[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: How long is it acceptable to leave *undistributable* files in the kernel package?

On Wed, Jun 16, 2004 at 04:49:27PM -0400, Michael Poole wrote:
> Andrew Suffield writes:
> > The compiled kernel is almost certainly a derivative of the firmware
> > included in it. A good lawyer might be able to get you out of
> > this. Debian can *not* afford to assume that it would win such a case,
> > not least because of a lack of funding for good lawyers.
> Anyone who wanted to go after Debian would have a hard time of it:
> They would have to prove not only that the combined kernel is a
> derived work, but also that there were no binary blobs in the kernel
> when they started contributing and that they still have standing to
> make a claim.

That probably wouldn't hold. They merely have to demonstrate that they
were not aware that this was a license violation - that's easy, they
can do it on their own testimony.

> Estoppel would bar a claim if the plaintiff first
> contributed code to a kernel that already had binary blob components.
> A merely decent lawyer may be able to invoke laches depending on how
> long an author was silent after the first binary blob was added to the
> kernel, but merely decent lawyers are not much cheaper than good
> lawyers.

These are stock defenses that are invoked as a matter of course, in
case something useful turns up during discovery, and which are
notoriously difficult to prove. You need to prove *malicious*
*intent*, not incompetence, for these defences to work. Not a good
idea to bet on them.

> By way of example, the DGRS driver firmware was added to Linux
> sometime before linux-2.0 was released in June 1996.  It stayed more
> than seven years.  That kind of delay in discovering alleged rights
> reminds me more of a certain Lindon, Utah-based company than of
> serious developers, and smacks of revisionist history.

We have not always been paying such close attention as we currently
do. Anybody on -legal can tell you that much. This is partially
because we're getting better at it, and partially because SCO has
demonstrated that it's important we be good at it.

Also, this sort of delay is far from unusual in this sort of
case. That's why the statute of limitations exists (we're nowhere near
the limit here; in the US it's three years from the more recent
infringement for civil copyright cases, five for criminal ones).

  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

Attachment: signature.asc
Description: Digital signature

Reply to: