Ada in Debian: most libraries will switch to the pure GPL in Etch

Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

I received detailed answers from AdaCore's Robert Dewar and Arnaud
Charlet regarding the licenses of software downloaded from their
servers.  In summary:

- All software downloaded from AdaCore is pure GPL, no matter what the
  headers say.

- This also applies to software downloaded from the CVS server in
  source-only form.

- They refuse to give any assurances regarding copyright ownership, so
  I feel that I now need to go ask the authors.

- They will not sign a license document, even if a lawyer asked them
  (they are not required to sign anything, of course).

As you all know, Gentoo has switched to the pure GPL for all Ada
libraries, in accordance with these statements.  IANAL but this seems
to be the only legal route, so Debian will follow suit and switch to
the pure GPL for future versions of the libraries.

ASIS-for-GNAT is copyright (c) Free Software Foundation.
Theoretically, I could ask the FSF for a GMGPL license, and maybe get
the sources from gnat-asis.sourceforge.net, but that's unlikely to
work with GCC 4.1 without major work.  So, switch to pure GPL.

AUnit was written by Ed Falis, an AdaCore employee.  Only source is
AdaCore, so switch to pure GPL.

AWS was written by Dmitryi Anisimkov and Pascal Obry.  Neither of them
being an AdaCore employee (at least AFAIK), there may be a way to
acquire a GMGPL license from them.  They may also decide to fork the
project on a new repository.  Pascal's site [1] has a download page
that points to AdaCore's web site.  In the mean time, AdaCore is the
only source, so switch to pure GPL.

Florist was written by Florida State University; it is possible to get
a version ported to GCC 4.1 from http://gnat-florist.sourceforge.net,
so I'll do that.  GMGPL.

GLADE was written jointly by AdaCore, the ENST, A. Strohmeier, T. Wolf
and J. Kienzle.  Only source is AdaCore, so switch to pure GPL.

GtkAda was written by Emmanuel Briot, Joel Brobecker, Arnaud Charlet,
and Nicolas Setton, who are all AdaCore employees.  Only source is
AdaCore, so switch to pure GPL.

libgnat is GMGPL, since we obtained it from the FSF's repository as
part of GCC 4.1.

PolyORB is copyright (c) Free Software Foundation.  The situation is
exactly the same as with ASIS: the FSF doesn't seem to be aware of
PolyORB, the only place to get it from is AdaCore, and that's pure
GPL.  PolyORB has a home page [2], but the download link points to
AdaCore.  (that doesn't matter for now, since Debian does not provide
PolyORB).

Templates_Parser, a library that is incorporated into both AWS and
GPS, is copyright (c) AdaCore.  It seems that the author, Pascal Obry,
has assigned copyright to them.  If that's untrue (and the headers are
in error), then my remarks on AWS apply to Templates_Parser, too.
(Debian includes Templates_Parser as part of AWS and GPS, but not as a
separate package).

XML/Ada was written by Emmanuel Briot (an AdaCore employee),
Christophe Baillon and Martin Krischik.  Only source is AdaCore, so
switch to pure GPL.

AdaCore, or AdaCore employees, are authors of AUnit, GLADE, GtkAda,
and XML/Ada.  They might volunteer a GMGPL version of their work, but
I'm not going to ask them for one.

Since GtkAda has been mentioned more than any other library, I'd like
to point out that Debian already includes two alternatives:

- AdaBindX [3], a binding to X11 and LessTif by Hans-Frieder Vogt
  which is under GMGPL.  X11 is under X11 (MIT) license, and
  LessTif[4] is under LGPL.  But AdaBindX has not been updated since
  2000, and is not portable.  And not as good-looking as GTK+.
  According to Debian's Popularity Contest [5], this package has zero
  users, so I am tempted to drop it from Etch.  Speak up if you want
  me to keep it.

- TASH, the Tcl Ada SHell [6], includes a binding to Tk; it is
  available as a Debian package, under GMGPL, from Ada-France [7] and
  AdaWorld [8].  TASH is portable, but has not been updated since
  2003.

[1] http://www.obry.net
[2] http://polyorb.objectweb.org/
[3] http://home.arcor.de/hfvogt/programming.html
[4] http://www.lesstif.org/
[5] http://qa.debian.org/developer.php?popcon=libadabindx
[6] http://www.adatcl.com/
[7] http://www.ada-france.org/debian/
[8] http://www.adaworld.com/debian/

Thoughts, comments, offers to help?

-- 
Ludovic Brenta.

AdaCore ... the Next SCO?

[Jeffrey Creem]

Ludovic Brenta wrote:
> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:
> 
> - All software downloaded from AdaCore is pure GPL, no matter what the
>   headers say.
> 
> - This also applies to software downloaded from the CVS server in
>   source-only form.
> 

Hopefully someone at Greenhills is paying attention to this discussion. 
This confusing license perhaps "exposing" a company to GPL terms when 
the headers clearly are not GPL will make a great writeup that will 
pretty much make it impossible to even use GNATPro within my company.

At this point, it would seem the only purpose of the public CVS archives 
is to entrap people to allow AdaCore to sue at will. Comments and 
threads like this are of course likely to cause AdaCore to pull the CVS 
archives and that will be somewhat of a shame for free software 
developers but in its current state, the public CVS archives are doing 
more harm than good.


> - They refuse to give any assurances regarding copyright ownership, so
>   I feel that I now need to go ask the authors.
> 
> - They will not sign a license document, even if a lawyer asked them
>   (they are not required to sign anything, of course).

These last two statements are of course less problematic. AdaCore really 
  has no reason to sign things for people with which they have no 
business relationship.

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

Jeffrey Creem wrote :
> Hopefully someone at Greenhills is paying attention to this discussion.
> This confusing license perhaps "exposing" a company to GPL terms when
> the headers clearly are not GPL will make a great writeup that will
> pretty much make it impossible to even use GNATPro within my company.

I don't understand. Did your company not receive a license statement
from AdaCore along with GNAT Pro? Which company do you think is being
"exposed to the GPL terms"? And what does GreenHills have to do with
it? I wouldn't expect any problems, since AdaCore and GreeHills have a
formal agreement.

> At this point, it would seem the only purpose of the public CVS archives
> is to entrap people to allow AdaCore to sue at will. Comments and
> threads like this are of course likely to cause AdaCore to pull the CVS
> archives and that will be somewhat of a shame for free software
> developers but in its current state, the public CVS archives are doing
> more harm than good.

No, I don't think AdaCore want to sue anyone. I would rather think they
are die-hard, purist Free Software believers, like RMS (i.e. "free up
your software or pay"). Furthermore, they do not seem very eager to
serve the market of SMEs and one-man shops, who have been asking for
affordable GMGPL licenses for a long time. Well, that's their business
decision. As Marc A. Criley said so rightly, this decision just leaves
more room to Aonix and RR Software in that market.

I do agree that AdaCore's decision does more harm than good to Ada's
attractivity to SMEs. They are the only ones being hurt. The change of
license has no effect on corporations large enough to use GNAT Pro, or
on students, hobbyists, or free software developers.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Alex R. Mosteo]

Ludovic Brenta wrote:

> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:
> 
(snip)

> [1] http://www.obry.net
> [2] http://polyorb.objectweb.org/
> [3] http://home.arcor.de/hfvogt/programming.html
> [4] http://www.lesstif.org/
> [5] http://qa.debian.org/developer.php?popcon=libadabindx
> [6] http://www.adatcl.com/
> [7] http://www.ada-france.org/debian/
> [8] http://www.adaworld.com/debian/
> 
> Thoughts, comments, offers to help?

Thanks for the valuable summary of the situation. I don't have anything
particular to offer, except that I'm now using a debian derivative for my
office and home computers, so I closely follow these matters. I am also
interested in the Ada GPL community, so I'd like to volunteer in the
future, time permitting. 

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:

Ah, we'll see. Obviously the letter to me is still in transit.

> - All software downloaded from AdaCore is pure GPL, no matter what the
>   headers say.

> - This also applies to software downloaded from the CVS server in
>   source-only form.

So, if I get the source from ACT CVS it's GPL and if I get it from one
of the Archives somehwere out there, it has GMGPL. I'm talking GtkAda
2.2.1 and GtkAda 1.x specifically where there is strong indication
that it has been distributed as GMGPL (by whomever ...).

> - They refuse to give any assurances regarding copyright ownership, so
>   I feel that I now need to go ask the authors.

Please. Also It would be useful to ask AdaCore to document the
provenience of the sources they distribute: At which distribution
points the got the source, what are their assurances that they are
allowed to distribute the sources under GPL and so on :-). What their
baseline version is (that is from which version on they started to
"contribute" to the code base).

> - They will not sign a license document, even if a lawyer asked them
>   (they are not required to sign anything, of course).

I think they are required to say very clearly on which basis they
think they have the right to distribute the software they haven't
completely written themselves (which applies to most from libre2
AFAIS).

> As you all know, Gentoo has switched to the pure GPL for all Ada
> libraries, in accordance with these statements.  

> IANAL but this seems to be the only legal route, so Debian will
> follow suit and switch to the pure GPL for future versions of the
> libraries.


> ASIS-for-GNAT is copyright (c) Free Software Foundation.
> Theoretically, I could ask the FSF for a GMGPL license, and maybe get
> the sources from gnat-asis.sourceforge.net, but that's unlikely to
> work with GCC 4.1 without major work.  So, switch to pure GPL.

As far ASIS is concerned: I consider that to belong to the tool
sector, so GPL is no problem. ASIS is, in a sense an extension of
GNAT. I did not grok all of you statement here, though: What exactly
is the situation wrt to ASIS: 

> AUnit was written by Ed Falis, an AdaCore employee.  Only source is
> AdaCore, so switch to pure GPL.

Also a tool, wouldn't get linked into an executable for the customer:
In my eyes GPL is OK here.

> AWS was written by Dmitryi Anisimkov and Pascal Obry.  Neither of them
> being an AdaCore employee (at least AFAIK), there may be a way to
> acquire a GMGPL license from them.  They may also decide to fork the
> project on a new repository.  Pascal's site [1] has a download page
> that points to AdaCore's web site.  In the mean time, AdaCore is the
> only source, so switch to pure GPL.

Older Versions where GMGPL AFAIK. We should ask the authors for either
a GMGPL version (from their archives) or at least for the last GMGPL
version.

> Florist was written by Florida State University; it is possible to get
> a version ported to GCC 4.1 from http://gnat-florist.sourceforge.net,
> so I'll do that.  GMGPL.

This is, I think the same version as 3.15p as also available from
Baker's site. Since ACT did hardly anything with their GPLed florist
apart from removing the linking exception and adding a copyright
statement of their own, I'd judge that version still good (after all:
POSIC hasn't changed, it's mostly a question how it will build on
various platforms).

> GLADE was written jointly by AdaCore, the ENST, A. Strohmeier, T. Wolf
> and J. Kienzle.  Only source is AdaCore, so switch to pure GPL.

I agree as far as the tools themselves are concerned. But I can get a
version of glade-3.15p from non-ACT servers and (if we really want to
start splitting hair then: here) I can see linking exceptions in a
number of files. Perhaps these are exactly the runtime that gets
linked to client and server. This aspect is exactly why I'd really
want to know more about the copyright of single files in that whole
mess.


> GtkAda was written by Emmanuel Briot, Joel Brobecker, Arnaud Charlet,
> and Nicolas Setton, who are all AdaCore employees.  Only source is
> AdaCore, so switch to pure GPL.

So there has never been a GMGPL version? Never ever? I find that
odd. I suppose you can now never get Emmanuel Briot, Joel
Brobecker, Arnaud Charlet to admit that there was ever a GMGPL
version?

As far as past versions are concerned, I pull the following Texts from
my archive, which have accomanied GtkAda on libre (the old one) for
some time:

  Version 1.3: This package is distributed under the GPL license,
  slightly modified so that you can create proprietary software with
  this toolkit. The license is actually the same as the GNAT library
  itself. You should also read the Gtk license itself if you intend to
  do proprietary software based on gtk and GtkAda.

  Version 2.0.0: This package is distributed under the GPL license,
  slightly modified so that you can create proprietary software with
  this toolkit. The license is actually the same as the GNAT library
  itself. You should also read the Gtk license itself if you intend to
  do proprietary software based on gtk and GtkAda.

  Version 2.2.1: This package is distributed under the GPL license,
  slightly modified so that you can create proprietary software with
  this toolkit. The license is actually the same as the GNAT library
  itself. You should also read the Gtk license itself if you intend to
  do proprietary software based on gtk and GtkAda.

  Version 2.4.0: This package is distributed under the GPL license,
  slightly modified so that you can create proprietary software with
  this toolkit. The license is actually the same as the GNAT library
  itself. You should also read the Gtk license itself if you intend to
  do proprietary software based on gtk and GtkAda.


I suppose if I just get the version of GtkAda from my archive, I'll be
covered by GMGPL, or what? Or if I get this one:
http://www.adapower.net/libre/gtkada/GtkAda-2.0.0.tgz

ACT didn't even bother to change the source of 2.4.0 distributed from
libre2. The just somehow lost (not even substituted) the license
notice when they started to mirror libre.act-europe.fr.

All this has a positively Orwellian touch (especially if I factor in
the now well known Robert Dewar quote): Changing the past
retroactively.



> libgnat is GMGPL, since we obtained it from the FSF's repository as
> part of GCC 4.1.

Yep.

> PolyORB is copyright (c) Free Software Foundation.  The situation is
> exactly the same as with ASIS: the FSF doesn't seem to be aware of
> PolyORB, the only place to get it from is AdaCore, and that's pure
> GPL.  PolyORB has a home page [2], but the download link points to
> AdaCore.  (that doesn't matter for now, since Debian does not provide
> PolyORB).

polyorb/README

   -- As a special exception,  if other files  instantiate  generics from this --
   -- unit, or you link  this unit with other files  to produce an executable, --
   -- this  unit  does not  by itself cause  the resulting  executable  to  b  --

This even in the version obtained from ACT.

Other version can be gotten from FTP-Servers /= ACT's. 

If I go to http://polyorb.objectweb.org I find "As a special
exception, if other files instantiate generics from this ..." and 

"PolyORB 1.3r (2005-06-15) (...) This is the latest release of
PolyORB.  Release is a snapshot of current research work on middleware
architecture carried out by the ENST and the LIP6-SRC. (...) Note:
this snapshot is distributed on Libre's page."

It seems the copyright somehow gets lost in between objectweb.org and
ACT. I find it hardly credible that the agreement (of whatever kind)
between objectweb and ACT is, that ACT distributes under "pure
GPL". I'd suggest that the community ask objectweb for either
subversion access or the last GMGPL snapshot.


> Templates_Parser, a library that is incorporated into both AWS and
> GPS, is copyright (c) AdaCore.  It seems that the author, Pascal Obry,
> has assigned copyright to them.  If that's untrue (and the headers are
> in error), 

ACT is in the habit of rewriting headers.

> then my remarks on AWS apply to Templates_Parser, too.  (Debian
> includes Templates_Parser as part of AWS and GPS, but not as a
> separate package).


> XML/Ada was written by Emmanuel Briot (an AdaCore employee), ...
> Christophe Baillon and Martin Krischik.  Only source is AdaCore, so
> switch to pure GPL.

From the xmlada README: 

  "As a special exception, if other files instantiate generics from ..."

ACT might be in their rights to strip the linking exception from all
source they distribute from their site. But since at least a part of
that software got distributed under different licenses earlier, I
doubt they can take back the linking eception on all that copies
floating around on various backup media and ftp servers. Not
retroactively.

That also applies to all the other packages where the copyright is now
being stripped retroactivly.

> AdaCore, or AdaCore employees, are authors of AUnit, GLADE, GtkAda,
> and XML/Ada.  They might volunteer a GMGPL version of their work, but
> I'm not going to ask them for one.

I understand that. Is BTW, anybody here who's intrested in
constituting a GMGPL baseline of some core libraries?


> Since GtkAda has been mentioned more than any other library, I'd like
> to point out that Debian already includes two alternatives:

Ludovico. I understand that. But: Gtk+ has been LGPL for very good
reasons and part of the success of Gnome is based on that. In my
opinion the future on Unix/Linux is Gtk+/Pango/Cairo. And it comes as
a special bonus that these libs are portable. Everything else is
probably good to hack some tool now, but not to build something that
can survive in for the next 10 years. 

I'm not talking about a single program but about a code base. As I
very small example, I've developed (among other things) a flowing
label widget (Text in label flows on resizing) for GtkAda. I wouldn't
see any sense in developing (and releasing to the community)
infrastructure of that kind for what I consider a niche solution.

> - AdaBindX [3], a binding to X11 and LessTif by Hans-Frieder Vogt
>   which is under GMGPL.  X11 is under X11 (MIT) license, and
>   LessTif[4] is under LGPL.  But AdaBindX has not been updated since
>   2000, and is not portable.  


> And not as good-looking as GTK+.

>   According to Debian's Popularity Contest [5], this package has zero
>   users, so I am tempted to drop it from Etch.  Speak up if you want
>   me to keep it.

Just announce it again when you drop it. A X11-Binding would have been
nice for experiments, especially for trying to write a Ada native
widget set, but that not here, rather science fiction (or more likely
alternate history ...), so ...

> 
> - TASH, the Tcl Ada SHell [6], includes a binding to Tk; it is
>   available as a Debian package, under GMGPL, from Ada-France [7] and
>   AdaWorld [8].  TASH is portable, but has not been updated since
>   2003.

> Thoughts, comments, offers to help?

I'd like to help establishing GMGPL libraries (for libraries that get
linked into distributed executables, GPL is OK with me for the tool
chain). 

My thoughts, rather jumbled, I fear I've offered above.

Sorry for all this. I feel like single handedly killing Ada on Debian
for asking that stupid licensing question at the beginning.

Regards -- Markus

What luck we had ...

[M E Leypold]

There was a project we bid on in the beginning of 2005, when
libre.act-europe still carried GtkAda 2.4.0 with the GMGPL license
notice.

We didn't get the project (for various reason that had nothing to do
with Ada in this case :-)), but I suppose we have to count ourselves
really lucky that we didn't gat it: This customer would have come in
for multiple damages if he had found that his codebase got suddenly
contaminated. Of course we could have done it again in another
language, but then he'd have come in for contractual damages because
of the project being behind schedule.

Not that I think the ACT letter as quoted here would hold up in court,
not with all that contradicting evidence. But nonetheless :-). Avoid
lawyers, where possible. :-).

Regards -- Markus.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Dr. Adrian Wrigley]

On Tue, 27 Jun 2006 03:58:40 -0700, Ludovic Brenta wrote:

> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:
> 
> - All software downloaded from AdaCore is pure GPL, no matter what the
>   headers say.
> 
> - This also applies to software downloaded from the CVS server in
>   source-only form.
> 
> - They refuse to give any assurances regarding copyright ownership, so
>   I feel that I now need to go ask the authors.
> 
> - They will not sign a license document, even if a lawyer asked them
>   (they are not required to sign anything, of course).

I'm a little confused by all this...

Can you tell us which combination of the following is true, from
what you understand:

1) the GMGPL licences issued by AdaCode and others are being revoked?
2) AdaCore and others say they never granted licences under GMGPL?
3) the licences were granted and are still in force?
4) all licencing terms embedded in the distributions are repudiated?
5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?

(does this cover the basic possibilities?)

the code in question being GtkAda, libgnat, GLADE, etc.

In some cases (GtkAda?), the original authors transferred copyright to
AdaCore(?) - did the original authors revoke or repudiate the licences
in effecting this transfer?

Thanks.
--
Dr. Adrian Wrigley,
Cambridge.

Re: AdaCore ... the Next SCO?

[M E Leypold]

Jeffrey Creem <jeff@thecreems.com> writes:

> Ludovic Brenta wrote:
> > I received detailed answers from AdaCore's Robert Dewar and Arnaud
> > Charlet regarding the licenses of software downloaded from their
> > servers.  In summary:
> > - All software downloaded from AdaCore is pure GPL, no matter what
> > the
> >   headers say.
> > - This also applies to software downloaded from the CVS server in
> >   source-only form.
> >
> 
> Hopefully someone at Greenhills is paying attention to this
> discussion. This confusing license perhaps "exposing" a company to GPL
> terms when the headers clearly are not GPL will make a great writeup
> that will pretty much make it impossible to even use GNATPro within my
> company.

What AdaCore does at the moment is (IMHO, IANAL) to spread pure FUD
(if in doubt, license from ACT) -- ironically using GPL as a
weapon. Of course if any of the "proprietary software is the only
good" shills will get hold of the subject, FUD will go well with
further FUD.

> At this point, it would seem the only purpose of the public CVS
> archives is to entrap people to allow AdaCore to sue at will. Comments

Not only. Of course they want patches flowing back to them and beta
testers are also welcome. The contribution list on the old site for
2.4.0 read like this:


  # Alex Bykat who helped us compile GtkAda 0.2 on Solaris 2.5.
  # Jeff Creem who did the first port of GtkAda to Win32. Note that the Win32 port of GtkAda is now handled by the GtkAda team.
  # Philippe Durif who wrote the first version of the GtkAda documentation.
  # Samuel Tardieu who wrote the man pages of GtkAda for the debian project.
  # Francisco Javier Loma Daza who sent us some useful patches for gtk-style and gtk-table.
  # Paul Pukite who sent us the Gdk.RGB package.
  # Thomas Brupbacher who implemented Glade's support for Clist, List, Paned, Vscrollbar and Viewport, a basic support for XML attributes and various other bug fixes.
  # David Botton who maintains a GtkAda mirror.
  # Bobby D. Bryant who fixed the handling of comments in Glib.XML.
  # Manuel Op de Coul who has contributed some documentation.
  # Aidan Skinner who has contributed some patches to the Gnome binding.
  # Preben Randhol who has contributed several patches, in particular for porting gate to glade-2.

Useful, this, isn't it?

> and threads like this are of course likely to cause AdaCore to pull
> the CVS archives and that will be somewhat of a shame for free
> software developers but in its current state, the public CVS archives
> are doing more harm than good.

ACK.

> 
> 
> > - They refuse to give any assurances regarding copyright ownership, so
> >   I feel that I now need to go ask the authors.
> > - They will not sign a license document, even if a lawyer asked them
> >   (they are not required to sign anything, of course).
> 
> These last two statements are of course less problematic. AdaCore
> really has no reason to sign things for people with which they have no
> business relationship.

I think, if they want to keep their rights to publicly available
source they can be expected to send a binding letter stating their
position on their ownership.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> Jeffrey Creem wrote :

> "exposed to the GPL terms"? And what does GreenHills have to do with
> it? 

Greenhills is in the habit of publishing fuddy little articles on
their site how risky (securitywise and regarding exposition to
licensing issues / suites) it is supposed to be using open source as
part of your infa structure or product.

> 
> > At this point, it would seem the only purpose of the public CVS archives
> > is to entrap people to allow AdaCore to sue at will. Comments and
> > threads like this are of course likely to cause AdaCore to pull the CVS
> > archives and that will be somewhat of a shame for free software
> > developers but in its current state, the public CVS archives are doing
> > more harm than good.
> 
> No, I don't think AdaCore want to sue anyone. I would rather think they
> are die-hard, purist Free Software believers, like RMS (i.e. "free up

I agree, it's much more complicated, but neither do I believe in the
"die-hard, purist Free Software believers". This is a business. By
trying retroactively to change licenses from GMGPL to GPL they are
sending a message. They could as well have left the GMGPL. So, if they
took it upon them to change the license (it doesn't change of its own)
there is a purpose behind that.  Add to that, that their Pro-Version
probably still has the linking exception.

> your software or pay"). 

Which never has been RMS's stance, AFAIS.

> Furthermore, they do not seem very eager to
> serve the market of SMEs and one-man shops, who have been asking for
> affordable GMGPL licenses for a long time. Well, that's their business
> decision. 

It is. Trying to change the past, though, seems a bit fishy to me.

> As Marc A. Criley said so rightly, this decision just leaves
> more room to Aonix and RR Software in that market.


> I do agree that AdaCore's decision does more harm than good to Ada's
> attractivity to SMEs. They are the only ones being hurt. The change of
> license has no effect on corporations large enough to use GNAT Pro, or
> on students, hobbyists, or free software developers.

As I mentioned before: Students and hobbyists, even free software
developers are not seldom looking forward to becoming SMEs (as I did
:-). A skill set which places a high entry barrier on that transition
(wether it bee restriction to GPL software only, huge licensing fees
or restriction to a non free platform/environment, i.e. forgetting the
other half of the respective skill set), such a skill set is probably
not very attractive anymore.

So it has an effect on "students, hobbyists, or free software
developers" (And more: I am one, but I cannot develeop only free
software. What I free are spin offs and reusable components from
projects).

Or to put it differently: If I start to using a language as a main
tool as a hobbyits, some of the questions I ask, are like this (I
stylize a bit :-)):

  - Will anybody love/adore/employ me for using that language?

Whereas Ada is great in the coolness department (that's the "adore"
aspect of the thing), it's already rather weak on the love/employ
department. "Ada, what is that?".  This will not become better by SMEs
not using Ada anymore or not even thinking about it. And this affects
the hobbyist. There's hardly ever such a thing as a pure hobbyist.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Dr. Adrian Wrigley wrote :
> On Tue, 27 Jun 2006 03:58:40 -0700, Ludovic Brenta wrote:
>
> > I received detailed answers from AdaCore's Robert Dewar and Arnaud
> > Charlet regarding the licenses of software downloaded from their
> > servers.  In summary:
> >
> > - All software downloaded from AdaCore is pure GPL, no matter what the
> >   headers say.
> >
> > - This also applies to software downloaded from the CVS server in
> >   source-only form.
> >
> > - They refuse to give any assurances regarding copyright ownership, so
> >   I feel that I now need to go ask the authors.
> >
> > - They will not sign a license document, even if a lawyer asked them
> >   (they are not required to sign anything, of course).
>
> I'm a little confused by all this...
>
> Can you tell us which combination of the following is true, from
> what you understand:
>
> 1) the GMGPL licences issued by AdaCode and others are being revoked?

Switched to pure GPL, not revoked. The "linking and generic
instantiation" exception is revoked.

> 2) AdaCore and others say they never granted licences under GMGPL?

They don't say that, but they refuse to give details on when the switch
took place. I don't know. Ask them. From a theoretical standpoint, I am
indeed quite worried that I have downloaded and redistributed AdaCore's
software, thinking in good faith I had the right to do so when in fact
I didn't, since I didn't have written permission from any of the
copyright holders. Remember, the headers amount to naught from a legal
perspective. Practically speaking, I don't think AdaCore will sue me,
or anyone else, for that. IANAL.

> 3) the licences were granted and are still in force?

I don't know. I asked but they wouldn't go into specifics. Ask them for
yourself, if you're concerned.

> 4) all licencing terms embedded in the distributions are repudiated?

They never had any legal force; only a signed statement from the
copyright holder has legal force.

> 5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?

Yes, since it is under GPL. That much they are willing to say and
certify, but not in writing :(

> (does this cover the basic possibilities?)
>
> the code in question being GtkAda, libgnat, GLADE, etc.
>
> In some cases (GtkAda?), the original authors transferred copyright to
> AdaCore(?) - did the original authors revoke or repudiate the licences
> in effecting this transfer?

As I explained above, I don't know, and AdaCore refused to tell me. Ask
each author individually.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> Dr. Adrian Wrigley wrote :
> > On Tue, 27 Jun 2006 03:58:40 -0700, Ludovic Brenta wrote:
> >
> > > I received detailed answers from AdaCore's Robert Dewar and Arnaud
> > > Charlet regarding the licenses of software downloaded from their
> > > servers.  In summary:
> > >
> > > - All software downloaded from AdaCore is pure GPL, no matter what the
> > >   headers say.
> > >
> > > - This also applies to software downloaded from the CVS server in
> > >   source-only form.
> > >
> > > - They refuse to give any assurances regarding copyright ownership, so
> > >   I feel that I now need to go ask the authors.
> > >
> > > - They will not sign a license document, even if a lawyer asked them
> > >   (they are not required to sign anything, of course).
> >
> > I'm a little confused by all this...
> >
> > Can you tell us which combination of the following is true, from
> > what you understand:
> >
> > 1) the GMGPL licences issued by AdaCode and others are being revoked?
> 
> Switched to pure GPL, not revoked. The "linking and generic
> instantiation" exception is revoked.
> 
> > 2) AdaCore and others say they never granted licences under GMGPL?
> 
> They don't say that, but they refuse to give details on when the switch
> took place. I don't know. Ask them. From a theoretical standpoint, I am
> indeed quite worried that I have downloaded and redistributed AdaCore's

Considering that act-europe.fr carried the GMGPL license notice for
GtkAda et al until 2005/2 at least, there is nothing to worry about
(except the attitude of ACT).

> software, thinking in good faith I had the right to do so when in
> fact I didn't, since I didn't have written permission from any of
> the copyright holders. Remember, the headers amount to naught from a
> legal perspective.

They say.

> Practically speaking, I don't think AdaCore will sue me,
> or anyone else, for that. IANAL.

IANAL2.

> 
> > 3) the licences were granted and are still in force?
> 
> I don't know. I asked but they wouldn't go into specifics. Ask them for
> yourself, if you're concerned.
> 
> > 4) all licencing terms embedded in the distributions are repudiated?
> 
> They never had any legal force; only a signed statement from the
> copyright holder has legal force.

They say. 

> 
> > 5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?
> 
> Yes, since it is under GPL. That much they are willing to say and
> certify, but not in writing :(

So they will take even that back in 2 years from now? ...

Regards -- Markus

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Ludovic Brenta wrote:

>  The change of
> license has no effect on corporations large enough to use GNAT Pro, or
> on students, hobbyists, or free software developers.

Does it? The GPL is viral and booch, charles, AdaCL - all currently GMGPL
would need to to relicensed to GPL now to be used with with GNAT/GPL.

AdaCL at least produced licence-warnings when compiled with GNAT/GPL. While
licence warning are not legal the at least shows how AdaCore sees things.
And this would mean that those libs become unavailable to closed source
users.

Martin

PS: Just in case you have not seen any warnings so far: all package
specifications copied from the RM come with licence "unrestricted" and they
are free in all respects. Only when you use a packaged not mentioned inside
the RM the warning pops up.
-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Dr. Adrian Wrigley]

On Tue, 27 Jun 2006 08:28:38 -0700, Ludovic Brenta wrote:

> Dr. Adrian Wrigley wrote :
>> On Tue, 27 Jun 2006 03:58:40 -0700, Ludovic Brenta wrote:
>>
>> > I received detailed answers from AdaCore's Robert Dewar and Arnaud
>> > Charlet regarding the licenses of software downloaded from their
>> > servers.  In summary:
>> >
>> > - All software downloaded from AdaCore is pure GPL, no matter what the
>> >   headers say.
>> >
>> > - This also applies to software downloaded from the CVS server in
>> >   source-only form.
>> >
>> > - They refuse to give any assurances regarding copyright ownership, so
>> >   I feel that I now need to go ask the authors.
>> >
>> > - They will not sign a license document, even if a lawyer asked them
>> >   (they are not required to sign anything, of course).
>>
>> I'm a little confused by all this...
>>
>> Can you tell us which combination of the following is true, from
>> what you understand:
>>
>> 1) the GMGPL licences issued by AdaCode and others are being revoked?
> 
> Switched to pure GPL, not revoked. The "linking and generic
> instantiation" exception is revoked.
> 
>> 2) AdaCore and others say they never granted licences under GMGPL?
> 
> They don't say that, but they refuse to give details on when the switch
> took place. I don't know. Ask them. From a theoretical standpoint, I am
> indeed quite worried that I have downloaded and redistributed AdaCore's
> software, thinking in good faith I had the right to do so when in fact
> I didn't, since I didn't have written permission from any of the
> copyright holders. Remember, the headers amount to naught from a legal
> perspective. Practically speaking, I don't think AdaCore will sue me,
> or anyone else, for that. IANAL.
> 
>> 3) the licences were granted and are still in force?
> 
> I don't know. I asked but they wouldn't go into specifics. Ask them for
> yourself, if you're concerned.
> 
>> 4) all licencing terms embedded in the distributions are repudiated?
> 
> They never had any legal force; only a signed statement from the
> copyright holder has legal force.
> 
>> 5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?
> 
> Yes, since it is under GPL. That much they are willing to say and
> certify, but not in writing :(
> 
>> (does this cover the basic possibilities?)
>>
>> the code in question being GtkAda, libgnat, GLADE, etc.
>>
>> In some cases (GtkAda?), the original authors transferred copyright to
>> AdaCore(?) - did the original authors revoke or repudiate the licences
>> in effecting this transfer?
> 
> As I explained above, I don't know, and AdaCore refused to tell me. Ask
> each author individually.

I'm sorry you have been put in this position - relaying and interpreting
apparent changes in licensing conditions.  The issues raised here
are potentially of very wide interest to many millions of licencees
and licensors under the GPL and other Free software licences.

Under English law (and presumably most other places), signed statements
are not required to form a contract.  In particular, if the parties
behaved and believed that there is a contract then one exists.
Evidence such as files, ftp sites, emails etc. can help support
the claim that a contract existed.  Surely the behaviour of
the authors and users backs the claim that licences were granted?
Signatures on bits of paper might help, but still don't provide
a full guarantee.  IANAL.

As regards the GPL, it appears to be a perpetual, sub-licensable,
non-revocable (absense of breaches) licence.  The GMGPL terms
add to this, but don't change these basic features.

I am at a loss to understand what basis there is for revoking the
licences already issued.  As a party to the licence
contract between myself and the authors, I feel aggrieved.
I have kept to my side of the bargain.  I'm not convinced they
have kept to theirs.

Some people here will want a formal opinion.  I don't feel I have
need or resources for a professional view myself.  But I am being
made extremely wary of exposing myself to the possible legal
risks involved in *any* substantive business project involving
these software components.  The way these licences seem to be
being revoked, changed or withdrawn, without adequate explanation
is certainly a breach of the implied social "contract" created
when software is published.
--
Adrian

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Dmitry A. Kazakov]

On Tue, 27 Jun 2006 16:40:51 GMT, Dr. Adrian Wrigley wrote:

> Under English law (and presumably most other places), signed statements
> are not required to form a contract.  In particular, if the parties
> behaved and believed that there is a contract then one exists.
> Evidence such as files, ftp sites, emails etc. can help support
> the claim that a contract existed.  Surely the behaviour of
> the authors and users backs the claim that licences were granted?
> Signatures on bits of paper might help, but still don't provide
> a full guarantee.  IANAL.

Exactly. That was my question in another thread.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Tue, 2006-06-27 at 18:38 +0200, Martin Krischik wrote:
> Ludovic Brenta wrote:
> 
> >  The change of
> > license has no effect on corporations large enough to use GNAT Pro, or
> > on students, hobbyists, or free software developers.
> 
> Does it? The GPL is viral

A virus doesn't need to be accepted, in general.
Usually you catch a cold, your not establishing a
contract with influenza. A license is different.


>  and booch, charles, AdaCL - all currently GMGPL
> would need to to relicensed to GPL now to be used with with GNAT/GPL.

I don't think that any GMGPL sources need to be relicensed
to be used with GNAT GPL Edition for free software work.
The GMGPL is certainly GPL compatible.
Even if you run into a pragma License(some invalid identifier),
you just create a derivative work by placing '--' before the
pragma. Just make sure that the result, if distributed, is
distributed as required by the GPL.


> And this would mean that those libs become unavailable to closed source
> users.

But that's a consequence of using GNAT GPL edition, not of using
the libraries? These can keep their licenses. (The GPL doesn't say
it is time limited, and revocable; I'd be surprised to learn that
the exception is time limited.)

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:
>
> - All software downloaded from AdaCore is pure GPL, no matter what the
>   headers say.
>
> - This also applies to software downloaded from the CVS server in
>   source-only form.
>
> - They refuse to give any assurances regarding copyright ownership, so
>   I feel that I now need to go ask the authors.
>
> - They will not sign a license document, even if a lawyer asked them
>   (they are not required to sign anything, of course).

How can any AdaCore software then remain in Debian at all? How is any
Debian user supposed to know any AdaCore software is GPL if file
headers are invalid? How do *you* know? You have nothing written and
signed and the authors deny any validity of anything written in the
software itself.

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Tue, 2006-06-27 at 17:00 +0200, M E Leypold wrote:

>   # Alex Bykat who helped us compile GtkAda 0.2 on Solaris 2.5.
>   # ...
>   # Preben Randhol who has contributed several patches, in particular for porting gate to glade-2.

> Useful, this, isn't it?

Could make a change if members of the above list decided to
make a GMGPL Edition from their earlier work.
But just in case, this isn't useful as a licensing argument.
A quote from the GNU site illustrates:

"If you modify this code, you may extend this exception to your version
of the code, but you are not obligated to do so. If you do not wish to
do so, delete this exception statement from your version."

So the release of a modified GtkAda may drop the
exception without consulting the originators.
I can't say whether or not the members of the above list
like the new facts.

-- Georg 

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

>> 4) all licencing terms embedded in the distributions are repudiated?
>
> They never had any legal force; only a signed statement from the
> copyright holder has legal force.


From a well know file named COPYING (note paragraph 2):

	    How to Apply These Terms to Your New Programs

  If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.

  To do so, attach the following notices to the program.  It is safest
to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least
the "copyright" line and a pointer to where the full notice is found.

    <one line to give the program's name and a brief idea of what it does.>
    Copyright (C) 19yy  <name of author>

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA


-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ed Falis]

M E Leypold wrote:

>> AUnit was written by Ed Falis, an AdaCore employee.  Only source is
>> AdaCore, so switch to pure GPL.
>
> Also a tool, wouldn't get linked into an executable for the customer:
> In my eyes GPL is OK here.

Note that AUnit has been GPL from its first release in 2000.  Since its
intent is user "in-house" testing rather than any kind of
re-distribution of test cases, I considered it the appropriate license.

- Ed

Re: What luck we had ...

[Pascal Obry]

Or maybe your customer will have been happy to get the application under
GPL license and have the right to ask for the sources. Do you think that
giving away the sources to your customer will have changed something ? I
bet he would not even consider asking for the sources :)

Pascal.

-- 

--|------------------------------------------------------
--| Pascal Obry                           Team-Ada Member
--| 45, rue Gabriel Peri - 78114 Magny Les Hameaux FRANCE
--|------------------------------------------------------
--|              http://www.obry.net
--| "The best way to travel is by means of imagination"
--|
--| gpg --keyserver wwwkeys.pgp.net --recv-key C1082595

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

>> 4) all licencing terms embedded in the distributions are repudiated?
>
> They never had any legal force; only a signed statement from the
> copyright holder has legal force.

Oh, and BTW I suppose the following then also has no legal force:

-- This library is distributed in the hope that it will be useful,   --
-- but WITHOUT ANY WARRANTY; without even the implied warranty of    --
-- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU --
-- General Public License for more details.                          --

(from gdk.ads)

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Tue, 2006-06-27 at 20:39 +0200, Michael Bode wrote:
> "Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:
> 
> > I received detailed answers from AdaCore's Robert Dewar and Arnaud
> > Charlet regarding the licenses of software downloaded from their
> > servers.  In summary:

>  How is any
> Debian user supposed to know any AdaCore software is GPL if file
> headers are invalid?

The file headers aren't invalid. That sounds like nitpicking,
but "invalid" has different meaning. Anyway, all this is relevant
only when someone uses software in a way that by applicable
law is (a) illegal, and (b) this fact is made a legal issue.

As long as Debian doesn't have any legal obligations, they can
proactively try to reduce the risk of being associated with
software that incurs a surprising legal status. I think that
one way to do this is to collect all available evidence,
and take Dewar's and Charlet's word for it.
Chances are that they represent the legal party that could make
a claim.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> The file headers aren't invalid. That sounds like nitpicking,
> but "invalid" has different meaning.

Ok. ... if file headers are lying.

> Anyway, all this is relevant only when someone uses software in a
> way that by applicable law is (a) illegal, and (b) this fact is made
> a legal issue.

Debian is distributing the software. If at some later point in time it
is claimed that the software today was not under GPL but some other
license which forbids distribution they have a problem. Seems like
this has just happend with GtkAda and GMGPL. 

> As long as Debian doesn't have any legal obligations, they can
> proactively try to reduce the risk of being associated with
> software that incurs a surprising legal status. I think that
> one way to do this is to collect all available evidence,
> and take Dewar's and Charlet's word for it.

They only can collect bits which are not signed with strong crypto.

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Ludovic Brenta:

> They don't say that, but they refuse to give details on when the switch
> took place. I don't know. Ask them. From a theoretical standpoint, I am
> indeed quite worried that I have downloaded and redistributed AdaCore's
> software, thinking in good faith I had the right to do so when in fact
> I didn't, since I didn't have written permission from any of the
> copyright holders.

You didn't download from Adacore, I think, but from NYU.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Tue, 2006-06-27 at 21:39 +0200, Michael Bode wrote:

> Debian is distributing the software. If at some later point in time it
> is claimed that the software today was not under GPL but some other
> license which forbids distribution they have a problem.

Yes, they will have a problem if someone accuses them of distributing
software they weren't allowed to distribute. But...

>  Seems like
> this has just happend with GtkAda and GMGPL. 

I'm not so sure. First, I don't think that the maintainers of Ada
stuff in Debian will have a problem. By design, they never distribute
anything without providing access to the source code as well.
So GMGPL or GPL is not a legal issue at all from a Debian maintainer's
perspective, as long as the issue is whether it is GPL or GMGPL.

Now if you wanted to make a claim that in spite of messages from
AdaCore officials to the contrary, software downloaded from the
AdaCore site is neither GPLed nor GMGPLed, then I think everyone
will be eager to hear what else it is.

> > to collect all available evidence,
> > and take Dewar's and Charlet's word for it.
> 
> They only can collect bits which are not signed with strong crypto.

I see the irony, but this is legal stuff. Consider a few sheets of
paper landing on your desk saying, this is a part of the NT kernel.
This part of the NT kernel source, as the header clearly indicates,
is GPLed. Would you believe it?

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

Georg Bauhaus <bauhaus@futureapps.de> writes:

>>  Seems like
>> this has just happend with GtkAda and GMGPL. 
>
> I'm not so sure. First, I don't think that the maintainers of Ada
> stuff in Debian will have a problem. By design, they never distribute
> anything without providing access to the source code as well.
> So GMGPL or GPL is not a legal issue at all from a Debian maintainer's
> perspective, as long as the issue is whether it is GPL or GMGPL.
>
> Now if you wanted to make a claim that in spite of messages from
> AdaCore officials to the contrary, software downloaded from the
> AdaCore site is neither GPLed nor GMGPLed, then I think everyone
> will be eager to hear what else it is.

I did not claim that Debian right now has a problem with GtkAda. I
claim that a sudden, secret and maybe retroactive (or maybe it was
between 2.4.0 and 2.4.1?, AdaCore won't tell us) change of license
happend to GtkAda. And if that happens once ...

Regarding messages from AdaCore to the contrary of anything:

http://web.archive.org/web/20050224092046/http://libre.act-europe.fr/GtkAda/

-- 
Michael Bode

Re: What luck we had ...

[Jeffrey R. Carter]

M E Leypold wrote:
> 
> We didn't get the project (for various reason that had nothing to do
> with Ada in this case :-)), but I suppose we have to count ourselves
> really lucky that we didn't gat it: This customer would have come in
> for multiple damages if he had found that his codebase got suddenly
> contaminated. Of course we could have done it again in another
> language, but then he'd have come in for contractual damages because
> of the project being behind schedule.

A change to the license on future distributions does not change the 
license under which you obtained it. If you obtained a GMGPL version of 
GtkAda 2.4.0, then you had a GMGPL version, regardless of the license 
used later. If you kept that version, then you still have a GMGPL 
version. You also have the right to redistribute that GMGPL version 
granting the GMGPL to those you distribute it to.

-- 
Jeff Carter
"Many times we're given rhymes that are quite unsingable."
Monty Python and the Holy Grail
57

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Florian Weimer <fw@deneb.enyo.de> writes:

> * Ludovic Brenta:
>
>> They don't say that, but they refuse to give details on when the switch
>> took place. I don't know. Ask them. From a theoretical standpoint, I am
>> indeed quite worried that I have downloaded and redistributed AdaCore's
>> software, thinking in good faith I had the right to do so when in fact
>> I didn't, since I didn't have written permission from any of the
>> copyright holders.
>
> You didn't download from Adacore, I think, but from NYU.

I did download quite a lot from AdaCore: GtkAda, AWS, XML/Ada, and
also GPS but that's not a library.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Tue, 2006-06-27 at 22:23 +0200, Michael Bode wrote:

>  I
> claim that a sudden, secret and maybe retroactive (or maybe it was
> between 2.4.0 and 2.4.1?, AdaCore won't tell us) change of license
> happend to GtkAda. And if that happens once ...
> 
> Regarding messages from AdaCore to the contrary of anything:
> 
> http://web.archive.org/web/20050224092046/http://libre.act-europe.fr/GtkAda/
> 

Ludovic lets us know a response from AdaCore that uses the _present_
tense,

- All software downloaded from AdaCore is pure GPL, no matter what the
  headers say.

I don't think that _if_ there was a legal status at a time in the
recent past, this status can be declared something different.
These things should happen only after one legal system has been
dropped in favor of another, for example after a war.
I am aware that many of the software packages used to be advertised
as being GMGPL.

With Debian in particular,
I think there might be more issues, like you don't want to stress your
good relations with AdaCore, if any, if you are maintaining Debian Ada 
software that nowadays is mostly produced by AdaCore. This, I guess,
is just a business issue for some who are involved.

In fact, we might be able to produce more GPLed software for profit
in spite of all the rogues around. It's worth a try, anyway, and
some report success.

Yet, what about a collaborative effort to produce good general
purpose Ada libraries to be used in Ada programs,
irrespective of the compiler. Hm. A PAL with funding from everyone
involved, including compiler makers and their customers.
We've all got to make a living, haven't we?


Georg 

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Michael Bode <m.g.bode@web.de> writes:
> How can any AdaCore software then remain in Debian at all? How is any
> Debian user supposed to know any AdaCore software is GPL if file
> headers are invalid? How do *you* know? You have nothing written and
> signed and the authors deny any validity of anything written in the
> software itself.

This is becoming metaphysical.  Cogito, ergo sum, and all that :) Your
argument applies to all software that is not at least
cryptographically signed by their author, and accompanied by a signed
license statement :) (note: software distributed by Debian _is_
cryptographically signed, but that from upstream is not.

AdaCore make no claim as to the license terms for software I download
from other sites than AdaCore.  And I don't think they can
retroactively revoke the licenses that Debian received when I
initially downloaded the software from NYU, AdaCore, or other places.
But IANAL.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

M E Leypold writes:
> Ludovic Brenta writes:
>> - They refuse to give any assurances regarding copyright ownership, so
>>   I feel that I now need to go ask the authors.
>
> Please. Also It would be useful to ask AdaCore to document the
> provenience of the sources they distribute: At which distribution
> points the got the source, what are their assurances that they are
> allowed to distribute the sources under GPL and so on :-). What their
> baseline version is (that is from which version on they started to
> "contribute" to the code base).

I'd rather spend my time on the technical side of things.  I'm a
software engineer :-)

I'll see how much interest there is in GMGPL versions of libraries,
and decide then.  You can ask these authors directly if you want to.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Dr. Adrian Wrigley writes:
> I'm sorry you have been put in this position - relaying and interpreting
> apparent changes in licensing conditions.  The issues raised here
> are potentially of very wide interest to many millions of licencees
> and licensors under the GPL and other Free software licences.

Yes :( But it also applies to non-free EULAs, or any kind of license
whatsoever.  It even applies to the movie I purchased on DVD a couple
of days ago.  The lawyers are after us!   Run!  Run!

> Under English law (and presumably most other places), signed statements
> are not required to form a contract.  In particular, if the parties
> behaved and believed that there is a contract then one exists.
> Evidence such as files, ftp sites, emails etc. can help support
> the claim that a contract existed.  Surely the behaviour of
> the authors and users backs the claim that licences were granted?
> Signatures on bits of paper might help, but still don't provide
> a full guarantee.  IANAL.

A license is not the same as a contract.  A contract has to have
measurable obligations for both parties; a license is only a grant of
rights to the licensee, possibly without compensation.  No
compensation, no contract, either written or implied.

> As regards the GPL, it appears to be a perpetual, sub-licensable,
> non-revocable (absense of breaches) licence.  The GMGPL terms
> add to this, but don't change these basic features.

Right.  That's why I think Debian is in the clear as regards the older
versions of AdaCore's software that are presently in the distribution.

> I am at a loss to understand what basis there is for revoking the
> licences already issued.  As a party to the licence
> contract between myself and the authors, I feel aggrieved.
> I have kept to my side of the bargain.  I'm not convinced they
> have kept to theirs.

Since they received no compensation for the software, there is no
bargain, no contract, ergo they have no obligations.

> Some people here will want a formal opinion.  I don't feel I have
> need or resources for a professional view myself.  But I am being
> made extremely wary of exposing myself to the possible legal
> risks involved in *any* substantive business project involving
> these software components.  The way these licences seem to be
> being revoked, changed or withdrawn, without adequate explanation
> is certainly a breach of the implied social "contract" created
> when software is published.

Yes, but that applies to all licenses, even non-free licenses from
well-known vendors.  The only thing that can provide assurances is a
contract.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> This is becoming metaphysical.  Cogito, ergo sum, and all that :) Your
> argument applies to all software that is not at least
> cryptographically signed by their author, and accompanied by a signed
> license statement :)

It applies only to software from authors who don't accept the section
'How to Apply These Terms to Your New Programs' from the file COPYING.

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Simon Wright]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> - TASH, the Tcl Ada SHell [6], includes a binding to Tk; it is
>   available as a Debian package, under GMGPL, from Ada-France [7] and
>   AdaWorld [8].  TASH is portable, but has not been updated since
>   2003.

TASH is in the process of moving to SourceForge at 
http://tcladashell.sourceforge.net/ 

I welcome anyone who wants to join in! My interest is in extending
Tcl/Tk with commands written in Ada so as to eg have a Tk gui to an
Ada application, rather than using Tcl features in Ada applications,
so there's a gap there ...

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Björn Persson]

Georg Bauhaus wrote:
> Yet, what about a collaborative effort to produce good general
> purpose Ada libraries to be used in Ada programs,
> irrespective of the compiler. Hm. A PAL with funding from everyone
> involved, including compiler makers and their customers.

If a project were to be started with the purpose of writing and 
maintaining GMGPL libraries, would you donate some of your time?

-- 
Bj�rn Persson                              PGP key A88682FD
                    omb jor ers @sv ge.
                    r o.b n.p son eri nu

Re: AdaCore ... the Next SCO?

[Jeffrey Creem]

Ludovic Brenta wrote:
> Jeffrey Creem wrote :
> 
>>Hopefully someone at Greenhills is paying attention to this discussion.
>>This confusing license perhaps "exposing" a company to GPL terms when
>>the headers clearly are not GPL will make a great writeup that will
>>pretty much make it impossible to even use GNATPro within my company.
> 
> 
> I don't understand. Did your company not receive a license statement
> from AdaCore along with GNAT Pro? Which company do you think is being
> "exposed to the GPL terms"? And what does GreenHills have to do with
> it? I wouldn't expect any problems, since AdaCore and GreeHills have a
> formal agreement.


I should have been clearer here (though I think others have somewhat 
answered some of this).

1) Actually, I am pretty sure when we have purchased GNATPro in the past 
that we have not gotten a license agreement **signed** by someone in 
authority at AdaCore. Certainly we do get license agreements.

2) I mentioned Greenhills because they have been publishing papers about 
this risks of Open source/Linux (http://www.ghs.com/linux.html) and or 
Non-Greenhills products(some would call this stuff FUD. At times I agree 
but generally you can't argue with the basic truths in the papers. You 
probably can disagree with some of the conclusions).

3) The biggest reason this murky license is likely to eventually result 
in AdaCore being banned is that many large companies are already leary 
of anything marked GPL in any fashion.  (e.g. http://tinyurl.com/jk9xy 
Note that this tiny URL is pointing to a PDF at a lockheed martin 
server)..I'll save you the trouble of looking. The summary is that they 
ban their suppliers from providing anything GPL to them. Yes there are 
ways out/around it..But it is not uncommon for this to be read as a 
defacto ban by some.

In any case, all of the assurances from the person who stands to make 
money on the SW won't matter if the FUD level (and truth level) gets 
high enough. There is an extremely good chance that another vendor's 
compiler will be picked due to the GPL "stink" on AdaCore and these 
misleading license terms on the CVS archive do nothing to help that cause.


> 
> 
>>At this point, it would seem the only purpose of the public CVS archives
>>is to entrap people to allow AdaCore to sue at will. Comments and
>>threads like this are of course likely to cause AdaCore to pull the CVS
>>archives and that will be somewhat of a shame for free software
>>developers but in its current state, the public CVS archives are doing
>>more harm than good.
> 
> 
> No, I don't think AdaCore want to sue anyone. I would rather think they
> are die-hard, purist Free Software believers, like RMS (i.e. "free up
> your software or pay"). Furthermore, they do not seem very eager to
> serve the market of SMEs and one-man shops, who have been asking for


Obviously one can't get into their heads but I'd be willing to bet that 
the change in terms has everything to do with business (not wanting to 
be their own competition) and nothing to do with free software purism. 
In fact, I sure hope that is the case since they are not a 
not-for-profit enterprise.  And I agree that they are unlikely to sue 
anyone but <wild speculation to illustrate a point>in 3 years when all 
the rest of the business dries up and we are all programming in XYZZY 
who knows the last dying breaths of the company might look like </wild 
speculation to illustrate a point>


> affordable GMGPL licenses for a long time. Well, that's their business
> decision. As Marc A. Criley said so rightly, this decision just leaves
> more room to Aonix and RR Software in that market.


True. Hopefully Aonix, RR Software, Greenhills and the other vendors 
remain vialble alternatives. Having more vendors is obviously a good thing.

> 
> I do agree that AdaCore's decision does more harm than good to Ada's
> attractivity to SMEs. They are the only ones being hurt. The change of
> license has no effect on corporations large enough to use GNAT Pro, or
> on students, hobbyists, or free software developers.
> 

I think I can say with high confidence that the change in license does 
indeed have an effect on corporations large enough to use GNATPro. It 
adds to the anti-gpl-of-any-kind FUD and will almost certainly cause 
other vendors to be selected in some cases. Hopefully the "free version" 
terms change gains them more customers than it costs them. I suspect it 
probably will.

Re: What luck we had ...

[M E Leypold]

"Jeffrey R. Carter" <spam.not.jrcarter@acm.not.spam.org> writes:

> M E Leypold wrote:
> > We didn't get the project (for various reason that had nothing to do
> > with Ada in this case :-)), but I suppose we have to count ourselves
> > really lucky that we didn't gat it: This customer would have come in
> > for multiple damages if he had found that his codebase got suddenly
> > contaminated. Of course we could have done it again in another
> > language, but then he'd have come in for contractual damages because
> > of the project being behind schedule.
> 
> A change to the license on future distributions does not change the
> license under which you obtained it. If you obtained a GMGPL version
> of GtkAda 2.4.0, then you had a GMGPL version, regardless of the
> license used later. If you kept that version, then you still have a
> GMGPL version. You also have the right to redistribute that GMGPL
> version granting the GMGPL to those you distribute it to.

I'd hope so, but that is exactly the thing that seems to be under
doubt at present. See all the other posts on that subject (not only
mine). AdaCore isn't helpful on that, also, since they don't remember
having ever distributed GMGPL version of anything. Take that together
with the reported attitude of "you cannot rely on COPYING, file
headers or accompaning files, only on letters from the licensor"
(whoever that might be), you see that a situation of doubt arises.

Whereas I'm convinced (after now finding the license notices from the
old sites in my archive) that indeed, the situation is as you
describe, there is no positive statement on that to be got from ACT. 

My statement above was half ironic. But only half: I'm indeed glad
that I've not to explain this customer an ambivalent situation in
which the supposed copyright holder cannot remember the former license
and our only assurance of our license status would be a backup copy on
a rather old and grubby tape or CD.

It smells like a situation where you'd need a lawyer for protection,
and the customer wouldn't like it. And of course would put everything
on the account of the unreliability of Open Source and the shiftyness
and even sneakiness of the GPL: One day you have GMGPL and suddenly
there is DOUBT (did I mention that customers often abhor doubt, since
it translates to risk in business?) and possibly what first was just
harmless code in the customers CVS might have become a GPL trojan
horse. I can explain all that. I can try to assure. But where is the
limit where the customer just has a better feeling not to use any Open
Source any more ("you never know what you get") and looks for clear
contracts and proprietary code bases again?

But never mind: The situation didn't happen anyway. (The comment on
damages for delay was real thaough, and it there is room for
discussion wether the _possibly_ changed license wouldn't have been a
good reason to ditch the whole project even months after
start. IANAL.)

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Tue, 2006-06-27 at 17:00 +0200, M E Leypold wrote:
> 
> >   # Alex Bykat who helped us compile GtkAda 0.2 on Solaris 2.5.
> >   # ...
> >   # Preben Randhol who has contributed several patches, in particular for porting gate to glade-2.
> 
> > Useful, this, isn't it?
> 
> Could make a change if members of the above list decided to
> make a GMGPL Edition from their earlier work.
> But just in case, this isn't useful as a licensing argument.
> A quote from the GNU site illustrates:


> "If you modify this code, you may extend this exception to your version
> of the code, but you are not obligated to do so. If you do not wish to
> do so, delete this exception statement from your version."

Wonderful. Now tell me what rights ACT had to strip the linking
exception from most files of florist. Please do that after diffing GPL
florist against 3.15p.

Also, why the GtkAda license changed w/o any change in source (2.4.0).

Thanks -- Markus

> So the release of a modified GtkAda may drop the
> exception without consulting the originators.

Which makes GMGPL it inferior to LGPL. :-) You contribute under
condition X and are never sure wether and when it will change to
condition Y.

> I can't say whether or not the members of the above list
> like the new facts.

Nor can I.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Martin Krischik <krischik@users.sourceforge.net> writes:

> Ludovic Brenta wrote:
> 
> >  The change of
> > license has no effect on corporations large enough to use GNAT Pro, or
> > on students, hobbyists, or free software developers.
> 
> Does it? The GPL is viral and booch, charles, AdaCL - all currently GMGPL
> would need to to relicensed to GPL now to be used with with GNAT/GPL.

Martin,

That is exactly why I asked earlier in the DTRAQ thread wether it is
really necessary to change GMGPL to GPL when linking sources with
these licenses in an excutable.

I never got an answer on that, instead got bogged down in questions
why I'd want that (I showed a possible scenario), the advice of the
kind "why don't you just buy ..." and discussions about the fairness
of it all.

But perhaps I did not ask explicitely or clearly enough, so let's try
again:

Can I:

  - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
    under the GMGPL whereas S3 is GPL.

  - I then compile S1, S2, S3 to X. 

  - Obviously X is covered by GPL: I must distribute S1, S2, S3 with X
    as the GPL demands.

  - But cant' I state that S1, S2 are under GMGPL -- that is, anyone
    receiving them is allowed to unbundle them from the source package
    of X (which is actually made up from 3 different trees) and can
    distribute them (S1, S2) as GMGPL sources or create other GMGPL,
    GPL or even closed executables from them.

Question: Is there any contradiction in the license terms or is that
permissible? 

Of course one would want to do that for whatever reason ever. I do not
want to discuss the fairness or unfairness of that or what one could
expect the authors of S1, S2 to do or not to do. I just want to know,
wether the authors of S1 and S2 (if they want) license S1 and S2 as
GMGPL and be sure that they can be used in aforesaid manner.

Or do they have to license as GPL to ensure "linkability" with a GPL
library?

Note that I thing, that the builder of the excutable X cannot strip
the linking exceaption from the libs S1 and S2 since he/she has not
changed the libs. He would have to refrain from using them if he is
not allowed to link with S3.

> AdaCL at least produced licence-warnings when compiled with GNAT/GPL. While
> licence warning are not legal the at least shows how AdaCore sees things.
> And this would mean that those libs become unavailable to closed source
> users.

ACT seems to disallow the situation I've been sketching above.

> 
> Martin
> 
> PS: Just in case you have not seen any warnings so far: all package
> specifications copied from the RM come with licence "unrestricted" and they
> are free in all respects. Only when you use a packaged not mentioned inside
> the RM the warning pops up.


Ah!

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:
> 
> >  and booch, charles, AdaCL - all currently GMGPL
> > would need to to relicensed to GPL now to be used with with GNAT/GPL.
> 
> I don't think that any GMGPL sources need to be relicensed
> to be used with GNAT GPL Edition for free software work.
> The GMGPL is certainly GPL compatible.


Martin asks wether it really is. I asked the same question once her,
never got an answer (I was not clear, perhaps). The warnings reported
by martin seem to indicate otherwise (or at least a different view of
ACT on the subject).


> > And this would mean that those libs become unavailable to closed source
> > users.
> 
> But that's a consequence of using GNAT GPL edition, not of using
> the libraries? These can keep their licenses. (The GPL doesn't say
> it is time limited, and revocable; I'd be surprised to learn that
> the exception is time limited.)


No. Just reread Martins post about the license warnings popping up
(and perhaps my follow up which tries to state the problem in more
ebatract terms). The fear is, that GMGPL libs must be relicensed to
GPL to be linkable with GPL libs. And we're talking about the new
GtkAd here not the Gnat GPL.

I do not say that relicensing would be necessary, but I'd like to see
that point cleared up.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

"Dr. Adrian Wrigley" <amtw@linuxchip.demon.co.uk.uk.uk> writes:
> > 
> > As I explained above, I don't know, and AdaCore refused to tell me. Ask
> > each author individually.
> 
> I'm sorry you have been put in this position - relaying and interpreting
> apparent changes in licensing conditions.  The issues raised here

Yes. He has my compassion. Just to find myself in the same position
and since I cannot trust mere rumor on c.l.a. (remember: Have to ask
the licensor, directly!) I have written to ACT to ask the same or
perhaps slightly different questions. I looking forward to my
answer(s).

> are potentially of very wide interest to many millions of licencees
> and licensors under the GPL and other Free software licences.


> Under English law (and presumably most other places), signed statements
> are not required to form a contract.  In particular, if the parties
> behaved and believed that there is a contract then one exists.
> Evidence such as files, ftp sites, emails etc. can help support
> the claim that a contract existed.  Surely the behaviour of
> the authors and users backs the claim that licences were granted?
> Signatures on bits of paper might help, but still don't provide
> a full guarantee.  IANAL.

I agree.


> As regards the GPL, it appears to be a perpetual, sub-licensable,
> non-revocable (absense of breaches) licence.  The GMGPL terms
> add to this, but don't change these basic features.

That was my impression.

> I am at a loss to understand what basis there is for revoking the
> licences already issued.  As a party to the licence

Well -- it seems, the situation is more complex. The source they
distribute now is unchanged BUT with another license (IMHO you can
only drop the exception clause when you change something, but perhaps
that does not apply to the copyright owner).

They don't say the other licenses where not GMGPL or are revoked, They
only don't remember.

> contract between myself and the authors, I feel aggrieved.

So I've been feeling for some time, especially since I have been
repeatedly lectured here on how unreasonable I've been to have
expected more than GPL .. -- What? I've not been expecting GMGPL, I do
have copies of web pages from ACT _telling_ me that I'm licensing
unter GMGPL and I might be forgiven that I find that loss of memory on
ACTs part slightly unsettling.

> I have kept to my side of the bargain.  I'm not convinced they
> have kept to theirs.

:-)


> Some people here will want a formal opinion.  I don't feel I have
> need or resources for a professional view myself.  But I am being
> made extremely wary of exposing myself to the possible legal
> risks involved in *any* substantive business project involving
> these software components.  The way these licences seem to be

That is exactly what I've been saying (look at my post with "luck" in
the title). I _am_ glad that I just barely missed a situation where
I've to explain all this to a certain customer, regardless of wether
we can resolve that with a lawyers help to a point where we see that
we still have rights to use GtkAda under GMGPL.

It's the lawyer that will make the customer very skittish indeed.

And there is my point that all this is damaging Ada beyond
redemption. You can all deny that. You can all say that the hobbyist
can still use that and that the commercial user can still license this
or should be able to pay that. At the end of the day this are just all
very convincing attempts to close the eyes to the inevitable: The
situation as it is, generates uncertainty and doubt. Fear will follow
and I predict the last commercial (and no, Government, NASA, ESA and
defense industry DON'T count, this is all a very closed and special
market of its own) -- I predict that after the water have reached a
certain amount of muddiness, the commercial users will leave Ada as
language or not adopt it, as will the hobbyist with a professional
attitude (the boundaries aren't quite so clear indeed).

> being revoked, changed or withdrawn, without adequate explanation
> is certainly a breach of the implied social "contract" created
> when software is published.

Thanks. You expressed my opinion very clearly.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> Dr. Adrian Wrigley writes:
> > I'm sorry you have been put in this position - relaying and interpreting
> > apparent changes in licensing conditions.  The issues raised here
> > are potentially of very wide interest to many millions of licencees
> > and licensors under the GPL and other Free software licences.
> 
> Yes :( But it also applies to non-free EULAs, or any kind of license

> A license is not the same as a contract.  A contract has to have

Before we talk so all inclusive on contract an licnesing laws, I'd
remind evryone that GtkAda has not been distributed in the US
only. European contract law, German contract law and the basic
concepts behind it are very different.

I do not say, that we shouldn't discuss our respective rights. But
even if the conclusion might hold in another jurisdiction, the
reasoning might be unparsable.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:
> 
> Yes, but that applies to all licenses, even non-free licenses from
> well-known vendors.  The only thing that can provide assurances is a
> contract.

I disagree. I am not a lawyer, but a public declaration of intent to
grant the public certain rights should actually suffice to constitute
a right. That is the way "putting something into the public domain"
works and even if the GPL is more complex I doubt that you have to
have a contract to have something to stand on legally.

But IANAL. Don't take that as legal advice.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Michael Bode <m.g.bode@web.de> writes:

> "Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:
> 
> >> 4) all licencing terms embedded in the distributions are repudiated?
> >
> > They never had any legal force; only a signed statement from the
> > copyright holder has legal force.
> 
> 
> From a well know file named COPYING (note paragraph 2):
> 
> 	    How to Apply These Terms to Your New Programs


>   To do so, attach the following notices to the program.  It is safest
> to attach them to the start of each source file to most effectively
> convey the exclusion of warranty; and each file should have at least
> the "copyright" line and a pointer to where the full notice is found.

Ah bah -- Michael,

You know that the FSF doesn't have a clue? :-). Just joking -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Michael Bode <m.g.bode@web.de> writes:

> "Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:
> 
> >> 4) all licencing terms embedded in the distributions are repudiated?
> >
> > They never had any legal force; only a signed statement from the
> > copyright holder has legal force.
> 
> Oh, and BTW I suppose the following then also has no legal force:
> 
> -- This library is distributed in the hope that it will be useful,   --
> -- but WITHOUT ANY WARRANTY; without even the implied warranty of    --
> -- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU --
> -- General Public License for more details.                          --


Ah yes, that has no legal force. Good idea. I'll sent ACT a bill for
the time I spent debugging 3.15p. After all their exclusion of
warranty had no legal force.

What a absolutely /&$"/&�$!"/�$ <censored>.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> Florian Weimer <fw@deneb.enyo.de> writes:
> 
> > * Ludovic Brenta:
> >
> >> They don't say that, but they refuse to give details on when the switch
> >> took place. I don't know. Ask them. From a theoretical standpoint, I am
> >> indeed quite worried that I have downloaded and redistributed AdaCore's
> >> software, thinking in good faith I had the right to do so when in fact
> >> I didn't, since I didn't have written permission from any of the
> >> copyright holders.
> >
> > You didn't download from Adacore, I think, but from NYU.
> 
> I did download quite a lot from AdaCore: GtkAda, AWS, XML/Ada, and
> also GPS but that's not a library.


Ah, but did you download from libre or from libre2? :-)

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ed Falis <falis@verizon.net> writes:

> M E Leypold wrote:
> 
> >> AUnit was written by Ed Falis, an AdaCore employee.  Only source is
> >> AdaCore, so switch to pure GPL.
> >
> > Also a tool, wouldn't get linked into an executable for the customer:
> > In my eyes GPL is OK here.
> 
> Note that AUnit has been GPL from its first release in 2000.  Since its
> intent is user "in-house" testing rather than any kind of
> re-distribution of test cases, I considered it the appropriate license.

Ed, I agree. 

Tools for inhouse use are appropriately GPL IMHO. We don't want anyone
to run away with tools chain after all :-). The issues here -- where
all that started -- are different (not with new licenses but with the
possible retroactive change of old licenses, license changes on
unchanged source, missing or misleading license notices and a general
amnesia).

I'm sure you'll see that after wading through all that fallout. 

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> M E Leypold writes:
> > Ludovic Brenta writes:
> >> - They refuse to give any assurances regarding copyright ownership, so
> >>   I feel that I now need to go ask the authors.
> >
> > Please. Also It would be useful to ask AdaCore to document the
> > provenience of the sources they distribute: At which distribution
> > points the got the source, what are their assurances that they are
> > allowed to distribute the sources under GPL and so on :-). What their
> > baseline version is (that is from which version on they started to
> > "contribute" to the code base).
> 
> I'd rather spend my time on the technical side of things.  I'm a

Me too.

> software engineer :-)

So am I. Unfortunately clean beaselines seem to be required to sort
that mess out, excapt if you want to go to GPL in all aspects (which I
won't, I'll rather drop Ada).

> I'll see how much interest there is in GMGPL versions of libraries,
> and decide then.  You can ask these authors directly if you want to.

I'll poll at c.l.a within the next week on how many people are
intrested in GMGPL baselines. If there is intrest I'll try to
establish them / ask authors.

If you want to poll for intrest in GMGPL versions of the libraries,
please do so.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Tue, 2006-06-27 at 20:39 +0200, Michael Bode wrote:
> > "Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:
> > 
> > > I received detailed answers from AdaCore's Robert Dewar and Arnaud
> > > Charlet regarding the licenses of software downloaded from their
> > > servers.  In summary:
> 
> >  How is any
> > Debian user supposed to know any AdaCore software is GPL if file
> > headers are invalid?
> 
> The file headers aren't invalid. That sounds like nitpicking,
> but "invalid" has different meaning.

Like?

> only when someone uses software in a way that by applicable
> law is (a) illegal, and (b) this fact is made a legal issue.

Sorry. That is, in example, quite relevant, if, in example, someone
just picks the files with linking exception from the package and
creates a GMGPL binding from that.

Which he/she should be allowed to do, if the headers have any meaning.

> As long as Debian doesn't have any legal obligations, they can
> proactively try to reduce the risk of being associated with software
> that incurs a surprising legal status.

Man, man. The "surprising legal status" is something that should never
have happened. And forgive me: It's not Debian which is at fault here.

> I think that one way to do this is to collect all available
> evidence, and take Dewar's and Charlet's word for it.  

For what? They are not actually giving any useful words, AFAIS.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Tue, 2006-06-27 at 21:39 +0200, Michael Bode wrote:
> 
> > Debian is distributing the software. If at some later point in time it
> > is claimed that the software today was not under GPL but some other
> > license which forbids distribution they have a problem.
> 
> Yes, they will have a problem if someone accuses them of distributing
> software they weren't allowed to distribute. But...
> 
> >  Seems like
> > this has just happend with GtkAda and GMGPL. 
> 
> I'm not so sure. First, I don't think that the maintainers of Ada
> stuff in Debian will have a problem. By design, they never distribute
> anything without providing access to the source code as well.
> So GMGPL or GPL is not a legal issue at all from a Debian maintainer's
> perspective, as long as the issue is whether it is GPL or GMGPL.

May I play the sweet melody of damages ACT might (in extremis) want
from the Debian maintainers (not that I believe that, but you were the
one starting to analyze the legal situation). Debian has distributed
the packages in question under a license which is probably more
liberal than ACT would have it to be. Other used them in good faith in
proprietary software. Two parties now have problems: The ones whose
software is now suddenly open and ACT whose software has been used
"illegally".

Don't fear: I don't have that situation, I'll not sue or whatever. But
proposing that Debian could not have incurred some liability is a bit
naive.

So Debian is right to go the safe way, even if that just illustrates
the power of FUD.


> Now if you wanted to make a claim that in spite of messages from
> AdaCore officials to the contrary, software downloaded from the
> AdaCore site is neither GPLed nor GMGPLed, then I think everyone
> will be eager to hear what else it is.

In a year from now? Who knows. You see: Files in the ditro and file
headers are not binding.  E-Mails from ACT probably have a similar
ephemeral status if not signed with strong crypto. So you'll never
know what the future brings.
> 
> > > to collect all available evidence,
> > > and take Dewar's and Charlet's word for it.
> > 
> > They only can collect bits which are not signed with strong crypto.> 

> I see the irony, but this is legal stuff. Consider a few sheets of
> paper landing on your desk saying, this is a part of the NT kernel.
> This part of the NT kernel source, as the header clearly indicates,
> is GPLed. Would you believe it?

No. On the other side I think, that MS would take action on that very
soon, when it becomes public, whereas the GtkAda source has been in
public with ACTs knowledge for quite a time. They know what the
headers say since version 1.3.x and I find it not creditable that they
never wondered about their meaning, sort of.

Implied consent?

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Tue, 2006-06-27 at 22:23 +0200, Michael Bode wrote:
> 
> >  I
> > claim that a sudden, secret and maybe retroactive (or maybe it was
> > between 2.4.0 and 2.4.1?, AdaCore won't tell us) change of license
> > happend to GtkAda. And if that happens once ...
> > 
> > Regarding messages from AdaCore to the contrary of anything:
> > 
> > http://web.archive.org/web/20050224092046/http://libre.act-europe.fr/GtkAda/
> > 
> 
> Ludovic lets us know a response from AdaCore that uses the _present_
> tense,
> 
> - All software downloaded from AdaCore is pure GPL, no matter what the
>   headers say.
> 
> I don't think that _if_ there was a legal status at a time in the
> recent past, this status can be declared something different.

Fine. Problem is the amnesia presently ravaging ACT headquarters. They
don't remember the past or don't talk about it. It certainly would be
going out on a limb if anyone distributes GtkAda 2.4.0 (from old
libre) now as GMGPL which is exactly the same version as 2.4.0
deistributed from ACT today which is GPL and relies on his/her memory
only regarding the license the thing had when it was downloaded. After
all, the licensing statements in the tarball itself are far from clear
(which started all that hullabaloo in the very beginning).


> These things should happen only after one legal system has been
> dropped in favor of another, for example after a war.  I am aware
> that many of the software packages used to be advertised as being
> GMGPL.

Well. Can I call you as witness? Will that suffice? -- You see my
point.


> With Debian in particular, I think there might be more issues, like
> you don't want to stress your good relations with AdaCore, if any,

That certainly seems to be on of the factors in a number of cases.

> if you are maintaining Debian Ada software that nowadays is mostly
> produced by AdaCore.

> This, I guess, is just a business issue for some who are involved.

Which perhaps should cut both ways. ACTs distributions never installed
flawlessly in a give system, so they should be _very_ grateful for the
work of debian maintainers which are doing a lot of free advertisment
for ACT by providing well maintained no-hassle installations of ACTs
public versions.


> Yet, what about a collaborative effort to produce good general
> purpose Ada libraries to be used in Ada programs,
> irrespective of the compiler. Hm. A PAL with funding from everyone
> involved, including compiler makers and their customers.
> We've all got to make a living, haven't we?

I would participate, but not if I can't sell a closed programm now and
then (or at least assure the customer that I am allowed to sign away
my rights to him). But I don't suppose you'd want GMGPL libraries, would you?

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> Michael Bode <m.g.bode@web.de> writes:
> > How can any AdaCore software then remain in Debian at all? How is any
> > Debian user supposed to know any AdaCore software is GPL if file
> > headers are invalid? How do *you* know? You have nothing written and
> > signed and the authors deny any validity of anything written in the
> > software itself.
> 
> This is becoming metaphysical.  Cogito, ergo sum, and all that :) Your
> argument applies to all software that is not at least
> cryptographically signed by their author, and accompanied by a signed
> license statement :) (note: software distributed by Debian _is_
> cryptographically signed, but that from upstream is not.
> 
> AdaCore make no claim as to the license terms for software I download
> from other sites than AdaCore.  And I don't think they can

Which leaves ypu with the fact that you don't know anything about
that, since the licensing terms in the GtkAda source tarball are far
from clearly stated.

> retroactively revoke the licenses that Debian received when I
> initially downloaded the software from NYU, AdaCore, or other places.

One would think. I think Michael has just overstating the problem
behind all that, but you see it too, don't you? I'd go as far as
saying, that ACT did purposefully muddy the water, since it is hardly
believable, that they don't know what and how they distributed from
libre.act-europe.fr. A statement of "it now like this, everthing up to
version x has been GMPL and/or everything you got from
libre.act-europe.fr or before $DATE is GMGPL" would not have cost them
much. But they didn't.

> But IANAL.

IANAL2.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Michael Bode <m.g.bode@web.de> writes:

> Ludovic Brenta <ludovic@ludovic-brenta.org> writes:
> 
> > This is becoming metaphysical.  Cogito, ergo sum, and all that :) Your
> > argument applies to all software that is not at least
> > cryptographically signed by their author, and accompanied by a signed
> > license statement :)
> 
> It applies only to software from authors who don't accept the section
> 'How to Apply These Terms to Your New Programs' from the file COPYING.

If I think about that, that would make the situation clear. The
COPYING file says that the copyright is in the headers, so the headers
are binding. These have a linking exception, so the single files are
under GMGPL. Just delete any non-GMGPL stuff from GtkAda that would
get linked to your executable and then you have GMGPL-GtkAda [1].

Hm. I think we'll just have to ask a lawyer about that.

Michael, mail me privately on that. I got me an idea :-).

Regards -- Markus


PS: [1] I'm not alawyer. Don't construe that legal advice. It isn't.

Re: AdaCore ... the Next SCO?

[M E Leypold]

> 
> 3) The biggest reason this murky license is likely to eventually
> result in AdaCore being banned is that many large companies are
> already leary of anything marked GPL in any fashion.
> (e.g. http://tinyurl.com/jk9xy Note that this tiny URL is pointing to
> a PDF at a lockheed martin server)..I'll save you the trouble of
> looking. The summary is that they ban their suppliers from providing
> anything GPL to them. Yes there are ways out/around it..But it is not
> uncommon for this to be read as a defacto ban by some.
> 
> In any case, all of the assurances from the person who stands to make
> money on the SW won't matter if the FUD level (and truth level) gets
> high enough. There is an extremely good chance that another vendor's

Apart from the fact that you can assure what you want: There is a
point from where the insureance of the person supplying the software
won't cover what they perceive as risk any more.

> compiler will be picked due to the GPL "stink" on AdaCore and these
> misleading license terms on the CVS archive do nothing to help that
> cause.

Full ACK.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Tue, 2006-06-27 at 22:53 +0000, Björn Persson wrote:

> If a project were to be started with the purpose of writing and 
> maintaining GMGPL libraries, would you donate some of your time?

Certainly. In a few month from now I should find some time.

Re: AdaCore ... the Next SCO?

[Simon Wright]

Jeffrey Creem <jeff@thecreems.com> writes:

> Hopefully someone at Greenhills is paying attention to this
> discussion. This confusing license perhaps "exposing" a company to GPL
> terms when the headers clearly are not GPL will make a great writeup
> that will pretty much make it impossible to even use GNATPro within my
> company.

I don't see this, GNATpro clearly comes with the sort of licence we
need (GMGPL and all). Noone could use it if it didn't.

The fear of GPL infection remains, of course.

Re: AdaCore ... the Next SCO?

[Simon Wright]

Martin Krischik <krischik@users.sourceforge.net> writes:

> Does it? The GPL is viral and booch, charles, AdaCL - all currently
> GMGPL would need to to relicensed to GPL now to be used with with
> GNAT/GPL.

I don't see that. As far as the BCs are concerned, all the
contributors (including Grady) are happy with the GPGPL and see no
reason to change. After all, the BCs will compile with the Aonix
compiler so there's not even any real dependency on GNAT extensions
(and it would make no difference if there were).

The GMGPL says that inclusion of itself doesn't impose GPL obligations
but doesn't affect other reasons that might. Issuing a binary compiled
with GNAT GPL would be such a reason.

If I release a unit to the public domain, I don't have to re-license
it just because I've compiled it with a GPL compiler at some point.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> I see the irony, but this is legal stuff. Consider a few sheets of
> paper landing on your desk saying, this is a part of the NT kernel.
> This part of the NT kernel source, as the header clearly indicates,
> is GPLed. Would you believe it?

If I had downloaded it myself from
https://gnu.microsoft.com/free_software/windows_nt/src/ ?

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Simon Wright]

Since the BCs don't use the non-standard pragma License (Licence?) the
question doesn't arise. And given this discussion they are not going
to.

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Georg Bauhaus wrote:

> >  and booch, charles, AdaCL - all currently GMGPL
> > would need to to relicensed to GPL now to be used with with GNAT/GPL.
>
> I don't think that any GMGPL sources need to be relicensed
> to be used with GNAT GPL Edition for free software work.
> The GMGPL is certainly GPL compatible.

No they are not. Look at this simple code sniplet:

with GNAT.OS_Lib;

package My_OS_Lib
   package Pass_Thrue  renames GNAT.OS_Lib;
end My_OS_Lib;

I used a rename to make shure the problem is exposed.

> Even if you run into a pragma License(some invalid identifier),
> you just create a derivative work by placing '--' before the
> pragma. Just make sure that the result, if distributed, is
> distributed as required by the GPL.

Right: My work has to become GPL as well.

Martin

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Simon Wright wrote:
> Since the BCs don't use the non-standard pragma License (Licence?) the
> question doesn't arise. And given this discussion they are not going
> to.

In the absence of pragma License GNAT scans the commens for magic words
which indicate the lincence used ;-) . As it is the BC will be
considered GMGPL by GNAT. But since no GNAT specific packages are used
no licence warnings are issued.

Martin

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Simon Wright wrote:

> The GMGPL says that inclusion of itself doesn't impose GPL obligations
> but doesn't affect other reasons that might. Issuing a binary compiled
> with GNAT GPL would be such a reason.

Yes, of corse - i missed that.

Martin

No relicense everything now! (was: Re: AdaCore ... the Next SCO?)

[Georg Bauhaus]

On Wed, 2006-06-28 at 00:28 -0700, Martin Krischik wrote:

> > The GMGPL is certainly GPL compatible.
> 
> No they are not.

>From a licensing point of view, the GMGPL is compatible with
the GPL.
You just can't arbitrarily add the GMGPL exception to GPLed
source that isn't yours.

No need to make the BCs GPLed.
GNAT GPL Edition has no technical problem compiling any
GMGPLed code.
The GMGPL sources do not influence the license status of
the GNAT run-time system, for example.

>  Look at this simple code sniplet:
> 
> with GNAT.OS_Lib;
> 
> package My_OS_Lib
>    package Pass_Thrue  renames GNAT.OS_Lib;
> end My_OS_Lib;
> 
> I used a rename to make shure the problem is exposed.

There is no problem with the BCs license:
If someone wishes to compile closed source binaries for
distribution using GNAT GPL Edition, they can't.
Reason: the GNAT GPL Edition is not licensed for closed source
software development.
The BCs could be in the public domain. That wouldn't change
anything.


> Right: My work has to become GPL as well.

But not the BCs!

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Dr. Adrian Wrigley]

On Tue, 27 Jun 2006 23:18:37 +0200, Ludovic Brenta wrote:
 
> A license is not the same as a contract.  A contract has to have
> measurable obligations for both parties; a license is only a grant of
> rights to the licensee, possibly without compensation.  No
> compensation, no contract, either written or implied.

After I sent my message, I found this article:
http://www.advogato.org/article/606.html

I have changed my position slightly.

It doesn't seem clear to "the experts" whether GPL is perpetual
or revocable or not :(   And it doesn't seem clear whether it is
a simple copyright licence, or is part of a contract :(
But it is clear that significanty different interpretations would
be placed on the GPL by different legal systems and different
(US) states :(

I do believe that there is "consideration" required of a (English)
contract. GPL *obligates* the licensee to make available source code to
everyone to whom a binary is distributed.  And it *obligates* the licensee
to license source modifications to the recipients of binaries, and
*obligates* the licensee to grant the rights to pass the source to third
parties.  These obligations on the licensee can be extremely valuable to
the licensor.  But the lack of any obligation on the licensee to
distribute in the first place may invalidate the contingent obligation.
Perhaps the contract is formed when the licensee takes actions of
possible benefit to the licensor (such as distribution, modification etc)?
In the end, I don't know how important the contract question is, but
it is unquestionable that damages are caused to parties who seem to
have lost rights to distribute their Ada programs by the recent changes.

The Free Software Definition from gnu.org says:

"In order for these freedoms to be real, they must be irrevocable as long
as you do nothing wrong; if the developer of the software has the power to
revoke the license, without your doing anything to give cause, the
software is not free."

It seems pretty clear from the general industry practice in software
that GPL is not revocable.  The Ada software in question does not
meet this definition of Free Software if the GMGPL licences really
have been revoked.

The strange thing is that nobody seems to benefit from the changes
and the consequent FUD :(
--
Dr. Adrian Wrigley,
Cambridge.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Wed, 2006-06-28 at 10:51 +0000, Dr. Adrian Wrigley wrote:

> "In order for these freedoms to be real, they must be irrevocable as long
> as you do nothing wrong; if the developer of the software has the power to
> revoke the license, without your doing anything to give cause, the
> software is not free."
> 
> It seems pretty clear from the general industry practice in software
> that GPL is not revocable.  The Ada software in question does not
> meet this definition of Free Software if the GMGPL licences really
> have been revoked.

There is no instance of anything being revoked, AFAICS, if AdaCore
is entitled to drop the exception from the source they distribute,
which might feel like something has been revoked. I doubt this
is an adequate description of the (GM)GPL legal situation.

The license text is the GPL in either case, GPL or GPL with the
special exception.

If any license has been granted to anyone, and they have been told
their license has been revoked, although they have done nothing
wrong, please speak up.


> The strange thing is that nobody seems to benefit from the changes
> and the consequent FUD :(

Right. This is why this thread should not continue to try to construct
or reconstruct, or invent, what supposedly was or is legally relevant
information in some past, freely mixing the past and the present.
All we have is hints, and AdaCore notes from two different points in
time.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Wed, 2006-06-28 at 13:14 +0200, Georg Bauhaus wrote:
> On Wed, 2006-06-28 at 10:51 +0000, Dr. Adrian Wrigley wrote:

> > It seems pretty clear from the general industry practice in software
> > that GPL is not revocable.  The Ada software in question does not
> > meet this definition of Free Software if the GMGPL licences really
> > have been revoked.
> 
> There is no instance of anything being revoked, AFAICS, if AdaCore
> is entitled to drop the exception from the source they distribute,
> which might feel like something has been revoked. I doubt this
> is an adequate description of the (GM)GPL legal situation.

Does someone know whether the right to use something is the
same as a license, and on what continent, or island?

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Michael Bode:

> How can any AdaCore software then remain in Debian at all?

Debian does not check the precise copyright status of most packages.
I doubt many authors are willing to provide assurances without proper
compensation, especially if they have incorporated any contributions
from third parties (which can pose legal risks even if you've got
statements in writing to the contrary).

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Georg Bauhaus wrote :
> Does someone know whether the right to use something is the
> same as a license, and on what continent, or island?

The right to use falls outside of the scope of copyright law. Copyright
law only deals with the right to copy, modify and redistribute. This is
explained in the preliminary notes of the GPL.

If a licensor wants to restrict or control usage of their software, the
only way they can do this is by signing a contract with the licensee.

For example, Java's license has, or used to have, a clause that forbids
use in the nuclear industry or in life-critical applications. Such
clause is null and void from a legal perspective, because the licensor
has no right to control or restrict usage of their software, but only
copying, modification, and distribution. (there are of course good
technical reasons why only a pointy-haired manager would use Java in
life-critical applications, but that's not the point).

-- 
Ludovic Brenta.

Re: AdaCore ... the Next SCO?

[M E Leypold]

Simon Wright <simon@pushface.org> writes:

> Jeffrey Creem <jeff@thecreems.com> writes:
> 
> > Hopefully someone at Greenhills is paying attention to this
> > discussion. This confusing license perhaps "exposing" a company to GPL
> > terms when the headers clearly are not GPL will make a great writeup
> > that will pretty much make it impossible to even use GNATPro within my
> > company.
> 
> I don't see this, GNATpro clearly comes with the sort of licence we
> need (GMGPL and all). Noone could use it if it didn't.

Well -- if you, as JC seemed to imply -- are working for a company
that is providing programming service to large aerospace companies and
of those some in turn seem to have a "no GPL" policy in place now, as
other posts seem to imply: You wouldn't want to explain to them the
difference between Gnat GPL and Gnat Pro. You'd like to avoid the "GPL
stink", someone put it here, the association to 

I'm not sure wether large contractors actually audit their
subcontractors development tools and process. But I wouldn't be
surprised.

> The fear of GPL infection remains, of course.

Exactly. If ACT had spoke up clearly or had clear GPL licenses for
_new_ version on their side, perhaps a short note announcing the
transition etc., they would have got flak earlier, but it could have
been sorted out. As it is, fear, uncertainty and doubt, the three
apocalyptic riders of information technology, are at large again.

You might not see it, but I take other posts as confirmation, that
other people also see potential users of Ada (be they commercial or
student or hobbyist) turned away by that. Nobody wants to learn riding
with a dead horse.

Regards  -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Martin Krischik" <krischik@users.sourceforge.net> writes:

> Simon Wright wrote:
> > Since the BCs don't use the non-standard pragma License (Licence?) the
> > question doesn't arise. And given this discussion they are not going
> > to.
> 
> In the absence of pragma License GNAT scans the commens for magic words
> which indicate the lincence used ;-) . As it is the BC will be
> considered GMGPL by GNAT. But since no GNAT specific packages are used
> no licence warnings are issued.

If if there would be: I do not grok it: Linking GMGPL with GPL
libraries should put the excutable under GPL, but the GMGPL part of
the source would still be under GMGPL, wouldn't it? Why the license
warnings then with GNAT specific units?

Regards -- Markus

Re: No relicense everything now! (was: Re: AdaCore ... the Next SCO?)

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Wed, 2006-06-28 at 00:28 -0700, Martin Krischik wrote:
> 
> > > The GMGPL is certainly GPL compatible.
> > 
> > No they are not.
> 
> >From a licensing point of view, the GMGPL is compatible with the GPL.

> You just can't arbitrarily add the GMGPL exception to GPLed
> source that isn't yours.

George, would you please listen to Martins point and not let fly on
the first sentence you read? Martin say that the warnings Gnat emits
in certain situation, seem to indicate that the _GMGPL_ source would
have to relicenced as GPL. _Not_ the other way around.

Wether that is so or not, is open to debate yet. Perhaps Gnat is wrong
in issuing that warning. But please stick to the point and try to
muddle Martins argument. I'd like that to get cleared up as well.


> No need to make the BCs GPLed.

<snip>

> 
> > Right: My work has to become GPL as well.
> 
> But not the BCs!

And if My  = Grady's ?

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

"Dr. Adrian Wrigley" <amtw@linuxchip.demon.co.uk.uk.uk> writes:

> The strange thing is that nobody seems to benefit from the changes
> and the consequent FUD :(

Exactly. I still wonder why AdaCore doesn't speak up clearly, saying
when the license change occurred, letting the old versions under GMGPL
(even if they now come from their site). Yes, there is a certain
probability of a community supported version arising. It won't hurt
them, since I don't think those wanting to buy Gnat Pro support would
really want a community supported version (read: slowly maintained, no
support on demand, no new Gtk features) anyway.

My guess would be that they actually forgot the anon cvs and forgot to
strip the linking excaption there and are now entrenching themselves
in a probably untenable legal theory that (a) the license notices in
the files have no meaning whatsoever and (b) everything is GPL now,
even the old versions, because they don't want anyone to become wise
on that and just compile himself GtkAda from the CVS source to get a
GMGPL library. (IANAL).

I'd wish that a really big player would do exactly that. It'd be
against the spririt of free software (I usually try even to get the
authors agreement if I'm mirroring sources from their distribution
sites) and an act of blatant and arrogant highway robbery, but it
would be a sort of just punishment. That wont happen, I know, but one
can dream. (I bet it would happen if I forgot to put my license
notices right).

Looking at the CVS I tend to think that even the 2.8.x source should
still be GMGPL mostly. But IANAL. 

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Wed, 2006-06-28 at 10:51 +0000, Dr. Adrian Wrigley wrote:
> 
> > "In order for these freedoms to be real, they must be irrevocable as long
> > as you do nothing wrong; if the developer of the software has the power to
> > revoke the license, without your doing anything to give cause, the
> > software is not free."
> > 
> > It seems pretty clear from the general industry practice in software
> > that GPL is not revocable.  The Ada software in question does not
> > meet this definition of Free Software if the GMGPL licences really
> > have been revoked.
> 
> There is no instance of anything being revoked, AFAICS, if AdaCore
> is entitled to drop the exception from the source they distribute,
> which might feel like something has been revoked. I doubt this
> is an adequate description of the (GM)GPL legal situation.

Come on. They don't want to admit they ever distributed under
GMGPL. Put that together with their theory of "you must always check
with te licensor", than that is nearest you can get to a revocation,
without actually saying so. Of course it's none, but it makes
FUD. It's not good enough for telling third parties that a given lib
is under GMGPL.

> The license text is the GPL in either case, GPL or GPL with the
> special exception.

We know, that you don't want to see the difference. Since you don't
care for it, you'd also not see that anything changed.

Regards -- Markus

Re: What luck we had ...

[Preben Randhol]

Pascal Obry <pascal@obry.net> wrote on 27/06/2006 (21:15) :
> 
> Or maybe your customer will have been happy to get the application under
> GPL license and have the right to ask for the sources. Do you think that
> giving away the sources to your customer will have changed something ? I
> bet he would not even consider asking for the sources :)

What stops them from publishing the source? Is GNAT dual licensed by the
way? I mean can a company buy a commercial license like the QT widget
library? (www.trolltech.no)

I really don't like the viral part of the GPL. Forced freedom is not freedom.

Preben

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Preben Randhol]

Ludovic Brenta <ludovic@ludovic-brenta.org> wrote on 27/06/2006 (17:30) :
> > 1) the GMGPL licences issued by AdaCode and others are being revoked?
> 
> Switched to pure GPL, not revoked. The "linking and generic
> instantiation" exception is revoked.

Can you update the packafe descriptions of the ada packages saying something
like: Please note: You will only be able to make pure GPL programs with
this compiler/library.

> They don't say that, but they refuse to give details on when the switch
> took place. I don't know. Ask them. From a theoretical standpoint, I am
> indeed quite worried that I have downloaded and redistributed AdaCore's
> software, thinking in good faith I had the right to do so when in fact
> I didn't, since I didn't have written permission from any of the
> copyright holders. Remember, the headers amount to naught from a legal
> perspective. Practically speaking, I don't think AdaCore will sue me,
> or anyone else, for that. IANAL.

If you are worried by this I think we should revoke all gnat packages
from Debian.

> > 5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?
> 
> Yes, since it is under GPL. That much they are willing to say and
> certify, but not in writing :(

How can we know that it is GPL? They "say" it is GPL now like it was
GMGPL before. Sorry, but I don't trust the ACT anymore. 

I think I'll stick with Python or other languages where one can relate
to licenses.

Preben

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

On Wed, 2006-06-28 at 15:50 +0200, Preben Randhol wrote:


> How can we know that it is GPL? They "say" it is GPL now like it was
> GMGPL before. Sorry, but I don't trust the ACT anymore. 
> 
> I think I'll stick with Python or other languages where one can relate
> to licenses.

Good Grief!
First thing to do if you receive software and worry about its
license is to find the license document. Then, if you need to
be fairly certain that you can do what you think you can do is
ask a lawyer. AdaCore has done so in the past, even informally
here on c.l.ada.

If you consider Python (no machine code, very little compile time
checking) as a replacement for Ada, I wonder whether this is really
about licensing. FUD at its best.

I hereby declare that Microsoft has several patents pending on the
execution of compiled python code within the .NET framework.
As the .NET framework will sooner of later start emulating x86
machine code on virtual hardware, jointly developed with VMware,
you can't trust the Python language any longer.

Re: AdaCore ... the Next SCO?

[Preben Randhol]

Georg Bauhaus <bauhaus@futureapps.de> wrote on 27/06/2006 (20:50) :
> On Tue, 2006-06-27 at 17:00 +0200, M E Leypold wrote:
> 
> >   # Alex Bykat who helped us compile GtkAda 0.2 on Solaris 2.5.
> >   # ...
> >   # Preben Randhol who has contributed several patches, in particular for porting gate to glade-2.
> 
> > Useful, this, isn't it?
> 
> Could make a change if members of the above list decided to
> make a GMGPL Edition from their earlier work.
> But just in case, this isn't useful as a licensing argument.
> A quote from the GNU site illustrates:
> 
> "If you modify this code, you may extend this exception to your version
> of the code, but you are not obligated to do so. If you do not wish to
> do so, delete this exception statement from your version."
> 
> So the release of a modified GtkAda may drop the
> exception without consulting the originators.
> I can't say whether or not the members of the above list
> like the new facts.

I don't like it. I contributed to a GMGPL project not a pure GPL
project.

Anyway, what I will do for the future is to use any license other than
GPL. Now, I guess I need to spend a bit time to find a license
compatible with GPL but without the viral part. I guess the BSD license is
one candidate. I guess I also have to read the Linux Format article
about GPL 3. Note that many programs are distributed with GPL saying the
program is licensed under (L)GPL 2.0 or later version of your choice. I
don't know what it entails, but if one keep it open like that one risk
running into more trouble when RMS makes a new version...

Just for the record I'm not against GPL per se, just that I don't like
posibility that one at some point end up in a quagmire where one cannot
make anything without a GPL license. That is not Freedom.

Also the article had an interview with the Apache developer and his
views on licensing since Apache is not under GPL.

[1] Link to the issue, unfortunately not available on the net:
http://www.linuxformat.co.uk/modules.php?op=modload&name=Sections&file=index&req=viewarticle&artid=5#80


One big question though. What is /where can on find the license of the
gcc - GNU Compiler Collection?

Does the gcc have an exception so one can make non-GPL programs with it?

Preben

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Wed, 2006-06-28 at 16:27 +0200, Preben Randhol wrote:

> One big question though. What is /where can on find the license of the
> gcc - GNU Compiler Collection?

A good source of information is the GCC web site.
If after reading you still have specific questions about the license,
first ask your laywer. If he or she cannot answer all your questions,
maybe your project can address a question to the FSF.


> Does the gcc have an exception so one can make non-GPL programs with it?

Which gcc?

Re: AdaCore ... the Next SCO?

[M E Leypold]

Preben Randhol <randhol+cla2@pvv.org> writes:

> Does the gcc have an exception so one can make non-GPL programs with it?

The C and C++ runtime are LGPL AFAIK, the gcc Gnat runtime is
GMGPL. So (IANAL) it should be possible to produce non-GPL executables
with FSF gcc in the languages C, C++ and Ada.

I remember that in the past, there were also some libraries included,
that where GPL only (libgcov? I might be wrong).

I don't know anything about the license status of the Java and ObjC
runtimes.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Frank J. Lhota]

Michael Bode wrote:
> Georg Bauhaus <bauhaus@futureapps.de> writes:
> 
>> I see the irony, but this is legal stuff. Consider a few sheets of
>> paper landing on your desk saying, this is a part of the NT kernel.
>> This part of the NT kernel source, as the header clearly indicates,
>> is GPLed. Would you believe it?
> 
> If I had downloaded it myself from
> https://gnu.microsoft.com/free_software/windows_nt/src/ ?


Folks, Georg was only kidding. We will have to wait a *long* time to see 
"gnu" and "Microsoft" in the same URL.

AFAIK Microsoft has not officially released the source code for any OS 
except the DOS 1.x BIOS. Two years ago, there was a panic over the 
leaking of the source code for Windows NT / 2000; see

	http://www.neowin.net/index.php?act=view&id=17509

This immediately raised the concern that hackers could use this source 
code to pinpoint vulnerabilities. So much for the concept of security 
through obscurity!

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ed Falis]

Frank J. Lhota wrote:

> AFAIK Microsoft has not officially released the source code for any OS
> except the DOS 1.x BIOS.

My brother works for MS.  I'm pretty sure he told me late last year that
MS does make the source code available for a price as part of their
shared source initiative.

- Ed

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

Florian Weimer <fw@deneb.enyo.de> writes:

>> How can any AdaCore software then remain in Debian at all?
>
> Debian does not check the precise copyright status of most packages.
> I doubt many authors are willing to provide assurances without proper
> compensation, especially if they have incorporated any contributions
> from third parties (which can pose legal risks even if you've got
> statements in writing to the contrary).

That I can understand. However I think most (L/GM/)GPL software
authors respect all parts of COPYING and thus respect the file headers
they had applied in accordance with COPYING.

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Frank J. Lhota:

> AFAIK Microsoft has not officially released the source code for any OS
> except the DOS 1.x BIOS.

Most Microsoft software is available in source code form.  And they
adhere to the GPL if necessary.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Ludovic Brenta:

> I did download quite a lot from AdaCore: GtkAda, AWS, XML/Ada [...]

Ah, okay.  The libraries are indeed problematic if you need the
linking exception.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

Frank J. Lhota wrote:
> Michael Bode wrote:
>> Georg Bauhaus <bauhaus@futureapps.de> writes:
>>> Consider a few sheets of
>>> paper landing on your desk saying, this is a part of the NT kernel.
>>> This part of the NT kernel source, as the header clearly indicates,
>>> is GPLed. Would you believe it?
>>
>> If I had downloaded it myself from
>> https://gnu.microsoft.com/free_software/windows_nt/src/ ?
> 
> Folks, Georg was only kidding.

Well, not really. 

> We will have to wait a *long* time to see 
> "gnu" and "Microsoft" in the same URL.

Though,
http://www.microsoft.com/spain/sharedsource/Articles/GNU.mspx

From the XEN laboratory at U of Cambridge:
"A port of Windows XP was developed for an earlier version of Xen,
but is not available for release due to licence restrictions."

Seems to me a word with those whose software you are using is
always a way. I'd prefer this to arguing about license headers,
and I guess that most compiler shops will prefer a simple, quiet,
reasonable deal, too.
 Wouldn't they know what software production is about?

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Georg Bauhaus]

Georg Bauhaus wrote:

> First thing to do if you receive software and worry about its
> license is to find the license document. Then, if you need to
> be fairly certain that you can do what you think you can do is
> ask a lawyer. AdaCore has done so in the past, even informally
> here on c.l.ada.
This last sentence is a dangling thought. Please,
for Sentence'Storage_Pool use null.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

"Frank J. Lhota" <flhota.NOSPAM@ll.mit.edu> writes:

>> If I had downloaded it myself from
>> https://gnu.microsoft.com/free_software/windows_nt/src/ ?
>
>
> Folks, Georg was only kidding. We will have to wait a *long* time to
> see "gnu" and "Microsoft" in the same URL.

Yeah, that would be surprising. The point was this: if I take someone
elses closed source software and stick a COPYING to it, then of course
this cannot result in the software suddenly becoming GPL. 

But if the author of said software sprinkles it all over the place
with licensing declarations (about 200 *.ads files in GtkAda contain
the GMGPL header) then I would believe that this means what it says.

-- 
Michael Bode

Re: What luck we had ...

[Michael Bode]

Preben Randhol <randhol+cla2@pvv.org> writes:

> I really don't like the viral part of the GPL. Forced freedom is not freedom.

I have no problem with the GPL and its viral aspects. I have a problem
when someone quietly changes the rules and sticks misleading labels
(or file headers) to his software.

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Simon Wright]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> They never had any legal force; only a signed statement from the
> copyright holder has legal force.

I had an off-the-cuff opinion from a lawyer that to publish
contradictory statements about copyright and licensing is at the least
going to make any legal case more difficult.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Simon Wright]

"Dr. Adrian Wrigley" <amtw@linuxchip.demon.co.uk.uk.uk> writes:

> As regards the GPL, it appears to be a perpetual, sub-licensable,
> non-revocable (absense of breaches) licence.  The GMGPL terms add to
> this, but don't change these basic features.

The GPL includes statements about transferability, the modifications
don't (but there is nothing to say they _don't_ transfer, so one might
expect the GPL provisions to apply).

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Frank J. Lhota]

Georg Bauhaus wrote:
>> Folks, Georg was only kidding.
> 
> Well, not really.

The URL https://gnu.microsoft.com/free_software/windows_nt/src does not 
currently appear to be up. I just assumed it was a joke. If it wasn't, I 
humbly apologize. Out of curiosity, if this was a real site, what source 
files did it offer?

>> We will have to wait a *long* time to see "gnu" and "Microsoft" in the 
>> same URL.
> 
> Though,
> http://www.microsoft.com/spain/sharedsource/Articles/GNU.mspx
> 
>  From the XEN laboratory at U of Cambridge:
> "A port of Windows XP was developed for an earlier version of Xen,
> but is not available for release due to licence restrictions."
> 
> Seems to me a word with those whose software you are using is
> always a way. I'd prefer this to arguing about license headers,
> and I guess that most compiler shops will prefer a simple, quiet,
> reasonable deal, too.
> Wouldn't they know what software production is about?

One would hope!

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Michael Bode]

"Frank J. Lhota" <flhota.NOSPAM@ll.mit.edu> writes:

> The URL https://gnu.microsoft.com/free_software/windows_nt/src does
> not currently appear to be up. I just assumed it was a joke.

Yes, that was a joke.

-- 
Michael Bode

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

M E Leypold writes:
> Ed Falis <falis@verizon.net> writes:
>
>> M E Leypold wrote:
>> 
>> >> AUnit was written by Ed Falis, an AdaCore employee.  Only source is
>> >> AdaCore, so switch to pure GPL.
>> >
>> > Also a tool, wouldn't get linked into an executable for the customer:
>> > In my eyes GPL is OK here.
>> 
>> Note that AUnit has been GPL from its first release in 2000.  Since its
>> intent is user "in-house" testing rather than any kind of
>> re-distribution of test cases, I considered it the appropriate license.
>
> Ed, I agree. 
>
> Tools for inhouse use are appropriately GPL IMHO. We don't want anyone
> to run away with tools chain after all :-). The issues here -- where
> all that started -- are different (not with new licenses but with the
> possible retroactive change of old licenses, license changes on
> unchanged source, missing or misleading license notices and a general
> amnesia).
>
> I'm sure you'll see that after wading through all that fallout. 
>
> Regards -- Markus

Ed is the author of AUnit, he was just correcting an oversight I made
in the post that started this thread.  He is well aware of all the
issues involved.

BTW, thanks for the correction, Ed.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Florian Weimer <fw@deneb.enyo.de> writes:

> * Michael Bode:
>
>> How can any AdaCore software then remain in Debian at all?
>
> Debian does not check the precise copyright status of most packages.

Oh yes, per Policy.

> I doubt many authors are willing to provide assurances without proper
> compensation, especially if they have incorporated any contributions
> from third parties (which can pose legal risks even if you've got
> statements in writing to the contrary).

So far, Debian has provided a some form of assurances by means of the
copyright file shipped with every package, and cryptographically
signed by the package maintainer (i.e. myself in the case of Ada
packages).

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Preben Randhol <randhol+cla2@pvv.org> writes:

> Ludovic Brenta <ludovic@ludovic-brenta.org> wrote on 27/06/2006 (17:30) :
>> > 1) the GMGPL licences issued by AdaCode and others are being revoked?
>> 
>> Switched to pure GPL, not revoked. The "linking and generic
>> instantiation" exception is revoked.
>
> Can you update the packafe descriptions of the ada packages saying something
> like: Please note: You will only be able to make pure GPL programs with
> this compiler/library.

I've started doing that with ASIS, which I am preparing for upload.  I
will continue to do that for the other packages as I update them.  But
the GMGPL remains in force for all packages that were previously GMGPL
in Sarge.

>> They don't say that, but they refuse to give details on when the switch
>> took place. I don't know. Ask them. From a theoretical standpoint, I am
>> indeed quite worried that I have downloaded and redistributed AdaCore's
>> software, thinking in good faith I had the right to do so when in fact
>> I didn't, since I didn't have written permission from any of the
>> copyright holders. Remember, the headers amount to naught from a legal
>> perspective. Practically speaking, I don't think AdaCore will sue me,
>> or anyone else, for that. IANAL.
>
> If you are worried by this I think we should revoke all gnat packages
> from Debian.

I am not worried, and since the same reasoning applies to all software
in Debian, then Debian and all software should disappear.  I don't
think that's realistic or true.

>> > 5) the SW is Free (in http://www.gnu.org/philosophy/free-sw.html terms)?
>> 
>> Yes, since it is under GPL. That much they are willing to say and
>> certify, but not in writing :(
>
> How can we know that it is GPL? They "say" it is GPL now like it was
> GMGPL before. Sorry, but I don't trust the ACT anymore. 

Ask the authors, if you trust the AUTHORS file :)

> I think I'll stick with Python or other languages where one can relate
> to licenses.

That will not solve the problem.

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> Georg Bauhaus wrote:
>
>> First thing to do if you receive software and worry about its
>> license is to find the license document. Then, if you need to
>> be fairly certain that you can do what you think you can do is
>> ask a lawyer. AdaCore has done so in the past, even informally
>> here on c.l.ada.
> This last sentence is a dangling thought. Please,
> for Sentence'Storage_Pool use null.

I'd ask the author before asking a lawyer.  Keep the lawyers at bay
whenever you can :-)

-- 
Ludovic Brenta.

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

M E Leypold writes:
> Can I:
>
>   - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
>     under the GMGPL whereas S3 is GPL.

Yes.

>   - I then compile S1, S2, S3 to X. 

Yes.

>   - Obviously X is covered by GPL: I must distribute S1, S2, S3 with X
>     as the GPL demands.

Yes, and you must distribute the sources of X, too, since X must be
under terms compatible with the GPL.  You can even choose to
distribute X under GMGPL.

>   - But cant' I state that S1, S2 are under GMGPL -- that is, anyone
>     receiving them is allowed to unbundle them from the source package
>     of X (which is actually made up from 3 different trees) and can
>     distribute them (S1, S2) as GMGPL sources or create other GMGPL,
>     GPL or even closed executables from them.

Yes, that's true.  To make things clear to the licensee, you can
distribute the libraries, and X, in separate source packages.

> Question: Is there any contradiction in the license terms or is that
> permissible? 

I don't think there is a contradiction.

> Of course one would want to do that for whatever reason ever. I do not
> want to discuss the fairness or unfairness of that or what one could
> expect the authors of S1, S2 to do or not to do. I just want to know,
> wether the authors of S1 and S2 (if they want) license S1 and S2 as
> GMGPL and be sure that they can be used in aforesaid manner.

Yes, they can.

> Or do they have to license as GPL to ensure "linkability" with a GPL
> library?

No, they don't.  The GMGPL already ensures "linkability", it is
GPL-compatible.

-- 
Ludovic Brenta.

Re: AdaCore ... the Next SCO?

[Simon Wright]

M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:

> Add to that, that their Pro-Version
> probably still has the linking exception.

Indeed it does. We would be quite unable to use GNAT without the
exception (well, I think so, and I wouldn't bust a gut to persuade the
business that GPL'd weapon systems are OK because the customer owns
the code anyway and won't want to distribute it).

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

Preben Randhol writes:
> Georg Bauhaus wrote on 27/06/2006 (20:50) :
>> So the release of a modified GtkAda may drop the
>> exception without consulting the originators.
>> I can't say whether or not the members of the above list
>> like the new facts.
>
> I don't like it. I contributed to a GMGPL project not a pure GPL
> project.

If you are one of the authors, and you happen to have a copy of the
software, then I believe it is your right to distribute it under a new
license or under the GMGPL.

> One big question though. What is /where can on find the license of
> the gcc - GNU Compiler Collection?

http://gcc.gnu.org/mirrors.html

The tarballs are crypto-signed and contain a COPYING file and headers.
If you're still not sure about the license terms, ask the people who
crypto-signed the archive.

> Does the gcc have an exception so one can make non-GPL programs with it?

Yes, both in libstdc++ and libgnat.

-- 
Ludovic Brenta.

Re: AdaCore ... the Next SCO?

[Simon Wright]

"Martin Krischik" <krischik@users.sourceforge.net> writes:

> In the absence of pragma License GNAT scans the commens for magic words
> which indicate the lincence used ;-) . As it is the BC will be
> considered GMGPL by GNAT. But since no GNAT specific packages are used
> no licence warnings are issued.

I added the notice the BCs use to a unit which withs GNAT.IO (and which
gives the warning with pragma License (Modified_GPL)). No warnings
(the 2005 edition).

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

Simon Wright <simon@pushface.org> writes:
> M E Leypold writes:
>
>> Add to that, that their Pro-Version
>> probably still has the linking exception.
>
> Indeed it does. We would be quite unable to use GNAT without the
> exception (well, I think so, and I wouldn't bust a gut to persuade the
> business that GPL'd weapon systems are OK because the customer owns
> the code anyway and won't want to distribute it).

Delivery of GPL'd software by a weapons system will probably terminate
the recipient anyway :)

-- 
Ludovic Brenta.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Ludovic Brenta:

> Florian Weimer <fw@deneb.enyo.de> writes:
>
>> * Michael Bode:
>>
>>> How can any AdaCore software then remain in Debian at all?
>>
>> Debian does not check the precise copyright status of most packages.
>
> Oh yes, per Policy.

It checks if the *license* meets certain criteria.  But anybody can
take someone else's code, pretend it's their own, slap a new license
on it, and put it on the Net for download.  The result is typically a
copyright infringement.  The purported license statement doesn't tell
you this, of course.

>> I doubt many authors are willing to provide assurances without proper
>> compensation, especially if they have incorporated any contributions
>> from third parties (which can pose legal risks even if you've got
>> statements in writing to the contrary).
>
> So far, Debian has provided a some form of assurances by means of the
> copyright file shipped with every package,

The copyright file is just what upstream provided.  If it's wrong,
Debian will make the same false claims.

In some cases, packages contain contradicting license claims, which
are discovered after some time (see #328923 for an example).  But if
no such claims exist, it's unlikely that we'd spot a copyright
violation.

> and cryptographically signed by the package maintainer (i.e. myself
> in the case of Ada packages).

In a couple of weeks, indeed. 8-)

But in reality, Debian provides few developer-to-user guarantees.  The
.changes files are awfully hard to get, and I'm not sure if those for
binary-only NMUs are archived at all.  But this is completely
off-topic here.

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Florian Weimer]

* Frank J. Lhota:

> Folks, Georg was only kidding. We will have to wait a *long* time to
> see "gnu" and "Microsoft" in the same URL.

*yawn*

<ftp://ftp.microsoft.com/developr/Interix/sfu35/sources/Interix/gnu/>

Re: AdaCore ... the Next SCO?

[Björn Persson]

M E Leypold wrote:
> Can I:
> 
>   - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
>     under the GMGPL whereas S3 is GPL.
> 
>   - I then compile S1, S2, S3 to X. 
> 
>   - Obviously X is covered by GPL: I must distribute S1, S2, S3 with X
>     as the GPL demands.
> 
>   - But cant' I state that S1, S2 are under GMGPL -- that is, anyone
>     receiving them is allowed to unbundle them from the source package
>     of X (which is actually made up from 3 different trees) and can
>     distribute them (S1, S2) as GMGPL sources or create other GMGPL,
>     GPL or even closed executables from them.
> 
> Question: Is there any contradiction in the license terms or is that
> permissible? 

That is permissible.

The result can be confusing, for example if S1 is a program, S2 is some 
library that the program needs, and S3 is Libgnat. I think most people 
are used to assuming that the machine code of a program will have the 
same license as the source code. In this case it won't, and that's going 
to confuse people, but no, there's nothing in the license terms that 
forbids this.

> Note that I thing, that the builder of the excutable X cannot strip
> the linking exceaption from the libs S1 and S2 since he/she has not
> changed the libs. He would have to refrain from using them if he is
> not allowed to link with S3.

I disagree here. I think you can take GMGPL code, strip the exception 
and redistribute it as pure GPL but otherwise unmodified. You shouldn't 
do it, because it would serve no purpose other than FUD, but I think 
it's allowed. I'm not entirely sure about that though.

-- 
Bj�rn Persson                              PGP key A88682FD
                    omb jor ers @sv ge.
                    r o.b n.p son eri nu

Re: AdaCore ... the Next SCO?

[M E Leypold]

Ludovic Brenta <ludovic@ludovic-brenta.org> writes:

> M E Leypold writes:
> > Can I:
> >
> >   - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
> >     under the GMGPL whereas S3 is GPL.
> 
> Yes.
> 
> >   - I then compile S1, S2, S3 to X. 
> 
> Yes.
> 
> >   - Obviously X is covered by GPL: I must distribute S1, S2, S3 with X
> >     as the GPL demands.
> 
> Yes, and you must distribute the sources of X, too, since X must be
> under terms compatible with the GPL.  

Yes, of course. My idea was that there is no extra source. It is S2
which has the main procedure and glues S1 and S3 together :-).

> You can even choose to distribute X under GMGPL.

> >   - But cant' I state that S1, S2 are under GMGPL -- that is, anyone
> >     receiving them is allowed to unbundle them from the source package
> >     of X (which is actually made up from 3 different trees) and can
> >     distribute them (S1, S2) as GMGPL sources or create other GMGPL,
> >     GPL or even closed executables from them.
> 
> Yes, that's true.  To make things clear to the licensee, you can
> distribute the libraries, and X, in separate source packages.

OK. That answers at least one of the questions with which I started 2
weeks (??) ago in the DTRAQ thread.

> > Question: Is there any contradiction in the license terms or is that
> > permissible? 
> 
> I don't think there is a contradiction.

> > Of course one would want to do that for whatever reason ever. I do not
> > want to discuss the fairness or unfairness of that or what one could
> > expect the authors of S1, S2 to do or not to do. I just want to know,
> > wether the authors of S1 and S2 (if they want) license S1 and S2 as
> > GMGPL and be sure that they can be used in aforesaid manner.
> 
> Yes, they can.
> 
> > Or do they have to license as GPL to ensure "linkability" with a GPL
> > library?
> 
> No, they don't.  The GMGPL already ensures "linkability", it is
> GPL-compatible.

Thanks. At least we have an agreement here in opinion, even if we are
not lawyers. Some of the arguments I heard in the discussion seemed to
imply different or skirted the issue altogether.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Bj�rn Persson <spam-away@nowhere.nil> writes:

> M E Leypold wrote:
> > Can I:
> >   - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
> >     under the GMGPL whereas S3 is GPL.
> >   - I then compile S1, S2, S3 to X.   - Obviously X is covered by
> > GPL: I must distribute S1, S2, S3 with X
> >     as the GPL demands.
> >   - But cant' I state that S1, S2 are under GMGPL -- that is, anyone
> >     receiving them is allowed to unbundle them from the source package
> >     of X (which is actually made up from 3 different trees) and can
> >     distribute them (S1, S2) as GMGPL sources or create other GMGPL,
> >     GPL or even closed executables from them.
> > Question: Is there any contradiction in the license terms or is that
> > permissible?
> 
> That is permissible.
> 
> The result can be confusing, for example if S1 is a program, S2 is
> some library that the program needs, and S3 is Libgnat. I think most
> people are used to assuming that the machine code of a program will
> have the same license as the source code. In this case it won't, and
> that's going to confuse people, but no, there's nothing in the license
> terms that forbids this.
> 
> > Note that I thing, that the builder of the excutable X cannot strip
> > the linking exceaption from the libs S1 and S2 since he/she has not
> > changed the libs. He would have to refrain from using them if he is
> > not allowed to link with S3.
> 
> I disagree here. I think you can take GMGPL code, strip the exception
> and redistribute it as pure GPL but otherwise unmodified. You
> shouldn't do it, because it would serve no purpose other than FUD, but
> I think it's allowed. I'm not entirely sure about that though.

Hm. I now begin to doubt wether you're ever allowed to strip the
exception from a derived work. I assume the LE (linking exception
forms on unit or document with the GPL. Where the GPL says "this
license" it should in the GMGPL case mean GPL + LE => GMGPL. 

If that is right, one could force an executable under GPL by inserting
files that are pure GPL into the source. But it should not be possible
to strip the linking exceaption from those parts that have them.

Of course everything can be relicensed under another license: If _all_
copyright holders agree. I do not want to wake sleeping dogs here, but
I suggest that anyone intrested in the question tries to check
copyright history of and contributors to GtkAda and Florist.

At leaset some people that wrote in c.l.a. seem not to have signed
over their right to ACT when contributing.


Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Simon Wright <simon@pushface.org> writes:

> "Dr. Adrian Wrigley" <amtw@linuxchip.demon.co.uk.uk.uk> writes:
> 
> > As regards the GPL, it appears to be a perpetual, sub-licensable,
> > non-revocable (absense of breaches) licence.  The GMGPL terms add to
> > this, but don't change these basic features.
> 
> The GPL includes statements about transferability, the modifications
> don't (but there is nothing to say they _don't_ transfer, so one might
> expect the GPL provisions to apply).

Exactly. 

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[M E Leypold]

Florian Weimer <fw@deneb.enyo.de> writes:

> > So far, Debian has provided a some form of assurances by means of the
> > copyright file shipped with every package,
> 
> The copyright file is just what upstream provided.  If it's wrong,
> Debian will make the same false claims.

But if upstream is the copyright holder one would expect it to be
legally binding noneless. You cannot (bad analogy) sign a contract
with you bank and after some time decide, that the sugnature has been
"wrong". The same applies to adding copyright statements to your
product: It is a declaration of intent and as such should have legally
binding force. Just your tough look if you weren't awake enough when
you did that. 

So as a first approximation I'd say: Declaration of intent -> legally
binding. Proofability is another matter altogether, but there is often
a number of witnesse which could state that the downloaded the very
same file with exctly these copyright notices.

Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ole-Hjalmar Kristensen]

And this is really the core of the problem at the moment. Can you
trust ACT? 

>>>>> "PR" == Preben Randhol <randhol+cla2@pvv.org> writes:

<snip>

    PR> How can we know that it is GPL? They "say" it is GPL now like it was
    PR> GMGPL before. Sorry, but I don't trust the ACT anymore. 

    PR> I think I'll stick with Python or other languages where one can relate
    PR> to licenses.

    PR> Preben

-- 
   *NOT* speaking for my employer

Re: AdaCore ... the Next SCO?

[Preben Randhol]

Ludovic Brenta <ludovic@ludovic-brenta.org> wrote on 28/06/2006 (22:20) :
> > Does the gcc have an exception so one can make non-GPL programs with it?
> 
> Yes, both in libstdc++ and libgnat.

Only?

Re: AdaCore ... the Next SCO?

[Martin Krischik]

M E Leypold wrote:

> 
> "Martin Krischik" <krischik@users.sourceforge.net> writes:
> 
>> Simon Wright wrote:
>> > Since the BCs don't use the non-standard pragma License (Licence?) the
>> > question doesn't arise. And given this discussion they are not going
>> > to.
>> 
>> In the absence of pragma License GNAT scans the commens for magic words
>> which indicate the lincence used ;-) . As it is the BC will be
>> considered GMGPL by GNAT. But since no GNAT specific packages are used
>> no licence warnings are issued.
> 
> If if there would be: I do not grok it: Linking GMGPL with GPL
> libraries should put the excutable under GPL, but the GMGPL part of
> the source would still be under GMGPL, wouldn't it? Why the license
> warnings then with GNAT specific units?

Because Ada support "rename" - the GMGPL package could pass thru GPL
licensed stuff. The Linux-Kernel Guys are really pissed of by many
companies circumventing the GPL by modules containing basicly:

void x ()
  {
  y();
  }

Martin
-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com

Re: AdaCore ... the Next SCO?

[Stephen Leake]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> Jeffrey Creem wrote :
>> Hopefully someone at Greenhills is paying attention to this discussion.
>> This confusing license perhaps "exposing" a company to GPL terms when
>> the headers clearly are not GPL will make a great writeup that will
>> pretty much make it impossible to even use GNATPro within my company.
>
> I don't understand. Did your company not receive a license statement
> from AdaCore along with GNAT Pro? 

I didn't. But I'm not selling or distributing anything, so I don't
care specifically.

On the other hand, I probably could get one, since I have a
contractual relationship with them.

> No, I don't think AdaCore want to sue anyone. I would rather think they
> are die-hard, purist Free Software believers, like RMS (i.e. "free up
> your software or pay"). 

That wasn't what they said when they first released GNAT with GMGPL
runtime libraries.

-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Stephen Leake]

Martin Krischik <krischik@users.sourceforge.net> writes:

> Ludovic Brenta wrote:
>
>>  The change of
>> license has no effect on corporations large enough to use GNAT Pro, or
>> on students, hobbyists, or free software developers.
>
> Does it? The GPL is viral and booch, charles, AdaCL - all currently GMGPL
> would need to to relicensed to GPL now to be used with with GNAT/GPL.

This is not true. 

You can use GMGPL sources with GPL sources; the license for the
resulting product must be GPL.

> AdaCL at least produced licence-warnings when compiled with GNAT/GPL. While
> licence warning are not legal the at least shows how AdaCore sees things.
> And this would mean that those libs become unavailable to closed source
> users.

It's not the AdaCL libraries that are unavailable to closed source
users; it's the GNAT runtime libraries. 

As others have pointed out, that just means there will now be a better
market for different Ada runtimes.

-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Stephen Leake]

M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:

>> But that's a consequence of using GNAT GPL edition, not of using
>> the libraries? These can keep their licenses. (The GPL doesn't say
>> it is time limited, and revocable; I'd be surprised to learn that
>> the exception is time limited.)
>
>
> No. Just reread Martins post about the license warnings popping up

Those warnings are just saying "you've got two different licenses
here". They do _not_ mean "you can't do this legally". 

In that sense, they are misleading warnings. They should be
suppressed, in general.

-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Stephen Leake]

"Martin Krischik" <krischik@users.sourceforge.net> writes:

> Georg Bauhaus wrote:
>
>> The GMGPL is certainly GPL compatible.
>
> No they are not. Look at this simple code sniplet:
>
> with GNAT.OS_Lib;
>
> package My_OS_Lib
>    package Pass_Thrue  renames GNAT.OS_Lib;
> end My_OS_Lib;
>
> I used a rename to make shure the problem is exposed.

What, exactly, are you trying to say here?

The GMGPL license is GPL compatible; see
http://www.gnu.org/philosophy/license-list.html#GPLCompatibleLicenses

That means you can use GMGPL source code in a project that also has
GPL source code; the license for the total project must be GPL.

>> Even if you run into a pragma License(some invalid identifier),
>> you just create a derivative work by placing '--' before the
>> pragma. Just make sure that the result, if distributed, is
>> distributed as required by the GPL.
>
> Right: My work has to become GPL as well.

If you distribute binaries compiled with the GNAT GPL compiler, that
is true.

-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Björn Persson]

Preben Randhol wrote:
> I don't like it. I contributed to a GMGPL project not a pure GPL
> project.
> 
> Anyway, what I will do for the future is to use any license other than
> GPL. Now, I guess I need to spend a bit time to find a license
> compatible with GPL but without the viral part. I guess the BSD license is
> one candidate.

Well, if you want to stop people from taking your code and slapping the 
GPL on it, then a BSD-style license is a bad choice.

If I understand you right, you want a license that allows compiling 
together with GPL code to form a GPL executable, but forbids changing 
the license of the source code to GPL. I don't know of such a license, 
and I think it could be difficult to write one. You'd have to 
differentiate between two different kinds of derivative work.

-- 
Bj�rn Persson                              PGP key A88682FD
                    omb jor ers @sv ge.
                    r o.b n.p son eri nu

Re: AdaCore ... the Next SCO?

[Jeffrey R. Carter]

Stephen Leake wrote:
> "Martin Krischik" <krischik@users.sourceforge.net> writes:
> 
>> Right: My work has to become GPL as well.
> 
> If you distribute binaries compiled with the GNAT GPL compiler, that
> is true.

No, it's not true. "My work" can mean my sources, or my binary, or both. 
Only the binary must be GPL. The sources can remain GMGPL.

-- 
Jeff Carter
"I unclog my nose towards you."
Monty Python & the Holy Grail
11

Re: AdaCore ... the Next SCO?

[Jeffrey Creem]

Stephen Leake wrote:

>>
>>Right: My work has to become GPL as well.
> 
> 
> If you distribute binaries compiled with the GNAT GPL compiler, that
> is true.
> 

To be clear (and I am not implying that Stephen does not know this) the 
work as distributed as part of that binary needs to be GPL as well but 
still nothing prevents that source code from being made available under 
other terms as well.

I.e. the following is perfectly fine

Distributed as GPL		Distributed as GMGPL
Library A			Library A
Compiled with GNAT GPL		Compiled with Aonix(for example)
into application XYZZY		into application EB90

When you distribute application XYZZY, you must do so in accordance with 
the terms of the GPL (meaning releasing Library A under the GPL).

But, nothing prevents you from also using Library A in application EB90 
and distributing application EB90 under a proprietary license.

This is why I think it is a little silly (even given my other concerns 
about the CVS "misleading?" headers) for anyone distributing GMGPL to 
change the license headers on their own files simply because they want 
it to "work" with the GNAT GPL edition.

Re: AdaCore ... the Next SCO?

[Simon Wright]

Martin Krischik <krischik@users.sourceforge.net> writes:

> M E Leypold wrote:
>
>> "Martin Krischik" <krischik@users.sourceforge.net> writes:
>> 
>>> Simon Wright wrote:
>>> > Since the BCs don't use the non-standard pragma License (Licence?) the
>>> > question doesn't arise. And given this discussion they are not going
>>> > to.
>>> 
>>> In the absence of pragma License GNAT scans the commens for magic words
>>> which indicate the lincence used ;-) . As it is the BC will be
>>> considered GMGPL by GNAT. But since no GNAT specific packages are used
>>> no licence warnings are issued.
>> 
>> If if there would be: I do not grok it: Linking GMGPL with GPL
>> libraries should put the excutable under GPL, but the GMGPL part of
>> the source would still be under GMGPL, wouldn't it? Why the license
>> warnings then with GNAT specific units?
>
> Because Ada support "rename" - the GMGPL package could pass thru GPL
> licensed stuff. The Linux-Kernel Guys are really pissed of by many
> companies circumventing the GPL by modules containing basicly:
>
> void x ()
>   {
>   y();
>   }

I don't quite understand this point. However, I suppose it protects me
as a GMGPL author from mistakenly depending on something which (in the
public GNAT GPL compiler) is GPL. For instance, my embedded web server
depends on GNAT.Regex, damnit.

Re: AdaCore ... the Next SCO?

[Simon Wright]

M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:

> Well -- if you, as JC seemed to imply -- are working for a company
> that is providing programming service to large aerospace companies
> and of those some in turn seem to have a "no GPL" policy in place
> now, as other posts seem to imply: You wouldn't want to explain to
> them the difference between Gnat GPL and Gnat Pro. You'd like to
> avoid the "GPL stink", someone put it here

I really don't see this as a problem; I work for a large aerospace
company, we are so far as I know quite happy with the service we get
from our vendors and the licence terms are OK too. I hope we aren't so
blinkered as to go off on a rant against anything with G, P and L in
its licence.

OK, every so often we contemplate programming our systems in VB or
Java, wiser heads prevail, but C++ is a real challenge; and not
because it's GPL-free, but because it's easier to get people with it
on their CVs.

Re: AdaCore ... the Next SCO?

[Martin Dowie]

Simon Wright wrote:
> M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:
> 
>> Well -- if you, as JC seemed to imply -- are working for a company
>> that is providing programming service to large aerospace companies
>> and of those some in turn seem to have a "no GPL" policy in place
>> now, as other posts seem to imply: You wouldn't want to explain to
>> them the difference between Gnat GPL and Gnat Pro. You'd like to
>> avoid the "GPL stink", someone put it here
> 
> I really don't see this as a problem; I work for a large aerospace
> company, we are so far as I know quite happy with the service we get
> from our vendors and the licence terms are OK too. I hope we aren't so
> blinkered as to go off on a rant against anything with G, P and L in
> its licence.
> 
> OK, every so often we contemplate programming our systems in VB or
> Java, wiser heads prevail, but C++ is a real challenge; and not
> because it's GPL-free, but because it's easier to get people with it
> on their CVs.

Is it easy to get really /good/ C++ people though...

Cheers
-- Martin

Re: AdaCore ... the Next SCO?

[M E Leypold]

Simon Wright <simon@pushface.org> writes:

> M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:
> 
> > Well -- if you, as JC seemed to imply -- are working for a company
> > that is providing programming service to large aerospace companies
> > and of those some in turn seem to have a "no GPL" policy in place
> > now, as other posts seem to imply: You wouldn't want to explain to
> > them the difference between Gnat GPL and Gnat Pro. You'd like to
> > avoid the "GPL stink", someone put it here
> 
> I really don't see this as a problem; I work for a large aerospace
> company, we are so far as I know quite happy with the service we get
> from our vendors and the licence terms are OK too. I hope we aren't so
> blinkered as to go off on a rant against anything with G, P and L in
> its licence.

I've only been trying to summarize the arguments of an earlier post
(not mine) and also relate them to another post here which used the
words "GPL stink". Personally I can imagine that denial or
forgetfulness of past licensing and distribution term might cast
shadows of doubt over all that source and also support GPL FUD. But
regarding the "large aerospace companies" I've actually no
experience. 

As I said: I only wanted to summarize someone elses arguments here.

Actually I'd be happy to hear, that the industry at large had not
problem with the GPL. My experience in the small unfortunately has
been partially different.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Jeffrey R. Carter" <spam.not.jrcarter@acm.not.spam.org> writes:

> Stephen Leake wrote:
> > "Martin Krischik" <krischik@users.sourceforge.net> writes:
> >
> >> Right: My work has to become GPL as well.
> > If you distribute binaries compiled with the GNAT GPL compiler, that
> > is true.
> 
> No, it's not true. "My work" can mean my sources, or my binary, or
> both. Only the binary must be GPL. The sources can remain GMGPL.

Thanks. That was what I actually expected. 

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Jeffrey Creem <jeff@thecreems.com> writes:

> This is why I think it is a little silly (even given my other concerns
> about the CVS "misleading?" headers) for anyone distributing GMGPL to
> change the license headers on their own files simply because they want
> it to "work" with the GNAT GPL edition.

Exactly. And I had the impression that some people here are living
exactly under that impression (that they have now to relicense) and
asking about that in the DTRAQ thread was how the licensing discussion
started there. 

So apart from the questions on the actual status of the sources
distributed by AdaCore or the stance of AdaCore on all that (which
everyone should ask AdaCore about, not rely on rumors here :-)), this
separate question has been clarified:

  GMGPL source doesn't need to be relicensed to GPL to be used with
  Gnat GPL or with other GPL libraries.

IANAL, but that seems to become the agreement here. Any objections?

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Bj�rn Persson <spam-away@nowhere.nil> writes:

> Preben Randhol wrote:
> > I don't like it. I contributed to a GMGPL project not a pure GPL
> > project.
> > Anyway, what I will do for the future is to use any license other
> > than
> > GPL. Now, I guess I need to spend a bit time to find a license
> > compatible with GPL but without the viral part. I guess the BSD license is
> > one candidate.
> 
> Well, if you want to stop people from taking your code and slapping
> the GPL on it, then a BSD-style license is a bad choice.
> 
> If I understand you right, you want a license that allows compiling
> together with GPL code to form a GPL executable, but forbids changing
> the license of the source code to GPL. 

I presently living under the impression that GMGPL is actually such a
license. "This license" in the GPL text refers to the license as a
whole, which includes the GMGPL linking exception when read in the
context of GMGPL (note: It's called Gnat _modified_ Gnu public license
-- the Gnat modified has license as designation, the result is
another, slightly different license :-)).

I doubt, that anyone can just strip the linking exception from a GMGPL
licensed work, but of course some GPL parts/files could be introduced
to make the executables GPL by force. This is IMHO what Gnat GPL
does. That is OK (from a licensing point of view nas I, IANAL, see
it). But removing the licensing exception from a derived file is not
OK. 

IANAL. Someone will have to take that up with AdaCore to get
clarification on that.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Stephen Leake <Stephe.Leake@nasa.gov> writes:

> M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:
> 
> >> But that's a consequence of using GNAT GPL edition, not of using
> >> the libraries? These can keep their licenses. (The GPL doesn't say
> >> it is time limited, and revocable; I'd be surprised to learn that
> >> the exception is time limited.)
> >
> >
> > No. Just reread Martins post about the license warnings popping up
> 
> Those warnings are just saying "you've got two different licenses
> here". They do _not_ mean "you can't do this legally". 

Ah, I see. 

> In that sense, they are misleading warnings. They should be
> suppressed, in general.

Or should better care for the licensing semantics. Would be nice to
have the compiler state in the linking stage "This excutable can be
distributed under licenses X, Y or Z".

Regards -- Markus

Re: AdaCore ... the Next SCO?

[Stephen Leake]

"Jeffrey R. Carter" <spam.not.jrcarter@acm.not.spam.org> writes:

> Stephen Leake wrote:
>> "Martin Krischik" <krischik@users.sourceforge.net> writes:
>>
>>> Right: My work has to become GPL as well.
>> If you distribute binaries compiled with the GNAT GPL compiler, that
>> is true.
>
> No, it's not true. "My work" can mean my sources, or my binary, or
> both. Only the binary must be GPL. The sources can remain GMGPL.

Yes, that is correct.

-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Stephen Leake]

Jeffrey Creem <jeff@thecreems.com> writes:

> Stephen Leake wrote:
>
>>>
>>>Right: My work has to become GPL as well.
>> If you distribute binaries compiled with the GNAT GPL compiler, that
>> is true.
>>
>
> To be clear (and I am not implying that Stephen does not know this)
> the work as distributed as part of that binary needs to be GPL as well
> but still nothing prevents that source code from being made available
> under other terms as well.

Yes, I was not sufficiently clear.


-- 
-- Stephe

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Simon Wright wrote:

>> Because Ada support "rename" - the GMGPL package could pass thru GPL
>> licensed stuff. The Linux-Kernel Guys are really pissed of by many
>> companies circumventing the GPL by modules containing basicly:
>>
>> void x ()
>> {
>> y();
>> }
> 
> I don't quite understand this point.

You combine three different licences to link closed source to GPL:

GPL <-> totally_free_licence <-> closed_source_licence.

Martin.
-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Preben Randhol]

On Wed, 28 Jun 2006 20:55:52 +0200
Florian Weimer <fw@deneb.enyo.de> wrote:

> * Frank J. Lhota:
> 
> > AFAIK Microsoft has not officially released the source code for any
> > OS except the DOS 1.x BIOS.
> 
> Most Microsoft software is available in source code form.  And they
> adhere to the GPL if necessary.

Where?

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

Martin Krischik wrote:

> You combine three different licences to link closed source to GPL:
> 
> GPL <-> totally_free_licence <-> closed_source_licence.

If is_derived_work(outcome(GPL <-> "totally_free_licence" <-> closed_source_licence))
then
   sources_must_be_available;
end;

How could an indirect call, closed( don_t_care( gpl(...))), work around
the fact that closed(), when called, will execute the code
that implements gpl()?

GPL: "We offer our services to anyone, provided they agree to L"
DONTCARE: "We offer our services to anyone, in any way for any use."

Now DONTCARE hires a GPLed service. The service works as part of one of
DONTCARE's own service functions. Someone notices this and starts asking
questions. What answers will she get?

CLOSED: "We only have a contract with DONTCARE. None of our business
  what they did"
DONTCARE: ________

Is CLOSED right?
Fill in the blanks. :-)

-- Georg 

Re: AdaCore ... the Next SCO?

[Simon Wright]

Martin Dowie <martin.dowie@btopenworld.com> writes:

> Simon Wright wrote:

>> OK, every so often we contemplate programming our systems in VB or
>> Java, wiser heads prevail, but C++ is a real challenge; and not
>> because it's GPL-free, but because it's easier to get people with
>> it on their CVs.
>
> Is it easy to get really /good/ C++ people though...

Hence the phrasing ...

Re: AdaCore ... the Next SCO?

[Simon Wright]

M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:

> Actually I'd be happy to hear, that the industry at large had not
> problem with the GPL.

I can't speak for others; indeed I can't speak _for_ my employer
either.

It's a Project responsibility to ensure that we have proper licences
for all the items delivered to the customer, it would be mine to make
sure -- for example -- that we don't try to deliver Xml/Ada, for which
we don't have a support contract, if it was GPL (the version we use
was obtained well before the Great Copyright Shift, and anyway it's
only in use for non-deliverables).

Re: AdaCore ... the Next SCO?

[Martin Krischik]

Georg Bauhaus wrote:

> Martin Krischik wrote:
> 
>> You combine three different licences to link closed source to GPL:
>> 
>> GPL <-> totally_free_licence <-> closed_source_licence.
> 
> If is_derived_work(outcome(GPL <-> "totally_free_licence" <->
> closed_source_licence)) then
>    sources_must_be_available;
> end;
> 
> How could an indirect call, closed( don_t_care( gpl(...))), work around
> the fact that closed(), when called, will execute the code
> that implements gpl()?
> 
> GPL: "We offer our services to anyone, provided they agree to L"
> DONTCARE: "We offer our services to anyone, in any way for any use."
> 
> Now DONTCARE hires a GPLed service. The service works as part of one of
> DONTCARE's own service functions. Someone notices this and starts asking
> questions. What answers will she get?
> 
> CLOSED: "We only have a contract with DONTCARE. None of our business
>   what they did"
> DONTCARE: ________
> 
> Is CLOSED right?
> Fill in the blanks. :-)

I never said it is right - it's just what is done - line with the nvidia
graphic drivers.

Martin
-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Matthew Goulet]

Ludovic Brenta wrote:
> I received detailed answers from AdaCore's Robert Dewar and Arnaud
> Charlet regarding the licenses of software downloaded from their
> servers.  In summary:
>
> ...
>
> - AdaBindX [3], a binding to X11 and LessTif by Hans-Frieder Vogt
>   which is under GMGPL.  X11 is under X11 (MIT) license, and
>   LessTif[4] is under LGPL.  But AdaBindX has not been updated since
>   2000, and is not portable.  And not as good-looking as GTK+.
>   According to Debian's Popularity Contest [5], this package has zero
>   users, so I am tempted to drop it from Etch.  Speak up if you want
>   me to keep it.
>
> ...
>
> Thoughts, comments, offers to help?
>
> --
> Ludovic Brenta.

Hello there,
  Long time listener, first time caller, so to speak.  Just wanted to
say I've found this whole thread quite interesting and intructive, even
though the subject itself still contains so much doubt.

  Also I wanted to say to Ludovic specifically, I'm actually someone
who's using the AdabindX libraries.  Perhaps I'm the only one :-)  As
for popularity contest, I use noatime, so it doesn't work for me.
Regardless, one person isn't enough reason to keep a package, but I
nevertheless wanted to say thanks for all the work on GNAT for Debian.

  I hope to be able to post meaningfully here in the future, its nice
to read postings with intellect, especially about my favorite language
:-)

-Matt

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

Preben Randhol wrote:
> Just for the record I'm not against GPL per se, just that I don't like
> posibility that one at some point end up in a quagmire where one cannot
> make anything without a GPL license. That is not Freedom.

It is, for the users of the software. You are forgetting, as
programmers so often do, that the freedom granted by the GPL is not for
the benefit of programmers. If, in the process of granting users the
freedom to run, read, modify, and share software it happens that
programmers are inconvenienced, the FSF does not care. If users wind up
with less software as a result, that is unfortunate, but not so
unfortunate as to require overriding the principles of users' software
freedom.

Remember that without the GPL, it's likely that some users down the
road will be denied the ability to exercise the freedoms that ought to
be theirs. AdaCore has done these users a favor by helping to limit how
much non-free software can be produced.

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

M E Leypold wrote:
> Wonderful. Now tell me what rights ACT had to strip the linking
> exception from most files of florist. Please do that after diffing GPL
> florist against 3.15p.

Florist itself grants that right. For example, I looked at
<http://www.cs.uofs.edu/~beidler/Ada/posix/ada_task_identification.html>.
It says at the beginning,
FLORIST is free software;  you can  redistribute  it and/or  modify it
under terms of the  GNU  General  Public  License as  published by the
Free Software Foundation;  either version  2, or (at  your option) any
later version.

That's what AdaCore is doing - they are choosing to distribute it
solely under the GPL. While the above header does grant the special
linking exemption as well, it does not require that this exemption must
continue to be given upon redistribution. In fact, a close reading of
that heading would imply that you are not even permitted to
redistribute with that exemption! That is, when the file is distributed
to you from the original party, it grants you the exemption, but it
says that you can only redistribute it under the GPL as published by
the FSF.

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> Preben Randhol wrote:
> > Just for the record I'm not against GPL per se, just that I don't like
> > posibility that one at some point end up in a quagmire where one cannot
> > make anything without a GPL license. That is not Freedom.
> 
> It is, for the users of the software. You are forgetting, as
> programmers so often do, that the freedom granted by the GPL is not for
> the benefit of programmers. If, in the process of granting users the
> freedom to run, read, modify, and share software it happens that
> programmers are inconvenienced, the FSF does not care. If users wind up
> with less software as a result, that is unfortunate, but not so
> unfortunate as to require overriding the principles of users' software
> freedom.
> 
> Remember that without the GPL, it's likely that some users down the
> road will be denied the ability to exercise the freedoms that ought to
> be theirs. AdaCore has done these users a favor by helping to limit how
> much non-free software can be produced.

Simply: No. 

Sorry, my friend. I've been pushing the GPL and more generally open
sourcing most of my professional live. But this kind of orwellian
newspeak simply annoys me and in my not very humble opinion doesn't do
good to GPL and Open Source.

There is a time to GPL, there is a time to do BSD licenses and LGPL
and GMGPL also have their place. Even the FSF had the wisdom not to
insist on GPL for everything. Your attempt to recast a _restriction_
as ultimate freedom undreminds me of Friedrich D�rrenmatt's novel "Vom
Beobachten des Beobachters" where the narrator characterizes rape as
the ultimate freedom since it frees the raped one from the necessity
of deciding.

"Freedom" is actually a complex thing, since, as other people already
have observed here, there are interactions and constraints between
different freedoms and not all can be realized at the same time. That
brings me to the suggestion, that for real freedom an ecology of
various coexisting licenses should be able to exist. Pushing the GPL
and GPL only as the singular way to salvation just is quasi religious
totalitarism: "If everybody would be to MY god, all evil would
perish". I personally doubt that.

As far as AdaCore is concerned, their pure-GPL stance has nothing to
with "freedom". It's a business decision as anybody can see, since,
perhaps, support without the additional "incentive" of a dual
licensing scheme (or say: The threat of GPL) doesn't sell any
more. You can draw your own conclusion from that.

Consider: Measuring AdaCores decision with your ethical scale would
make the current license arrangement totally fishy: For a substantial
sum of money one can buy onself off the downstream users freedom. If
the GPL is good, because it _forces_ the GPL developer to "respect"
the freedom of the downstream users, then not respecting it would be
"sin" but you can _buy_ yourself the freedom to commit this sin. 

How does that sound? The catholic church did this in the middle ages,
it was called "selling of indulgences" and led to the reformation.

Arguing purely on ethical issues is bound to fail with open source
licenses. One has to realize, that GPL and others are actually legal
hacks of a copyright system which had been invented to restrict
freedom. The hacks try to use the rules of this system in a way to
achieve another, novel effect, that is ensuring the one or the other
kind of freedom of the sources and/or the users.

Since the licenses are hacks, every single one has only a restricted
area of application and can also be misused (as an economical weapon).

If one is talking about license one better had to talk about cause and
effect -- that is, which license has which effects and wether one
would like to have that effects happen. Ethical considerations might
guide your reasoning, but frankly I cannot see a 1:1 mapping of any IP
ethics to a single license exactly because of their hackish
nature. 

Finally one has to consider wether (and how) your license works for
you or for the community: GPL protagonists are adviced to have a look
to the BSD communities. These also work and some bad things are not
happening generally, despite the licensees (users of source) are not
forced to do good all the time and ever. That makes me think, and it
should make you think also.

Back to the discussion here: It is rather simple: AdaCore has stripped
the linking exception from a number of source tarballs or pretend (so
far) that they are not valid any more. This is a license change from
GMGPL -> GPL. Other people have contributed under GMGPL and feel
cheated now. Generally the GPL cannot be changed (there is a guarantee
in the licensing clauses) and it open to discussion wether not the
same (including linking exception) applies to the GMGPL.

Just because the change was _to GPL_ (in your opinion the better
license) that doesn't dispense AdaCore from sticking to the rules. The
GMGPL is also protected by the rules. And that is the issue here.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> it does not require that this exemption must continue to be given
> upon redistribution.

Are you sure? I don't think so. The exemption is part of the
conditions under which the thingy is licensed to you. So the words
"this license" in the GPL apply to the exemption also.


> In fact, a close reading of
> that heading would imply that you are not even permitted to
> redistribute with that exemption! 

Could you try to give some arguments in favor of this rather strange view?


Regards -- Markus

Re: Ada in Debian: most libraries will switch to the pure GPL in Etch

[Ludovic Brenta]

Thanks, Matt.

-- 
Ludovic Brenta.

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Tue, 2006-07-04 at 01:58 +0200, M E Leypold wrote:


> > In fact, a close reading of
> > that heading would imply that you are not even permitted to
> > redistribute with that exemption! 
> 
> Could you try to give some arguments in favor of this rather strange view?

I found that formally convincing: close reading shows,

   "This program[!] is free software; you can redistribute it[!] and/or modify
    it[!] under the terms->of->the GNU General Public License[!] 
...
   "As a special exception[!]...

Considering the word in my native language:

   "Ausnahmsweise[!]...

I note that "exception" in not in general a "general permission", is it?

Yes, making GNAT strictly GPL might be a decision that is relevant to
Ada business.
GPLing generally useful libraries (though, with GNAT only) will
invite either thinking about selling more GPL software if possible,
or it might invite writing a Ada libraries portable among compilers
and operating systems.

--  Georg 

Re: AdaCore ... the Next SCO?

[Michael Bode]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> Remember that without the GPL, it's likely that some users down the
> road will be denied the ability to exercise the freedoms that ought to
> be theirs. AdaCore has done these users a favor by helping to limit how
> much non-free software can be produced.

Make that "to limit how much free and non-free software can be
produced in Ada". 

The part about free software comes from people who write free and
non-free software or write free software but choose a GPL-incompatible
Free Software licence, but don't want to use entire different tools
for both and thus choose tools that can do both.

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

M E Leypold wrote:
> "Hyman Rosen" <hyman.rosen@gmail.com> writes:
> 
>> it does not require that this exemption must continue to be given
>> upon redistribution.
> 
> Are you sure? I don't think so. The exemption is part of the
> conditions under which the thingy is licensed to you. So the words
> "this license" in the GPL apply to the exemption also.
> 
>> In fact, a close reading of
>> that heading would imply that you are not even permitted to
>> redistribute with that exemption! 
> 
> Could you try to give some arguments in favor of this rather strange view?

Well, again, here's the part of the header of the file
which deals with distribution:

......................................................................
FLORIST is free software;  you can  redistribute  it and/or  modify it
under terms of the  GNU  General  Public  License as  published by the
Free Software Foundation;  either version  2, or (at  your option) any
later version.

As a special exception, if other files instantiate generics from  this
unit, or you link this unit with other files to produce an  executable,
this  unit does not by itself cause the  resulting  executable  to  be
covered  by the  GNU  General  Public License. This exception does not
however invalidate any other  reasons why the executable file might be
covered by the GNU Public License.
......................................................................

As you can see, the only way you are allowed to distribute this file
is under the GPL, and the GPL does not contain the special exemption.
So you are not allowed to distribute this file to other people with
the exemption in place. The people who distributed it to you must then
have been the original copyright holders, who were able to choose any
additional terms they wished.

As you can also see, there is no mention of "this license". The only
license that's mentioned is the (pure) GPL.

So it's pretty clear: the authors of Florist distributed the files to
you, allowing you to generate executables from them not bound by the
GPL solely because of that, but only permitting you to redistribute
them under the GPL without that exemption. Had they wished otherwise,
they would have inserted license terms that said that you must grant
the same exemption to anyone to whom you redistribute.

Or maybe they just didn't ask a lawyer.

Re: AdaCore ... the Next SCO?

[Simon Wright]

Hyman Rosen <hyrosen@mail.com> writes:

> As you can see, the only way you are allowed to distribute this file
> is under the GPL, and the GPL does not contain the special
> exemption.  So you are not allowed to distribute this file to other
> people with the exemption in place. The people who distributed it to
> you must then have been the original copyright holders, who were
> able to choose any additional terms they wished.

It's Florist as a whole that is GPL'd. Part of Florist is 'this file'
which (as distributed) contained the exception. So the whole position
seems somewhat murky.

If Florist contains (as some of my code does) a file that is public
domain would it be forbidden to distribute it without _adding_ the
GPL?  I think not ..

None of us is a lawyer, of course, but I'm beginning to think that the
standard COPYING file is not enough, the licensing terms for a
mixed-license project ought to be spelt out in much more detail.

Re: AdaCore ... the Next SCO?

[M E Leypold]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> On Tue, 2006-07-04 at 01:58 +0200, M E Leypold wrote:
> 
> 
> > > In fact, a close reading of
> > > that heading would imply that you are not even permitted to
> > > redistribute with that exemption! 
> > 
> > Could you try to give some arguments in favor of this rather strange view?
> 
> I found that formally convincing: close reading shows,
> 
>    "This program[!] is free software; you can redistribute it[!] and/or modify
>     it[!] under the terms->of->the GNU General Public License[!] 
> ...
>    "As a special exception[!]...
> 
> Considering the word in my native language:
> 
>    "Ausnahmsweise[!]...
> 
> I note that "exception" in not in general a "general permission", is it?

Inhowfar would that forbid distribution with the exemption clause intact?

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Michael Bode <m.g.bode@web.de> writes:

> "Hyman Rosen" <hyman.rosen@gmail.com> writes:
> 
> > Remember that without the GPL, it's likely that some users down the
> > road will be denied the ability to exercise the freedoms that ought to
> > be theirs. AdaCore has done these users a favor by helping to limit how
> > much non-free software can be produced.
> 
> Make that "to limit how much free and non-free software can be
> produced in Ada". 
> 
> The part about free software comes from people who write free and
> non-free software or write free software but choose a GPL-incompatible
> Free Software licence, but don't want to use entire different tools
> for both and thus choose tools that can do both.

Exactly. Personally I'm not interested to use different tool sets for
my hobby than for my professional work (if possible :-)): My skills /
know how had to be acquired painfully and I see sense trying to
minimize the pain quantitatively. On the other side, if I really want
to use a totally different tool to write free software, there would be
a plethora of rather interesting languages and tools which aren't as
standardized as Ada is (like Haskell, Free Pascal - just to name two),
but who would care: After all it would be a hobby only :-).

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Hyman Rosen <hyrosen@mail.com> writes:

> M E Leypold wrote:
> > "Hyman Rosen" <hyman.rosen@gmail.com> writes:
> >
> >> it does not require that this exemption must continue to be given
> >> upon redistribution.
> > Are you sure? I don't think so. The exemption is part of the
> > conditions under which the thingy is licensed to you. So the words
> > "this license" in the GPL apply to the exemption also.
> >
> >> In fact, a close reading of
> >> that heading would imply that you are not even permitted to
> >> redistribute with that exemption!
> > Could you try to give some arguments in favor of this rather strange
> > view?
> 
> Well, again, here's the part of the header of the file
> which deals with distribution:
> 
> ......................................................................
> FLORIST is free software;  you can  redistribute  it and/or  modify it
> under terms of the  GNU  General  Public  License as  published by the
> Free Software Foundation;  either version  2, or (at  your option) any
> later version.
> 
> As a special exception, if other files instantiate generics from  this
> unit, or you link this unit with other files to produce an  executable,
> this  unit does not by itself cause the  resulting  executable  to  be
> covered  by the  GNU  General  Public License. This exception does not
> however invalidate any other  reasons why the executable file might be
> covered by the GNU Public License.
> ......................................................................
> 
> As you can see, the only way you are allowed to distribute this file
> is under the GPL, and the GPL does not contain the special exemption.
> So you are not allowed to distribute this file to other people with
> the exemption in place. The people who distributed it to you must then
> have been the original copyright holders, who were able to choose any
> additional terms they wished.
> 


> As you can also see, there is no mention of "this license". The only
> license that's mentioned is the (pure) GPL.

"This license" is the language in the GPL. So you don't think the
exemption is part of the license? 


 
> So it's pretty clear: the authors of Florist distributed the files to
> you, allowing you to generate executables from them not bound by the
> GPL solely because of that, but only permitting you to redistribute
> them under the GPL without that exemption. Had they wished otherwise,
> they would have inserted license terms that said that you must grant
> the same exemption to anyone to whom you redistribute.

I say: Nonsense. 

Regards -- Markus

Re: AdaCore ... the Next SCO?

[M E Leypold]

Simon Wright <simon@pushface.org> writes:

> Hyman Rosen <hyrosen@mail.com> writes:
> 
> > As you can see, the only way you are allowed to distribute this file
> > is under the GPL, and the GPL does not contain the special
> > exemption.  So you are not allowed to distribute this file to other
> > people with the exemption in place. The people who distributed it to
> > you must then have been the original copyright holders, who were
> > able to choose any additional terms they wished.
> 

> It's Florist as a whole that is GPL'd. Part of Florist is 'this file'

Florist 3.15p seems to be GPL+Exception = GMGPL. The README actually
quotes the exemption clause. But the clause itself only makes sense
for Ada compilation units, so in a sense, the other files are GPL. I
don't think there is any doubt how the exemption clause has to be
applied (even if there would be no extra header in single source
files).

FYI: In the current ACT florist the exemption clause has been stripped
also from the source files. 

> which (as distributed) contained the exception. So the whole position
> seems somewhat murky.

I would think that the intent is quite clear. Admittedly in retrospect
the GMGPL could have been designed better (by modifying the GPL text
directly), but I wonder why the problem crops up now, after more than
10 years ...


> If Florist contains (as some of my code does) a file that is public
> domain would it be forbidden to distribute it without _adding_ the
> GPL?  I think not ..

Nor do I.

> None of us is a lawyer, of course, but I'm beginning to think that the
> standard COPYING file is not enough, the licensing terms for a
> mixed-license project ought to be spelt out in much more detail.

The FOX library added a special text about how the LGPL should be
interpreted in the context of FOX. I find the exemplary. "More
details" are usually found in the README (here in many projects we
find texts saying that some files are under a different license, see
file headers) and the file headers. 

Usually all this is not a problem, if a thinking person has written
the README.

The current murkiness surrounding the ACT maintained libraries stem
mostly from bad documentation practice: No summary of licensing terms
in the README, sometimes stripped, sometimes non stripped file headers
and the stout maintainance that everything is pure GPL _regardless_
what the file headers say.

The last is either total cheekiness, somebody messed up big or just
bad will.

But generally, apart from ACTs fault, READMEs, COPYING or LICENSE
files and file headers are plenty for establishing a clear licensing
situation, if written well.

Perhaps we should now write to the FSF and ask for advice what the
GMGPL mean. I bet they don't say that it means, we cannot redistribute
GMGPLed files :-).

Regards -- Markus

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

M E Leypold wrote:
> I say: Nonsense.

For GPLv3, the FSF is planning to roll the LGPL into the GPL,
using the new GPL's ability to add extra permission clauses.

Here's a quote from Stallman (copied from Groklaw):
    One of the other benefits we get from this is that we
    make it clear that any time someone adds extra
    permissions on top of the GNU GPL, that when
    you modify the program you can take off those added
    permissions. You can release your version under the
    strict GPL and nothing more.

So you can see that it's very much the intention of the FSF that
people are allowed to remove extra permission clauses if they
want to. On the other hand, having these extra clauses as part
of the main license should clear up the question of whether one
can redistribute with the extra permissions. I do agree with you
that the originial authors of Florist and the like intended for the
software to be redistributable with the extended permissions. I
just think they didn't word their license properly. The new LGPL
should help with that.

Re: AdaCore ... the Next SCO?

[michael bode]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> So you can see that it's very much the intention of the FSF that
> people are allowed to remove extra permission clauses if they
> want to. 

I'd say there is a big difference between being *allowed* to distribute
without the extra permission and being *required* to distribute
without the extra permissin.

And at least Debian and Gentoo maintainers thought they were allowed
to distribute *with* the extra permission.

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> M E Leypold wrote:
> > I say: Nonsense.
> 
> For GPLv3, the FSF is planning to roll the LGPL into the GPL,
> using the new GPL's ability to add extra permission clauses.
> 
> Here's a quote from Stallman (copied from Groklaw):

<rest snipped>

Hi Hyman,

Yes, I've read the Interview at Groklaw. I've the following things to
remark:


 (1) Your hypothesis (somewhere earlier in this thread) was, that the
     the GMGPL linking exception is contradicting the GPL, i.e. one
     can't redistribute with the exception clause intact.  The
     interview has no bearing on that.


 (2) I think, that this direction on part of the FSF is a bad
     move. What will come from that is, that we'll see more and more
     software where the LGPL has been converted to GPL (not to give
     the end user more freedom, but to poison tools for commercial use
     and introduce a dual licensing scheme). Imagine Linux
     distributions with LibC under GPL, and you can only get the LGPL
     version if you buy support for the enterprise edition.


 (3) I think that allowing anyone to redistribute without linking
     exception will emphasize the viral nature of GPL and won't do
     FLOSS much good. After all, if I don't want somebody to strip the
     linking exception from my code, I'll have to avoid the GPL in
     future, since I cannot add a non-removable exception to GPL 3.

     Furthermore it would be annoying, if, a business adds improved to
     a library under LGPL. Then a competitor adds more improvements
     and strips the linking exception. The first company can now only
     continue to use the non improved version. Somehow that seems to
     me a break of the implied "sharing contract" -- I'd expect other
     people to add under the same conditions (so that I can also
     profit from their improvements, not introducing different rules
     that exclude application areas for earlier contributors).


 (4) As far as GtkAda (2.4.0) and Florist are concerned: Their license
     is still GMGPL based on GPL 2. So it is still open, wether the
     linking exception can be stripped from source you don't own. (As
     a reference: You can't convert LGPL 2 to GPL, that much is
     certain in my eyes).


Regards -- Markus

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

M E Leypold wrote:
>  (1) Your hypothesis (somewhere earlier in this thread) was, that the
>      the GMGPL linking exception is contradicting the GPL, i.e. one
>      can't redistribute with the exception clause intact.  The
>      interview has no bearing on that.

True. It's not even a hypothesis, really. I just wanted to
point out that the plain text of the license in the file could
be read in the way I described. It's certainly an affirmative
defense against people who say that you must not strip
the exception from the license.

>  (2) I think, that this direction on part of the FSF is a bad move.

Notice that this is not at all a change in direction, merely
a clarification of policies that have always been in place.
Notice that these exemptions can always eventually lead
to a user who is denied the ability to read, modify, and
redistribute a program.

> won't do FLOSS much good

Again, remember that the goal of the FSF is that when a
user receives a piece of software, he may execute, read,
modify, and redistribute it. The FSF does not care for the
"good of FLOSS" except as regards these four freedoms.
Anything which facilitates the creation of software which
does not come with these freedoms is bad, not good. The
LGPL exists because sometimes this bad thing is a
lesser evil, but the FSF isn't particulary happy about it.
For the most part, the FSF prefers no software to non-free
software. The LGPL is for cases where they fear non-free
alternatives may become popular if there is no open version
that can be used selfishly.

> After all, if I don't want somebody to strip the
>      linking exception from my code, I'll have to avoid the GPL in
>      future, since I cannot add a non-removable exception to GPL 3.

Or just dual-license, the way many other packages do.

>      Furthermore it would be annoying, if, a business adds improved to
>      a library under LGPL. Then a competitor adds more improvements
>      and strips the linking exception. The first company can now only
>      continue to use the non improved version.

Annoying to whom? Is there a user who loses any of the
four freedoms when this happens? If not, the FSF doesn't
care. Is the first company hampered by being unable to
use improvements without distributing under the GPL?
Good. More freedom for users.

> Somehow that seems to
>      me a break of the implied "sharing contract" -- I'd expect other
>      people to add under the same conditions (so that I can also
>      profit from their improvements, not introducing different rules
>      that exclude application areas for earlier contributors).

What you are describing is a cartel whose members
collude for their mutual benefit and to the detriment of
their users, who may be denied the four freedoms.
You really must learn to separate your needs and
desires as a programmer from the four freedoms that
the FSF wants users to have. The purpose of the extra
conditions is to avoid having to distribute a program
under the GPL. The main goal of avoiding the GPL is
generally to deny users the freedoms it would grant.

>  (As a reference: You can't convert LGPL 2 to GPL, that much is
>      certain in my eyes).

Have you read the LGPL? <http://www.gnu.org/copyleft/lesser.html>
    3. You may opt to apply the terms of the ordinary GNU
    General Public License instead of this License to a given
    copy of the Library. To do this, you must alter all the
    notices that refer to this License, so that they refer to the
    ordinary GNU General Public License, version 2, instead
    of to this License. (If a newer version than version 2 of the
    ordinary GNU General Public License has appeared, then
    you can specify that version instead if you wish.) Do not
    make any other change in these notices.

    Once this change is made in a given copy, it is irreversible
    for that copy, so the ordinary GNU General Public License
    applies to all subsequent copies and derivative works made
    from that copy.

Re: AdaCore ... the Next SCO?

[Michael Bode]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> Have you read the LGPL? <http://www.gnu.org/copyleft/lesser.html>
>     3. You may opt to apply the terms of the ordinary GNU
>     General Public License instead of this License to a given
>     copy of the Library. To do this, you must alter all the
>     notices that refer to this License, so that they refer to the
>     ordinary GNU General Public License, version 2, instead
>     of to this License. (If a newer version than version 2 of the
>     ordinary GNU General Public License has appeared, then
>     you can specify that version instead if you wish.) Do not
>     make any other change in these notices.
>
>     Once this change is made in a given copy, it is irreversible
>     for that copy, so the ordinary GNU General Public License
>     applies to all subsequent copies and derivative works made
>     from that copy.

This, as I have learned here, has no legal force. You always have to ask
the author.

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

Michael Bode wrote:
> This, as I have learned here, has no legal force.
> You always have to ask the author.

You have heard this claimed here. I would suggest
that you do not "learn" it. Should things ever come to
enforcement actions, a judge is going to look very
dubiously at anyone claiming that a plain statement
of license embedded in a file does not apply. Go have
a look over at Groklaw and their discussion of "squishy
law".
<http://www.groklaw.net/article.php?story=20060716182835630>.

The law is not a computer program which is rigidly
executed by lawyers and judges. Any plaintiff will be
asked why, if he did not want files to be distributed in
accordance with the license in the header, he allowed
that license to remain there. He will be asked this in
the context of the FSF's instructions on how to apply
the GPL, <http://www.gnu.org/licenses/gpl.html>, which
say to put the license and copyright information into the
source file headers. The judge at best will think that the
plaintiff was too lazy to properly establish his rights and
at worst that he was deliberately trying to entrap the
recipients of the software. And the plaintiff will have his
case summarily dismissed.

Re: AdaCore ... the Next SCO?

[M E Leypold]

"Hyman Rosen" <hyman.rosen@gmail.com> writes:

> Michael Bode wrote:
> > This, as I have learned here, has no legal force.
> > You always have to ask the author.
> 
> You have heard this claimed here. I would suggest
> that you do not "learn" it. Should things ever come to
> enforcement actions, a judge is going to look very
> dubiously at anyone claiming that a plain statement
> of license embedded in a file does not apply.

Hi Hyman,

I hope I'll find time for a more in depth answer at another time. For
now only the following:

  - License notices (file headers, COPYING file) between the following
    versions of GtkAda have not changed: 2.2.1, 2.4.0 as from
    libre.act-europe at the beginning of 2005, 2.4.0 as from obtained
    today from libre2, 2.8.0 as obtained from AdaCore CVS today.

  - Nonetheless AdaCore claims that the version(s) now available from
    their site are GPL only (not linking exception).

    When asked for the other 2 Version (2.2.1 and 2.4.0/2005) they
    suddenly fall silent and become unresponsive.

  - 2.2.1 to carries a GMGPL notice on freshmeat up to this day and is
    actually served from libre2 (so they cannot say they don't
    know about this ... )

  - 2.4.0/2005 from libre1 had an explicit GMGPL notice on the
    website.

I think that actually supports the hypothesis that AdaCore obviously
thinks that the license notices in the files are meaningless. Explicit
requests to clarify this point are met with glaring silence.

This is the point, AFAIS, to which Michael refers. And let's be
honest: I presently see no movement from the community to pressure
AdaCore to get their act clean. Since everybody loves GPL (me included
(well, sort of), despite the impression I seem to have left
sometimes), it just seems to be save to assume GPL and forget about
the slightly different notices in the files. That of course is, how
licenses start to loose their meaning, by bad practice. And that is
(even worse) how reliance to notices in the distributions will be
undermined.

> Should things ever come to enforcement actions, a judge is going to
> look very dubiously at anyone claiming that a plain statement of
> license embedded in a file does not apply.

I doubt that shoddy practice with regards to license notices in the
files ever comes to an "enforcement action". After all hardly anybody
will complain, since the actual license is "more GPL" not "less" and
GPL is the morally good license.

Personally I do not think it is good to be fuzzy in these things. If
the trust in current distribution practice is undermined, it'll not
create freedom if I've to find an actual rights holder and get him
speaking to me (which is even more difficult) every time I want to use
an open source package.

Regards -- Markus

Re: AdaCore ... the Next SCO?

[michael bode]

M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de> writes:

> Personally I do not think it is good to be fuzzy in these things. If
> the trust in current distribution practice is undermined, it'll not
> create freedom if I've to find an actual rights holder and get him
> speaking to me (which is even more difficult) every time I want to use
> an open source package.

And even if you find the actual rights holder (and you/he/someone can
prove he really is the actual rights holder) he might be believing in
the FSF interpretation and tell you to look into the headers, because
that's were he wrote the license. 

I imagine Linus Torvalds answering millions of requests if kernel
x.y.z *really* is GPL or maybe something else because license notices
in the sources "have no legal force". He will send you to hell.

So *if* one would believe the "no legal force" thing, the logical
consequence is to use commercial shrink-wrap EULA software and avoid
OSS at any cost.

One might wonder if some compiler vendor came to the exact same
conclusion.

Re: AdaCore ... the Next SCO?

[Jacob Sparre Andersen]

Hyman Rosen wrote:

> It is, for the users of the software. You are forgetting, as
> programmers so often do, that the freedom granted by the GPL is not
> for the benefit of programmers.

Exactly!

But it is not surprising that we in a _programming_ newsgroup are a
bit annoyed, when our freedom is limited in favour of the freedom of
the users.

> If, in the process of granting users the freedom to run, read,
> modify, and share software it happens that programmers are
> inconvenienced, the FSF does not care. If users wind up with less
> software as a result, that is unfortunate, but not so unfortunate as
> to require overriding the principles of users' software freedom.

I understand the principle and idea in this, but I strongly suspect
that ACT are doing both themselves and Ada a disservice by abandoning
GMGPL.

It is not all costumers who can be convinced that they have to accept
_two_ new unknowns at the same time (GNU GPL and Ada), so we may end
up in a situation where Ada gets completely locked to its current
commercial niche.

I have looked a bit at alternatives to GtkAda and GNAT, but I can't
find anything which supports the platforms I need at a price within
the budgets of my typical commercial projects.

Greetings,

Jacob
-- 
�I like it when the support group complains that they have
 insufficient data on mean time to repair bugs in Ada
 software.�                              -- Robert I. Eachus

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Tue, 2006-07-18 at 09:40 +0200, michael bode wrote:

> I imagine Linus Torvalds answering millions of requests if kernel
> x.y.z *really* is GPL or maybe something else because license notices
> in the sources "have no legal force". He will send you to hell.

What is "legal force"?

(I Can't speak for Linus.)

> So *if* one would believe the "no legal force" thing, the logical
> consequence is to use commercial shrink-wrap EULA software and avoid
> OSS at any cost.

Why should EULicenseA be different from GPLicense in this
regard?


-- Georg 

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

Michael Bode a écrit :
> And even if you find the actual rights holder (and you/he/someone can
> prove he really is the actual rights holder) he might be believing in
> the FSF interpretation and tell you to look into the headers, because
> that's were he wrote the license.

If he says "look in the headers", that is a valid response provided he
knows exactly what the headers contain *in your copy of the software*.
If he doesn't know (i.e. if your copy is not crypto-signed with his
key), then that's equivalent to giving you blanket approval to do
whatever you want.  If he says "GPL", then it's GPL.

The reasons why the headers have no legal force, and for the blanket
approval above, according to Robert Dewar at FOSDEM 2006, are:
- because anyone besides the copyright holder may have changed them in
the software.
- because license terms can change with the weather, or be different
for each licensee. They can even change after a licensee has had a
license for some time.

This is true of all software, free or proprietary, cost or no cost. So
basically, the user needs assurance about two things: who the copyright
holder really is, and what license terms are applicable at any given
point in time and for this particular licensee.

To give proper assurance on both accounts, I think that the copyright
holder should crypto-sign his software, and grant a perpetual,
non-revokable, immutable license to everyone. Unfortunately, the GPL
(GNAT-modified or not) is revokable and mutable (think "or any later
version"), so it is possible that it not apply to all licensees.

Debian crypto-signs all software it carries, but that doesn't make it
the copyright holder, and does not guarantee that the license terms
shipped with the software are really applicable. It only guarantees who
did the packaging, and suggests that the Debian maintainer has done his
best. The final authority is the copyright holder.

In the case of AdaCore, there is past evidence that suggests, but does
not prove for lack of crypto signatures, that AdaCore is indeed the
copyright holder of most of the software they offer on their CVS or web
servers (but not GNAT itself, which belongs to the FSF). I think that
AdaCore could successfully convince a court that they are indeed the
copyright holder. If that's the case, then they are the ultimate
authority on the license terms.

The problem is that AdaCore have destroyed confidence in the license
terms they apply to their software.

My interpretation of the license change is this:

- if I download anything from AdaCore, I receive it under the terms of
the pure GPL, no matter what the headers say. I have a statement from
them in my records to back this claim (unfortunately, it is not
crypto-signed, so one could argue it has no legal force either :))

- the software I downloaded in the past, and which is now in Debian
Sarge, was under the terms of the GMGPL, and by these terms I am
allowed to redistribute this software under GMGPL terms. I think this
is correct because AdaCore has not explicitly revoked my GMGPL license
to this old software (I also have emails on record to back this claim).
The software in Sarge is crypto-signed with my GPG key, do if you get
the software from Debian, you have assurance that *I* am redistributing
the software to *you* under the GMGPL terms.

But let me reiterate: I don't think AdaCore will sue anyone. In fact,
this whole ongoing conversation is evidence that several people are
going out of their way to be really, really honest and respectful of
AdaCore's copyright, and trying to understand the ramifications. I know
many people who wouldn't be so considerate.

> I imagine Linus Torvalds answering millions of requests if kernel
> x.y.z *really* is GPL or maybe something else because license notices
> in the sources "have no legal force". He will send you to hell.

Actually, the situation of the Linux kernel is quite uncertain, as
there are many copyright holders, not all of whom can be reached, and
it would be impossible to get assurances for every bit of the kernel.
But, as for the portions that Linus owns, I think it would be easy to
convince a court that he is the copyright holder.

More importantly, there is no question, currently, about which license
terms are applicable to the Linux kernel. Everyone is convinced that
the GPL version 2 (and not a later version) is applicable, and I think
a court could be convinced easily, since Linus has public statements on
record to that effect.

> So *if* one would believe the "no legal force" thing, the logical
> consequence is to use commercial shrink-wrap EULA software and avoid
> OSS at any cost.

No; the problem remains the same whether or not the software is free or
open-source. You can buy pirated copies of Microsoft Windows,
shrink-wrapped, in China and other places, or you can get pirated
software bundled with a new computer from some small vendors. And the
license terms, theoretically, could change because of your name or
nationality or the weather in Iceland. In fact, the licence terms of
Microsoft Windows have changed in the past ("Software Assurance" they
called it, how ironic), and could change again in the future. So,
theoretically, you are still and always responsible for getting a
proper license statement from the copyright holder.

-- 
Ludovic Brenta.

Revocable or non-revocable licenses from AdaCore? (Re: AdaCore ... the Next SCO?)

[Jacob Sparre Andersen]

Ludovic Brenta wrote:

> - because license terms can change with the weather, or be different
> for each licensee. They can even change after a licensee has had a
> license for some time.
>
> This is true of all software, free or proprietary, cost or no cost.

Not quite.  GPL, LGPL and GMGPL are all explicitly non-revocable.
AdaCore _cannot_ change a license after it has been given, if it is
one of these three - no matter what they claim.

> So basically, the user needs assurance about two things: who the
> copyright holder really is, and what license terms are applicable at
> any given point in time and for this particular licensee.

If a file for a long time has been available for download from the
copyright holder's own web site with a clear and consistent license
notice, then I doubt very much that there is any doubt about which
license you have it under if you have downloaded it during that
period.

> To give proper assurance on both accounts, I think that the
> copyright holder should crypto-sign his software, and grant a
> perpetual, non-revokable, immutable license to
> everyone. Unfortunately, the GPL (GNAT-modified or not) is revokable
> and mutable (think "or any later version"), so it is possible that
> it not apply to all licensees.

GPL is _not_ revokable.  The "or any later version" clause is purely
an option for the licensee, not something the copyright holder can
use.

Greetings,

Jacob
-- 
�Saving keystrokes is the job of the text editor, not the
 programming language.�                    -- Preben Randhol

Re: AdaCore ... the Next SCO?

[michael bode]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> > I imagine Linus Torvalds answering millions of requests if kernel
> > x.y.z *really* is GPL or maybe something else because license notices
> > in the sources "have no legal force". He will send you to hell.
> 
> What is "legal force"?

If my memory does not fail me "legal force" is the term used in
remarks about the "special exception" in GtkAda header files beeing
irrelevant to the license under which GtkAda-programs can
be distributed.

> (I Can't speak for Linus.)

Me too, but would *you* like to answer several millions of such
enquiries?

> > So *if* one would believe the "no legal force" thing, the logical
> > consequence is to use commercial shrink-wrap EULA software and avoid
> > OSS at any cost.
> 
> Why should EULicenseA be different from GPLicense in this
> regard?

Because then typically you get some piece of paper (or some original
media with hologram) with "Certificate of Authenticity" with the
written license terms and I've never heard of a case where the vendor
would claim his own EULA irrelevant. Anyway it is much easier to prove
that this piece of paper or hologram CD came from $VENDOR than that
some .tgz file was downloaded before $DATE-$TIME. Especially if
$VENDOR refuses to tell you $DATE-$TIME.

Re: AdaCore ... the Next SCO?

[michael bode]

"Ludovic Brenta" <ludovic@ludovic-brenta.org> writes:

> The reasons why the headers have no legal force, and for the blanket
> approval above, according to Robert Dewar at FOSDEM 2006, are:
> - because anyone besides the copyright holder may have changed them in
> the software.

This is simple to fix:

s/look into the headers/look into the headers as they are at $URL/
where $URL refers to a server under control of the author.

> - because license terms can change with the weather, or be different
> for each licensee. They can even change after a licensee has had a
> license for some time.

There is also a simple fix:
When you change the license also change the version, so that

$VERSION < x.y => $LICENSE = A
$VERSION >= x.y => $LICENSE = B

It could be even clearer if you host a public accessible CVS
repository. And of course do proper documentation.
 
> In the case of AdaCore, there is past evidence that suggests, but does
> not prove for lack of crypto signatures, that AdaCore is indeed the
> copyright holder of most of the software they offer on their CVS or web
> servers (but not GNAT itself, which belongs to the FSF). I think that
> AdaCore could successfully convince a court that they are indeed the
> copyright holder. If that's the case, then they are the ultimate
> authority on the license terms.

I've no doubt they are.

> The problem is that AdaCore have destroyed confidence in the license
> terms they apply to their software.

Exactly.

> My interpretation of the license change is this:
> 
> - if I download anything from AdaCore, I receive it under the terms of
> the pure GPL, no matter what the headers say. I have a statement from
> them in my records to back this claim (unfortunately, it is not
> crypto-signed, so one could argue it has no legal force either :))
> 
> - the software I downloaded in the past, and which is now in Debian
> Sarge, was under the terms of the GMGPL, and by these terms I am
> allowed to redistribute this software under GMGPL terms. I think this
> is correct because AdaCore has not explicitly revoked my GMGPL license
> to this old software (I also have emails on record to back this claim).

It would be much cleaner if they would simply put out a new version of
their libs with the headers cleaned up to what they want the license
to be. Then version 2.4.0 = GMGPL and 2.x.y = GPL. And for their
paying customers they call it 2.x.y.Pro or whatever and make it GMGPL.

Now since they have a CVS server where you can even today get 2.8 with
GMGPL headers, then 2.x.y must be > 2.8. That's bad luck for them but
I doubt it would hurt them too much.

> The software in Sarge is crypto-signed with my GPG key, do if you get
> the software from Debian, you have assurance that *I* am redistributing
> the software to *you* under the GMGPL terms.

According to one interpretation of the GMGPL that was discussed here,
this would be illegal if your are not the author. I would say this is
a rather stupid interpretation but IANAL.

> But let me reiterate: I don't think AdaCore will sue anyone. In fact,
> this whole ongoing conversation is evidence that several people are
> going out of their way to be really, really honest and respectful of
> AdaCore's copyright, and trying to understand the ramifications. I know
> many people who wouldn't be so considerate.

And let me clarify that Adacore can do what they want with their
software but please they should do it in a clear way. (And if they
would ask me: let the libs be GMGPL :-))

> Actually, the situation of the Linux kernel is quite uncertain, as
> there are many copyright holders, not all of whom can be reached, and
> it would be impossible to get assurances for every bit of the kernel.
> But, as for the portions that Linus owns, I think it would be easy to
> convince a court that he is the copyright holder.

My point was that free software does work with downloads from the
internet and license terms written in header files and not with
hologram CDs and license terms written on paper with Authenticity
Certificates and stickers with metal threads. If you want to abolish
the license in the header files, some guys from the more popular
projects will get some massive paperwork to do.

> No; the problem remains the same whether or not the software is free or
> open-source. You can buy pirated copies of Microsoft Windows,
> shrink-wrapped, in China and other places, or you can get pirated
> software bundled with a new computer from some small vendors.

Ok, you have to look here 
http://www.microsoft.com/resources/howtotell/en/default.mspx
if you got counterfeit products, but this is a different thing than
proving the date of the download of some .tgz.

> So, theoretically, you are still and always responsible for getting
> a proper license statement from the copyright holder.

You get one. It's in the box or on the hologram CD.

Re: AdaCore ... the Next SCO?

[Marius Amado-Alves]

> It would be much cleaner if they would simply put out a new version of
> their libs with the headers cleaned up to what they want the license
> to be.

Isn't that what they did?

/*
Curiously enough in the academic distro the new headers have blank  
lines where the exception used to be (instead of deleting those lines  
entirely). Gato escondido com rabo de fora.

------------------------------------------------------------------------ 
------
--                                                                       
     --
--                         GNAT LIBRARY  
COMPONENTS                          --
--                                                                       
     --
--                G N A T . S P I T B O L . P A T T E R N  
S                 --
--                                                                       
     --
--                                 S p e  
c                                  --
--                                                                       
     --
--           Copyright (C) 1997-2005 Ada Core Technologies,  
Inc.            --
--                                                                       
     --
-- GNAT is free software;  you can  redistribute it  and/or modify it  
under --
-- terms of the  GNU General Public License as published  by the Free  
Soft- --
-- ware  Foundation;  either version 2,  or (at your option) any  
later ver- --
-- sion.  GNAT is distributed in the hope that it will be useful, but  
WITH- --
-- OUT ANY WARRANTY;  without even the  implied warranty of  
MERCHANTABILITY --
-- or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public  
License --
-- for  more details.  You should have  received  a copy of the GNU  
General --
-- Public License  distributed with GNAT;  see file COPYING.  If not,  
write --
-- to  the Free Software Foundation,  59 Temple Place - Suite 330,   
Boston, --
-- MA 02111-1307,  
USA.                                                      --
--                                                                       
     --
--
--
--
--
--
--
--
-- GNAT was originally developed  by the GNAT team at  New York  
University. --
-- Extensive contributions were provided by Ada Core Technologies  
Inc.      --
--                                                                       
     --
------------------------------------------------------------------------ 
------
*/

Re: AdaCore ... the Next SCO?

[Hyman Rosen]

Ludovic Brenta wrote:
> The reasons why the headers have no legal force
...
> I think that AdaCore could successfully convince a
> court that they are indeed the copyright holder.

The natural mistake people here (and ACT as well) are making is to
assume that law works like computers. For example, you are recommending
technological solutions to ward off "spoofing attacks" on the license
and copyright of programs.

That's not how things work. In law, especially in license and contract
disputes, the assumption is that everyone is honest and that things
mean what they say. People wind up in court over disputes in meaning
and interpretation. Deliberate dishonesty and fraud is a much rarer
thing. Furthermore, law accomodates itself to the way things are.
Should a licensing dispute about Ada generic packages wind up in court,
it will be demonstrated that there are many open source libraries
available on the internet, and that it's routine to download them and
aggregate them under the terms of the licenses they contain in their
headers and associated license files. It will also be demonstrated that
for many of the libraries that ACT supplies and claims are distributed
only under the GPL, there are either identical or ancestral versions
available elsewhere with the extra exemption. The onus will be on ACT
to explain why, if they wished to change the license terms, they did
not follow normal practice and indicate the changed terms in the files
they distribute. The answer is going to be laziness or attempted
entrapment, and neither will sit well with the court.

Dewar's comments are those of a cautious (Ada!) programmer - he says
that since you have no way of knowing who might have altered the
headers of the files, you cannot trust them and so they are irrelevant,
and therefore they should be ignored, regardless of what they say. A
lawyer would laugh at this. A lawyer's assumption would be that things
must mean what they say and say what they mean. Regardless of whether
someone down the road might fraudulently alter the license of a file,
it is the responsibility of the distributors to make sure that the
license terms are given clearly and unambiguously if they intend to
later enforce their ownership rights. You've all seen the automatic
disclaimers that company lawyers attach to outgoing e-mails, the
warning labels on possibly dangerous products, and so on. Do you really
think that a lawyer would sanction sending out files that contain
deliberate misstatements of their licensing terms? It's laughable.

Re: AdaCore ... the Next SCO?

[Michael Bode]

Marius Amado-Alves <marius@amado-alves.info> writes:

>> It would be much cleaner if they would simply put out a new version of
>> their libs with the headers cleaned up to what they want the license
>> to be.
>
> Isn't that what they did?

Maybe in some places and not in others (GtkAda).

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Georg Bauhaus]

On Tue, 2006-07-18 at 13:55 +0200, michael bode wrote:

> I've never heard of a case where the vendor
> would claim his own EULA irrelevant.

The Free Software license text is typically spelled out
in the file COPYING. 

>  Anyway it is much easier to prove
> that this piece of paper or hologram CD came from $VENDOR than that
> some .tgz file was downloaded before $DATE-$TIME. Especially if
> $VENDOR refuses to tell you $DATE-$TIME.

Yes, Ludovic has mentioned digital signatures. Some licenses
are printed on paper. If you buy a compiler, you should get
some. (AdaCore ships paper, too, or has, according
to the archives.)

Microsoft offers lengthy license texts, including on redistributables.
Some paragraphs are known to not apply in some contries, still the
license text suggests they do (to the layman, at least).
The holographic trick is, I believe, a way to make selling
forged copies of software disks more difficult. A lot like obfuscation
is useful in making software theft more difficult. And probably
the looks of holographic disks help marketing, too.
But I am not aware of holographic print being used as a signature.
Is it a new form of signing?

I wonder why industry is so reluctant in adopting digital signatures.
Are they afraid of cryptography in general? An unintelligible myth?
Less opportunity for travel because you cannot shake hands over signed
contracts when using eMail? Increased dependence on inferiors to sort
out the technical stuff?  ;-)


-- Georg 

Re: AdaCore ... the Next SCO?

[Michael Bode]

Georg Bauhaus <bauhaus@futureapps.de> writes:

> Microsoft offers lengthy license texts, including on redistributables.
> Some paragraphs are known to not apply in some contries, still the
> license text suggests they do (to the layman, at least).
> The holographic trick is, I believe, a way to make selling
> forged copies of software disks more difficult. 

Yes and if you have this holographic logo on your CD this is a very
strong hint that the EULA on that CD is in fact Microsoft's EULA and
not changed by someone else. No need to call Bill Gates to check the
license.

> A lot like obfuscation is useful in making software theft more
> difficult. And probably the looks of holographic disks help
> marketing, too.  But I am not aware of holographic print being used
> as a signature.  Is it a new form of signing?

Sorry, I did not mean holographic signature. 

> I wonder why industry is so reluctant in adopting digital signatures.
> Are they afraid of cryptography in general? An unintelligible myth?
> Less opportunity for travel because you cannot shake hands over signed
> contracts when using eMail? Increased dependence on inferiors to sort
> out the technical stuff?  ;-)

All of that ;-)

-- 
Michael Bode

Re: AdaCore ... the Next SCO?

[Ludovic Brenta]

Michael Bode <m.g.bode@web.de> writes:

> Marius Amado-Alves <marius@amado-alves.info> writes:
>
>>> It would be much cleaner if they would simply put out a new version of
>>> their libs with the headers cleaned up to what they want the license
>>> to be.
>>
>> Isn't that what they did?
>
> Maybe in some places and not in others (GtkAda).

Not in their CVS server, in fact, but in all the software they ship as
part of GPL Editions, AFAICT.

-- 
Ludovic Brenta.

Re: Revocable or non-revocable licenses from AdaCore? (Re: AdaCore ... the Next SCO?)

[Samuel Tardieu]

Ludovic> Unfortunately, the GPL (GNAT-modified or not) is revokable
Ludovic> and mutable (think "or any later version"), so it is possible
Ludovic> that it not apply to all licensees.

Jacob> GPL is _not_ revokable.  The "or any later version" clause is
Jacob> purely an option for the licensee, not something the copyright
Jacob> holder can use.

Exactly. Moreover, as a developper you are free to release your code
under the GPLv2 without including the "or any later version". That is
what I have been doing for most of the software I have released under
the GPL.

I like the FSF, but I haven't been able to look at its bylaws (did
anyone ever have access to them?) so I don't know whether I can trust
them concerning future versions of the GPL.

  Sam
-- 
Samuel Tardieu -- sam@rfc1149.net -- http://www.rfc1149.net/

Re: AdaCore ... the Next SCO?

[Björn Persson]

Marius Amado-Alves wrote:
> Curiously enough in the academic distro the new headers have blank lines 
> where the exception used to be (instead of deleting those lines 
> entirely).

Probably so that the line numbers will be the same in GPL and GMGPL 
editions.

-- 
Bj�rn Persson                              PGP key A88682FD
                    omb jor ers @sv ge.
                    r o.b n.p son eri nu