Subject: Re: Licenses and subterfuge
From: Alex Rousskov <rousskov@measurement-factory.com>
Date: Wed, 25 Feb 2004 13:34:23 -0700 (MST)


On Wed, 25 Feb 2004 jcowan@reutershealth.com wrote:

> However, a judge might decide that an instruction to the customer to
> "do your own linking" was a transparent attempt to bypass the GPL,
> and decide to treat it as a violation.  Saying "I can't do this and
> send you the result, but I can tell you how to do it instead" is a
> fishy thing to do.  IANAL, TINLA, but I'd say "Don't do that."

Agreed. A judge might decide anything. In practice, lots of software
is distributed with dynamic linking requirements (for _technical_
reasons) so, IMHO, it would be relatively difficult to win a claim
that this common practice is related to licensing and is somehow fishy
by its own nature.

Of course, it would be foolish to provoke a law suite by attacking
viral goals in software documentation. Documentation should focus on
technical issues. For example, it would be foolish to write:

  "Do your best to find a FSF FooLib to link with our software.
   While other Foo API libraries might work, our software has been
   designed to tightly integrate with FooLib and makes extensive
   use of FooLib extensions and internal features that are not a part
   of the standard FOO API. If you are not using FSF FooLib, we
   will not be able to offer any support."

One could write:

  "Our software can be linked with any library supporting Foo
  API. FooLib library supports Foo API on Linux and BarLib
  library supports Foo API on NetBSD. Users also report
  success with CygFoo.dll library on Windows, but FooLib may be
  faster and more stable. Ported versions of those libraries or
  other Foo API libraries may be available for your OS. Known
  compatibility problems are available by searching our bug database
  at ..."

Alex.
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3