[GRASS5] New info on openDWG

[Michael Barton]

Thanks John and the rest of you who reviewed the information I sent out
regarding openDWG libraries. It is clear that the two licenses are
incompatible, though it is too bad given that both GRASS and Open Design
Alliance share many overall objectives.

I guess I'd have to agree with John (below) that we probably need to port
v.in.dxf (and a v.out.dxf) to GRASS 5.7. The objective would be not to
maintain compatibility with a regularly changing AutoCAD--something that
Open Design Alliance has taken on--but a recognition that many vectgor maps
are created in CAD software or other GIS's, and dxf is a very common file
exchange format for these programs. This is the same reason that GRASS has
long had import-export abilities for ESRI file formats.

Paolo Cavallini has a very good suggestion--about creating a way to make
'plugins' for GRASS that could be distributed as binaries without the need
to compile all of GRASS from source. I think this would be very nice for
programs like the predictive modeling module that Benjamin Ducke is doing.
Then there would be less need to include it in the CVS, although it could be
added to the standard GRASS package if enough people thought it would  be
useful.

In the case of the v.in.dwg module that uses openDWG libraries, I'm not sure
how this could be done legitimately, however. At least I'm not sure that it
could use the GPL-licensed GRASS code in v.in.dwg to do so (again, one of
the folks who are more familiar with the details of the GPL license might
shed some light). Perhaps someone could create a stand-alone program
(without GRASS code) that exports dwg/dxf files to GRASS vector format (or
maybe GRASS ascii vector format if doing it in binary is a problem), then
distribute this as free software (not GPL)--in compliance with Open Design
Team requirements. 

Thanks again to all that responded.
Michael

On 8/27/04 6:16 AM, "jg-linux at triad.rr.com" <jg-linux at triad.rr.com> wrote:

> Hi Michael.  
> (1) I read Moritz and Jan's comments but I thought
> there was a more fundamental problem.  I thought if
> you linked a GPLed program to the ODA's program it
> violates the GPL because you must be able to supply
> source code for everything including the ODA's code,
> which you obviously can't do. The "problem" is with
> the GPL license and it was designed this way on
> purpose to encourage changes and additions to be public.
> 
> (2) I am of the opinion that we should eventually
> port v.in.dxf from 5.3 to 5.7 to have at least
> something that can be used in 5.7 for users who
> don't wish to sigh up with the Open Design Alliance.
> 
> (3) Which library do you use from the ODA?  The
> older C library or the newer C++ library.
> 
> (4) Which brings up my next thought.  It seems we
> are linking to more and more C++ code. We used to
> use v.in.sdts for which the US government had an
> SDTS C library.  They now have a C++ library.  Now
> we use r.in.gdal (or v.in.ogr) which is C++. And
> more people are trying QGIS which is C++. I assume
> we are going to stick to C for Grass.  Are there any
> issues here to be concerned about?
> 
> John Gillette 
> 

____________________
C. Michael Barton, Professor
School of Human Diversity and Social Change
PO Box 872402
Arizona State University
Tempe, AZ  85287-2402
USA

Phone: 480-965-6262
Fax: 480-965-7671
www: <www.public.asu.edu/~cmbarton>