Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Wed, 19 Dec 2001 10:00:18 -0600

john sinclair wrote:
> 
> So, I believe that the structure of edm complies with this idea.

But depending on how you answer the linking question it may not be legal
to actually compile and/or run a program that uses both Channel Access and
the Qt Free Edition, because to do so requires you to create a derivative
work that combines the two, either in the executable itself (containing
information from the headers and library stubs for dynamic linking) or at
least the final in-memory image of the process with both libraries loaded.

> > > ... The ordinary
> > > General Public License therefore permits such linking only if the
> > > entire combination fits its criteria of freedom.
> 
> But, do you think that the license statement should be LGPL, BSD,
> or perhaps something else? I do feel the GPL "linking" issue is subject
> to some degree of interpretation.

I'm not sure which license statement you're talking about here.  You can
pick any license you like for your code, but you might need to discuss and
agree this with your lab's legal department first.  However they won't be
able to tell you whether you can actually run the resulting program
legally because no court has ruled on the linking question.  We are trying
to get our legal department to approve a BSD-like (ideally straight BSD)
license for unbundled EPICS modules, but base is a different issue because
of the export regulations.

To some extent I'm playing Devils Advocate here because given the LGPL
wording I think it's clear what Richard Stallman's *intent* was for the
GPL - he doesn't want people to be able to take advantage of any GPL code
in non-Free Software.  I agree that the GPL "linking" issue is open to
interpretation, but personally I would probably try to avoid doing so even
if the courts were to rule it legal, out of respect for RMS.

- Andrew
-- 
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away.
- Antoine de Saint-Exupery

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System