[License-discuss] Linking question

[Lawrence Rosen]

Bruce Perens wrote:
> I know of multiple parties who have settled by Ceasing to manufacture
> the product or Releasing a new version of their binary software along
> with complete and corresponding source code.

Interesting. I've been told that you have a copyright interest in Busybox. Then would you please help me advise an imaginary client with the following fact pattern:

1. My imaginary client is shipping over a million proprietary wireless devices that include, somewhere in the code, Busybox software. No question of "distribution" in this example!

2. The wireless devices ship with statically linked software, in-so-far as anything can be locked firmly in place in memory without padlock and key. No question about "static linking" in this example!

3. My client discloses the presence of Busybox in his devices and offers the source code to that Busybox software (along with any modifications he may have made to "standard" Busybox software) on his website along with copies of the appropriate GPL license. No question about disclosure and compliance with *those* aspects of the GPL in this example!

Is anything else required under the GPL or by the Busybox copyright owners? Specifically, is any of my client's proprietary software subject to disclosure? Must my client help anyone -- through product documentation or the disclosure of his proprietary code that he has purposely linked statically to Busybox -- to replace or upgrade Busybox itself in those millions of distributed proprietary wireless devices? [Please ignore the local rules about permitted forms of reverse engineering.]

/Larry


> -----Original Message-----
> From: Bruce Perens [mailto:bruce at perens.com]
> Sent: Friday, March 02, 2012 10:06 AM
> To: license-discuss at opensource.org
> Cc: Lawrence Rosen
> Subject: Re: [License-discuss] Linking question
> 
> Larry,
> 
> I know of multiple parties who have settled by Ceasing to manufacture
> the product or Releasing a new version of their binary software along
> with complete and corresponding source code.
> 
> Some of these cases involved replacing binaries that were static linked
> to LGPL software with binaries that were dynamic linked, to satisfy the
> requirement that the two pieces be seperable so that the LGPL piece
> could be modified by the end-user.
> 
> In all of the cases that I know of, the parties had available to them
> the strategy of asserting that their work was not derivative, and
> providing the GPL source only without reference to the proprietary
> components of their products. Surely, many of them have read your book.
> However, none of these parties chose to do so.
> 
> You may well be right. Nobody, however, has found it economical to
> invest the resources to prove that in court.
> 
>      Thanks
> 
>      Bruce
> 
>   On 03/02/2012 09:16 AM, Lawrence Rosen wrote:
> > NOBODY has ever settled such a case such that their code must be
> released.