Ada in Debian: most libraries will switch to the pure GPL in Etch
Ludovic Brenta
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.
Jeffrey Creem
> 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.
Ludovic Brenta
> 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.
Alex R. Mosteo
> 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.
Dr. Adrian Wrigley
> 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.
Ludovic Brenta
> 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.
Dr. Adrian Wrigley
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
Martin Krischik
> 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://kris...@users.sourceforge.net
Ada programming at: http://ada.krischik.com
Dmitry A. Kazakov
> 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
Georg Bauhaus
> 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.)
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.
M E Leypold
"Ludovic Brenta" <lud...@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
M E Leypold
"Ludovic Brenta" <lud...@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
M E Leypold
Jeffrey Creem <je...@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
M E Leypold
"Ludovic Brenta" <lud...@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
Michael Bode
> 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
Georg Bauhaus
> # 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
Michael Bode
>> 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
Ed Falis
>> 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
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
Michael Bode
>> 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
Georg Bauhaus
> "Ludovic Brenta" <lud...@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.
Michael Bode
> 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
Florian Weimer
> 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.
Georg Bauhaus
> 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?
Michael Bode
>> 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
Jeffrey R. Carter
>
> 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
Ludovic Brenta
I did download quite a lot from AdaCore: GtkAda, AWS, XML/Ada, and
also GPS but that's not a library.
--
Ludovic Brenta.
Georg Bauhaus
> 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
Ludovic Brenta
> 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.
Ludovic Brenta
> 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.
Ludovic Brenta
> 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.
Michael Bode
> 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
Simon Wright
> - 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 ...
Björn Persson
> 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?
--
Björn Persson PGP key A88682FD
omb jor ers @sv ge.
r o.b n.p son eri nu
Jeffrey Creem
> 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.
>
>
> 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
> 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.
M E Leypold
"Jeffrey R. Carter" <spam.not...@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
M E Leypold
Georg Bauhaus <bau...@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
M E Leypold
Martin Krischik <kris...@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
M E Leypold
Georg Bauhaus <bau...@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
M E Leypold
"Dr. Adrian Wrigley" <am...@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
M E Leypold
Ludovic Brenta <lud...@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
M E Leypold
Ludovic Brenta <lud...@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
M E Leypold
Michael Bode <m.g....@web.de> writes:
> "Ludovic Brenta" <lud...@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
M E Leypold
Michael Bode <m.g....@web.de> writes:
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
M E Leypold
Ludovic Brenta <lud...@ludovic-brenta.org> writes:
Ah, but did you download from libre or from libre2? :-)
Regards -- Markus
M E Leypold
Ed Falis <fa...@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
M E Leypold
Ludovic Brenta <lud...@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
M E Leypold
Georg Bauhaus <bau...@futureapps.de> writes:
> On Tue, 2006-06-27 at 20:39 +0200, Michael Bode wrote:
> > "Ludovic Brenta" <lud...@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
M E Leypold
Georg Bauhaus <bau...@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
M E Leypold
Georg Bauhaus <bau...@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
M E Leypold
Ludovic Brenta <lud...@ludovic-brenta.org> writes:
> Michael Bode <m.g....@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
M E Leypold
Michael Bode <m.g....@web.de> writes:
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.
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
Georg Bauhaus
> 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.
Simon Wright
> 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.
Michael Bode
> 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
Simon Wright
> 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.
Simon Wright
question doesn't arise. And given this discussion they are not going
to.
Martin Krischik
> > 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
Martin Krischik
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
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
Georg Bauhaus
> > 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!
Dr. Adrian Wrigley
> A license is not the same as a contract. A contract has to have
> 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.
Georg Bauhaus
> "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.
Georg Bauhaus
> 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?
Florian Weimer
> 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).
Ludovic Brenta
> 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.
Preben Randhol
> > 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
Carroll, Andrew
>> - All software downloaded from AdaCore is pure GPL, no matter what
the
>> headers say.
If this were true then why can't "a person" just change the headers to
say whatever they want the headers to say?
Also, if this the common understanding then why is this understanding
not written in the "free software" licensing headers (for any level of
"General Purpose")?
Sounds to me like the issue isn't what the headers say/contain or
complying with the headers. The problem is in the fact that there's
really no regulation of what the headers say or who complies or doesn't.
Compare this issue with say the Bill of Rights in the U.S. Constitution.
If my right to free speech was just a superficial header (had no
regulation) then the right to free speech I claim could be contended
either way with no one having a basis for their argument other than a
superficial header that was never enforced or regulated.
Just my two cents worth.
Andrew
Preben Randhol
>
> 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
Georg Bauhaus
> 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.
Preben Randhol
> 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
Georg Bauhaus
> 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?
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.
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!
Ed Falis
> 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
Michael Bode
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
Florian Weimer
> 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.
Georg Bauhaus
> Michael Bode wrote:
>> Georg Bauhaus <bau...@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?
Florian Weimer
> 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.
Georg Bauhaus
> 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.
Michael Bode
>> 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
Michael Bode
> 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
Simon Wright
> 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.
Simon Wright
> 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).
Frank J. Lhota
>> 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!
Michael Bode
> 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
Ludovic Brenta
> Ed Falis <fa...@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.
Ludovic Brenta
> * 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.
Ludovic Brenta
> Ludovic Brenta <lud...@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.
Ludovic Brenta
I'd ask the author before asking a lawyer. Keep the lawyers at bay
whenever you can :-)
--
Ludovic Brenta.
Ludovic Brenta
>>> Georg quoted
>>> - All software downloaded from AdaCore is pure GPL, no matter what
> the
>>> headers say.
>
> If this were true then why can't "a person" just change the headers to
> say whatever they want the headers to say?
>
> Also, if this the common understanding then why is this understanding
> not written in the "free software" licensing headers (for any level of
> "General Purpose")?
>
> Sounds to me like the issue isn't what the headers say/contain or
> complying with the headers. The problem is in the fact that there's
> really no regulation of what the headers say or who complies or doesn't.
No, the issue is the one you described in your first sentence. The
headers or COPYING file have no legal force, precisely because you
cannot prove that they apply to the software, and that they were
written by the author of the software. So, what Robert Dewar said is:
to be sure, you must ask the author.
--
Ludovic Brenta.
Ludovic Brenta
> 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.
Simon Wright
> 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).
Ludovic Brenta
> 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.
Ludovic Brenta
> 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.
Simon Wright
> 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).
Florian Weimer
> Florian Weimer <f...@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.
Florian Weimer
> 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/>
Björn Persson
> Can I:
>
> - Have sources S1, S2, S3 with S1 and S2 being dirtributed to me
> under the GMGPL whereas S3 is GPL.
>
>
> as the GPL demands.
>
> 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.
>
> 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.
--
Björn Persson PGP key A88682FD
omb jor ers @sv ge.
r o.b n.p son eri nu