From: Roland Schulz (roland_at_utk.edu)
Date: Mon Jul 06 2009 - 04:19:52 CDT

On Sun, Jul 5, 2009 at 9:14 PM, John Stone <johns_at_ks.uiuc.edu> wrote:

>
> Hi,
>
> On Sat, Jul 04, 2009 at 09:13:37AM -0400, Roland Schulz wrote:
> > Hi,
> >
> > is our understanding correct that the 10% clause in VMD is not GPL
> > compatible?
>
> It is compatible in that you're free to use it for anything you want,
> but, that said, you would not be able to re-license it as GPL code.
> So while you're free to use it, it would remain governed under its
> original license, just as code using the BSD license, UIUC open source
> license, and other licenses remain under their original license.

We don't want to re-license in the sense of removing the current license but
adding GPL as an additional license, so that the whole package can be used
under one license. (see below for additional option)

My understand is that the UIUC license allows that. Is that correct? FSF
lists the UIUC as GPL compatible.

My understanding is that the VMD license does not allow that, because the
10% clause can't be combined with GPL. Since the right of the licensee to
combine any software is in conflict with the licensor right to prohibit
combining packaged resulting to more than 10%.

> Also, it is true that if ten different people each borrowed 10% of the
> code,
> that you'd be violating the original license to incorporate those
> ten different 10%s into a single code base. I would not see a problem
> with us making modules like vmddlopen.[c] Plugin*.[Ch] etc available
> under the UIUC open source license (BSD derivative) however, if that
> would make it easier for you to use the VMD plugin APIs in other
> packages.

I think this would help us a lot.

> There's nothing particularly magical or proprietary in
> the plugin-related source code in VMD, so I would be happy to
> accomodate this if it made things easier for you and others to use
> the VMD plugin APIs.

That would be great.

> I have already been planning on reengineering things a bit so that
> the interesting parts of PluginMgr.* were migrated to the plugin
> source tree and compiled into the static molfile_plugin.a so
> that the same plugin management code would be shared by
> VMD, NAMD, psfgen, cionize, and any other tools that wanted to
> make significant use of plugins, particularly the dynamically
> loaded shared library variety of plugins, which is the only case
> that's very tricky.

That would be an alternative. My current plan (proposed by Axel Kohlmeyer -
thanks!) is to integrate the PluginMgr in the source. Thus the VMD header or
libraries would not be needed at compile time. And if the Plugins (as shared
libraries) would be available at run-time additional formats could be read.
If the PluginMgr would be part of the static library, it would be needed at
compile time. Thus it might be little bit less convenient for the user, but
only barely so.

So in case you would prefer not to change the license for PuginMgr to UIUC
license or would prefer us not to distribute it in our source, we could use
the compile time option.

If this would address your needs, let me know and I would be happy
> to begin working on this with you.

Yes both option would be great. Let me know which one you prefer.

Thanks a lot

Roland

>
>
> Cheers,
> John Stone
> vmd_at_ks.uiuc.edu
>
>
> >
> > ---------- Forwarded message ----------
> > From: Erik Lindahl <lindahl_at_cbr.su.se>
> >
> > 1) We need to be able to distribute the code as GPL/LGPL (instead of a
> > combination of licenses). The 10% clause probably doesn't fly with
> this,
> > since the one could imagine 10 different people copy 10% each,
> distribute
> > as GPL, then somebody else (who gets it under strict GPL, where there
> > cannot be side clauses) merges it and has the original code under GPL.
> > ------------------------------------------------
> >
> > If so, is there a chance that you could release a PluginMgr equivalent
> > example code under a GPL compatible license (e.g. University of
> Illinois
> > Open Source License). That would allow other GPL software packages to
> > easily use the VMD plugins.
> >
> > Roland
> >
> > --
> > ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
> > 865-241-1537, ORNL PO BOX 2008 MS6309
>
> --
> NIH Resource for Macromolecular Modeling and Bioinformatics
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> WWW: http://www.ks.uiuc.edu/~johns/ <http://www.ks.uiuc.edu/%7Ejohns/>
> Fax: 217-244-6078
>

-- 
ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
865-241-1537, ORNL PO BOX 2008 MS6309