a tantrum. If it is, it has been one in the making for 3 
years.

> Its also trivial to move the decompressor to user space 
> where it should be anyway. 

*sigh* As I have been saying a 100 times before, it is illogical, cumbersome 
for both users and developers, and will probably take a very long time to 
adopt (notwithstanding V4L2 [*]). 

I mean, I still remember when the YUV->RGB conversion code was snipped from 
PWC when I supplied it for inclusing in the kernel, back in 2001. It took a 
long, long time for webcam tools to adjust their code to check for the YUV 
palette and do the conversion themselves, and _to_this_very_day_ I'm 
getting mails about programs who still don't get it right.

*IF* there was a commonly accepted video "middle-layer", this would not pose 

much of a problem. But there is no such thing yet.

(maybe that's something for a 2.7 kernel...)

> Similarly the driver is useful without the binary stuff.

True. But judging from the mails I have received the last couple of days,
people don't really care about the binary stuff, as long as it works. They 
want to use the cam to its full potential, so PWCX is more or less a 
necessity. However, there's has now been added an extra hurdle in getting 
it work, for reasons I find questionable, and really, 3 years too late.

Seriously, this probably would not have happened if, back in 2001, the 
driver was rejected on the basis of this hook (you were there, Alan...) I 
never made a secret of it, it has been in the driver from day 1 and its 
purpose was clearly spelled out. If it had been rejected, I would probably 
have just switched to '3rd party module' mode and maintained it outside the 
kernel indefinetely. I would not have liked it, but it would have been 
acceptable.

Another acceptable solution would have been, if after the 'discovery' of the 
hook, Greg or anybody else had said: "Look, we really don't want this kind 
of thing in the kernel. However, since we're a bit late to react, we'll 
leave it in the 2.4 and 2.6 series, but versions beyond that (2.7-devel, 
etc) will not have PWC included in this form. In the mean time, we're 

asking you to think of a solution". Chances are the situation would have 
been fully resolved before that (and I mean fully *hint*).

> Or do we need a -ac tree again where this time -ac is "added camera" ;)

*lol* The code is still floating around on the Net, so nobody's stopping 
you... 

 - Nemosoft


[*] Some advice: if you really want to speed up V4L2 adoption by video 
tools, start disabling V4L1 in the kernel... 



From: Alan Cox [email blocked]
Subject: Re: pwc+pwcx is not illegal
Date: 	Sun, 29 Aug 2004 16:42:24 +0100

On Sul, 2004-08-29 at 17:33, Nemosoft Unv. wrote:
> That's one of the reasons I requested PWC to be removed. For me, it's also a 
> matter of quality: what good is a half-baked driver in the kernel when you 
> need to patch it first to get it working fully again? I don't want my name 
> attached to that.

It works very well for some users without that code. The raw pass
through for the compressed bitstreams solved the problems for the rest.
You appear to be seeking to hurt your userbase for your own ends. Thats
not pleasant behaviour.  I can more than understand
"take my name off it, make it clear its nothing to do with me".

> > Its also trivial to move the decompressor to user space 
> > where it should be anyway. 
> 
> *sigh* As I have been saying a 100 times before, it is illogical, cumbersome 
> for both users and developers, and will probably take a very long time to 
> adopt (notwithstanding V4L2 [*]). 

Video4linux has -always- specified decompressors in user space. This was
pointed out ages ago. V4L2 rationalised it even more clearly.

> *IF* there was a commonly accepted video "middle-layer", this would not pose 
> much of a problem. But there is no such thing yet.
> 
> (maybe that's something for a 2.7 kernel...)

No its for userspace. Just add it to the relevant video frameworks.

> Seriously, this probably would not have happened if, back in 2001, the 
> driver was rejected on the basis of this hook (you were there, Alan...) I 
> never made a secret of it, it has been in the driver from day 1 and its 
> purpose was clearly spelled out. If it had been rejected, I would probably 
> have just switched to '3rd party module' mode and maintained it outside the 
> kernel indefinetely. I would not have liked it, but it would have been 
> acceptable.

Back in 2001 I was saying that this was broken and it belonged in user
space.

> of thing in the kernel. However, since we're a bit late to react, we'll 
> leave it in the 2.4 and 2.6 series, but versions beyond that (2.7-devel, 
> etc) will not have PWC included in this form. In the mean time, we're 
> asking you to think of a solution". Chances are the situation would have 
> been fully resolved before that (and I mean fully *hint*).

There isn't a plan to have a 2.7 development tree but to do gradual
development until something major comes up. That makes the suggestion
rather more tricky - as does the legal question.

• Archive of above thread
• KernelTrap interview with Alan Cox