[GRASS5] license question

GRASS GRASS developer
1

grass/src/general/init/license.txt
and
grass51/lib/init/license.txt

state:
"Geographic Resources Analysis Support System (GRASS) is Copyright,
1999-2004 by the GRASS Development Team, and licensed under terms of the
GNU General Public License (GPL)."

Should that specifically say GPL version 2?

The COPYING files do; they also say:

"You should have received a copy of the GNU General Public License
along with this program; if not, write to ..."

where is GPL.TXT in 5.7? or do we (meaning the primary authors of 5.7,
not me) want to figure out how to licence any new libraries before that
file is added to the archive?

?,
Hamish

2

On Sunday 21 March 2004 09:29, Hamish wrote:

grass/src/general/init/license.txt
and
grass51/lib/init/license.txt

state:
"Geographic Resources Analysis Support System (GRASS) is Copyright,
1999-2004 by the GRASS Development Team, and licensed under terms of the
GNU General Public License (GPL)."

Should that specifically say GPL version 2?

The COPYING files do;

I dont know if it is important in license.txt, but I have changed it
in grass51 to "GNU General Public License (GPL) version 2."

COPYING says "either version 2 of the License, or (at your option) any later version"
and many source files "GNU General Public License (>=v2)"
(recommended in grass/SUBMITTING).

I think that to use a license which nobody knows is nonsense.
If the next version will be compatible, people can
change the license in future - no need to give permission now.
I don't think that we want to give now permission to relicense
GRASS, using a license non compatible with current license.

they also say:

"You should have received a copy of the GNU General Public License
along with this program; if not, write to ..."

where is GPL.TXT in 5.7?

I have submitted GPL.TXT to CVS.

or do we (meaning the primary authors of 5.7,
not me) want to figure out how to licence any new libraries before that
file is added to the archive?

These days I was inspecting the possibility to change the license
of vector libraries from GPL to LGPL, it brings me to 5.0 history
in general. I'll send some results and questions in separate mail.

Radim

3

On Wed, Apr 28, 2004 at 05:03:01PM +0200, Radim Blazek wrote:

On Sunday 21 March 2004 09:29, Hamish wrote:
> grass/src/general/init/license.txt
> and
> grass51/lib/init/license.txt
>
> state:
> "Geographic Resources Analysis Support System (GRASS) is Copyright,
> 1999-2004 by the GRASS Development Team, and licensed under terms of the
> GNU General Public License (GPL)."
>
> Should that specifically say GPL version 2?

Probably verson 2 or later as recommended.

> The COPYING files do;

I dont know if it is important in license.txt, but I have changed it
in grass51 to "GNU General Public License (GPL) version 2."

COPYING says "either version 2 of the License, or (at your option) any later version"
and many source files "GNU General Public License (>=v2)"
(recommended in grass/SUBMITTING).

I think that to use a license which nobody knows is nonsense.

Your statement is not well thought through.

This option for a later license is necessary to be able to fix legal holes.
Note that the new license can only be in the spirit of the
reamble of the old one, so you know what the new license will say
in general, though some legalise might change.
In addition everybody can continue to use the software under the old license.

Some other information on this:
http://www.fsf.org/licenses/gpl-faq.html#VersionTwoOrLater

It is about the legal maintainability of a Software.
You are making the same mistake as the linux people:
They were not watching their legal status and got that SCO case
as one of the results of this.

If the next version will be compatible, people can
change the license in future - no need to give permission now.

Practical reason make that harder than necessary.

I don't think that we want to give now permission to relicense
GRASS, using a license non compatible with current license.

True, though this clause does not give that.

Thanks for added GPL.TXT, is was obviously missing.
  Bernhard

4

Bernhard Reiter wrote:

It is about the legal maintainability of a Software.
You are making the same mistake as the linux people:
They were not watching their legal status and got that SCO case
as one of the results of this.

I don't think that particular statement is true; I think that they
would have gotten the SCO case anyway. SCO's arguments seem to be:

1. That they own some of the stuff which IBM contributed.
2. They automatically own anything and everything related to Unix.

I don't think that anything which the Linux developers could have done
(e.g. FSF-style copyright assignments) would have changed anything in
that particular case.

--
Glynn Clements <glynn.clements@virgin.net>

5

On Wed, Apr 28, 2004 at 06:19:32PM +0100, Glynn Clements wrote:

Bernhard Reiter wrote:

> It is about the legal maintainability of a Software.
> You are making the same mistake as the linux people:
> They were not watching their legal status and got that SCO case
> as one of the results of this.

I don't think that particular statement is true; I think that they
would have gotten the SCO case anyway.

Well we cannot be sure, but chances are a lot less.

SCO's arguments seem to be:

1. That they own some of the stuff which IBM contributed.
2. They automatically own anything and everything related to Unix.

I don't think that anything which the Linux developers could have done
(e.g. FSF-style copyright assignments) would have changed anything in
that particular case.

Yes it would change something:
Towards 1: They would have required the written official
statement of IBM that they wrote the code and never looked at other stuff.
This would have triggered a more thorough legal examination within IBM.
Towards 2: They never claimed that in court and this is also the
reason why "GNU is Not Unix".

3. FSF could enter the stage as copyright holder suing SCO.
The linux developers have a harder time doing so, because they do
not own the copyright on all portions.

6

On Sunday 21 March 2004 09:29, Hamish wrote:

grass/src/general/init/license.txt
and
grass51/lib/init/license.txt

state:
"Geographic Resources Analysis Support System (GRASS) is Copyright,
1999-2004 by the GRASS Development Team, and licensed under terms of the
GNU General Public License (GPL)."

Should that specifically say GPL version 2?

The COPYING files do; they also say:

"You should have received a copy of the GNU General Public License
along with this program; if not, write to ..."

But it also says:

"Parts of GRASS are not copyright by the GRASS development team.
  The original authors hold the copyrights and you have to abide
  to their licensing terms where noted."

this is not quite clear to me.

Some parts have less restrictions, for example
grass51/lib/external/bwidget (LGPL)
grass/src/libes/bwidget (LGPL)
grass51/lib/external/shapelib (MIT/X, LGPL)
grass/src/libes/vect32_64/shapelib-1.2.8 (MIT/X, LGPL)

We redistribute those libs as GPL, is it true?
Is it OK if we don't change their license statement?
Or should we add a note about other licenses to COPYING etc.?

Are there pieces of the code with more restrictions than GPL?

In 5.7 I used also grass51/lib/form/html_library.tcl
( grass51/lib/form/license.terms ), the code may be redistributed
for free, but I am not sure if it may be redistributed as GPL.
Can you look at it Bernhard?

Radim

7

On Thu, Apr 29, 2004 at 12:17:08PM +0200, Radim Blazek wrote:

On Sunday 21 March 2004 09:29, Hamish wrote:
> grass/src/general/init/license.txt
> and
> grass51/lib/init/license.txt
>
> state:
> "Geographic Resources Analysis Support System (GRASS) is Copyright,
> 1999-2004 by the GRASS Development Team, and licensed under terms of the
> GNU General Public License (GPL)."
>
> Should that specifically say GPL version 2?
>
> The COPYING files do; they also say:
>
> "You should have received a copy of the GNU General Public License
> along with this program; if not, write to ..."

But it also says:

"Parts of GRASS are not copyright by the GRASS development team.
  The original authors hold the copyrights and you have to abide
  to their licensing terms where noted."

this is not quite clear to me.

Some parts have less restrictions, for example
grass51/lib/external/bwidget (LGPL)
grass/src/libes/bwidget (LGPL)
grass51/lib/external/shapelib (MIT/X, LGPL)
grass/src/libes/vect32_64/shapelib-1.2.8 (MIT/X, LGPL)

We redistribute those libs as GPL, is it true?

When we redistribute them as part a full system of binaries, we do.
That is a bit tricky, we do not do it to the sources,
as everybody can also get them under GNU LGPL.
Also see http://www.fsf.org/licenses/gpl-faq.html#GPLModuleLicense

So infact the sources can be used as GNU LGPL and GNU GPL.

Is it OK if we don't change their license statement?

Yes.
The user with the GRASS binaries gets both.

Or should we add a note about other licenses to COPYING etc.?

It would be nice to have an overview somewhere.
All software is redistributable under the GNU GPL to our best
knowledge (see below) so we are probably not required
to keep such an overview.

Are there pieces of the code with more restrictions than GPL?

We did a best effort approach to ensure that this is not the case.
I searched for the string "copyright" and some others
to identify all sections that could be problematic a few years back.
Because GRASS is a huge haystack,
the best effort approach was pragmatic to not get
completely stuck in licensing issues, but it is not a proof.
It is impossible to prove this anyway.

In 5.7 I used also grass51/lib/form/html_library.tcl
( grass51/lib/form/license.terms ), the code may be redistributed
for free, but I am not sure if it may be redistributed as GPL.
Can you look at it Bernhard?

The main part reads like:

  The authors hereby grant permission to use, copy, modify,
  distribute, and license this software and its documentation for any
  purpose, provided that existing copyright notices are retained in
  all copies and that this notice is included verbatim in any
  distributions. No written agreement, license, or royalty fee is
  required for any of the authorized uses. Modifications to this
  software may be copyrighted by their authors and need not follow the
  licensing terms described here, provided that the new terms are
  clearly indicated on the first page of each file where they apply.

This seems to be a non-protecting Free Software license
compatible with the GNU GPL.
Note it is very close in wording to the
  Standard ML of New Jersey Copyright License.
listed at http://www.fsf.org/licenses/license-list.html

Best,
  Bernhard