[GRASS5] v.in.dwg license problem

[Roger Bivand]

Just a very short comment. While understanding Radim isn't difficult, and
we are all very grateful for hard work, commitment, and real
break-throughs for the future of GRASS, we should listen to Bernhard.
  
Comparing situations, someone working in a commercial, closed-source
business is typically very impeded in contributing anything to anyone,
even in time not being paid for by his/her employer, usually any
publication requires approval from various corporate "minders". 

We are fortunate to benefit from freedoms, many are not so fortunate, and
being careful about licensing conditions does matter - CRAN (the R
archive) is very careful about this, although for contributed packages,
the package maintainer is the one who has to answer for conditions being
met and protected (for instance conditions that code not be used for 
commercial purposes, very difficult to pursue, is contract research a 
commercial activity when done in a firm, and not in a university?). 

Maybe emails on topics like this could go through a cooling-off period 
(not serious, but you know what I mean)? We are so very dependent on 
keeping momentum and goodwill that a clear consensus ought to be in 
everybody's interests, which essentially means GPL for core code and 
programs within GRASS. Something like that, anyway. Licenses do matter.

Roger


On Thu, 15 May 2003, Bernhard Reiter wrote:

> On Thu, May 15, 2003 at 04:41:29PM +0200, Radim Blazek wrote:
> > On Wednesday 14 May 2003 01:33 pm, Bernhard Reiter wrote:
> > ....
> > > On Wed, May 14, 2003 at 10:45:17PM +1200, H Bowman wrote:
> > > > I suppose the worry is that if it is included in the main source
> > > > distribution, an over zealous packager might build it into a binary at
> > > > some point.
> > 
> > Surely, over zealous people can make fatal mistakes, but I don't 
> > think it is a good reason to complicate other peoples lifes too much.
> 
> There is a long term PR issue at hand.
> I don't like to go around and tell people 
> that the core GRASS contains non GPL compatible licenses.
> 
> > > It should be separate from main GRASS core ....
> > 
> > I decided to keep it where it is. I added to source code of v.in.dwg 
> > exception to link it to 'OpenDWG toolkit', as recommended in
> > http://www.fsf.org/licenses/gpl-faq.html#WritingFSWithNFLibs
> > (thanks for hint Bernhard). Situation is also described in README.
> > During the compilation 'make' prints this warning:
> > 
> > -----------------------------------------------------------------------------
> > ------------------------------   WARNING  -----------------------------------
> > v.in.dwg is linked to proprietary library 'OpenDWG toolkit' (www.opendwg.org)
> > and to GRASS libraries released under GPL. As a consequence, it is not allowed
> > to distribute binary version of v.in.dwg.
> > If you realy want to build the binaries for v.in.dwg and loose possibility
> > to distribute GRASS binary package, type 'yes'<Enter>.
> > -----------------------------------------------------------------------------
> > 
> > if user doesn't enter 'yes', compilation fails.
> 
> That is an improvement.
> 
> > I am ready to move v.in.dwg to some other place, in case that majority
> > of DEVELOPERS contributing to 5.1, i.e. both Markus Neteler and 
> > Paul Kelly, express their wish to do that.
> > I can also do the same whenever in future.
> 
> Part of my contribution to GRASS5.1 is running the infrastructure
> and providing licensing and process counsel. 
> Both I continue to do and I consider it to be important.
> So why would't I have a vote in this? 
> 
> > PS: I must say, that when I ever asked for help with 5.1, I did not
> > have in mind this kind of obstructions ....
> 
> I already help at least a bit as explained above,
> thus you probably don't accuse me but somebody else.
> 
> As you can see I am taking some interesting in what you do,
> but keeping GRASS' code clean of licensing traps is important,
> though it might be seen as an obstacle from your point of view.
> 
> Be reassured that I think that you are doing a great work on GRASS5.1,
> but just because you are the most active developer on this
> does not mean you should not be criticised!
> 
> 	Bernhard
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand at nhh.no