Boost license 1.0 is not OSI compliant (i.e. not an open source license)
Alf P. Steinbach
"mostly constant time" class with Boost license, <url:
home.no.net/alfps/cpp/lib/alfs_v06.zip>; Boost license because I gather
that most C++ practitioners are familiar with Boost and can feel safe
with that license -- no more hidden legal traps than we so far have
had the impression that using Boost has.
Now I thought I'd put that project on SourceForge, mainly to gain some
experience with open source development. Also, perhaps others could
then participate. But it was rejected due to the Boost license, which
apparently does not permit distribution of the source code!
Quoting the SourceForge project rejection:
<quote>
Current status: Rejected by SF.net
Review comment: Greetings,
This project is being rejected at this time as the selected license is
not OSI compliant. The main point here is that the license you selected
does not explicitly state that that the source code may also be freely
distributed.
I recommend that you select a different license, the BSD license may be
more along the lines you seek, and resubmit this project request. We
will approve it at that time if you select that license or other similar
such OSI compliant license.
Best regards,
Daniel Hinojosa - Sr. Manager Support, SourceForge.net
</quote>
Checking... Sure enough, there's no Boost license in the OSI list, at
<url: http://www.opensource.org/licenses/alphabetical>. And this
license is a few years old. If it was OK as open source, presumably it
would be in the OSI list by now?
Since OSI doesn't recognize the Boost license as open source, are all we
who have simply downloaded and installed the Boost library in source
code form, not paying a penny, in violation of Boost's license?
Or should we not use the Boost license for open source C++ code?
Cheers (scratching his head),
- Alf
--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
Maciej Sobczak
> Quoting the SourceForge project rejection:
[...]
This guy must be new there. :-)
Show them this link (hint: look the domain name):
Boost is not the only project hosted by SF that uses the Boost
license.
Also, the following project:
uses the Boost license and is very explcit about it (scroll down the
page).
The SourceForge guys agreed for Boost and SOCI, so they can agree for
your project as well.
I think you should try again. Good luck. ;-)
I also think that it makes sense to push them to display the Boost
license in their official list of licenses. Otherwise all this just
shows their inconsistency.
--
Maciej Sobczak
http://www.msobczak.com/
--
Alex Shulgin
Hm... in my opinion boost license is compatible with the modified BSD
license. Also, please note that gnu.org lists it not just as a free
software license, but even GPL-compatible one! (see
http://www.gnu.org/licenses/license-list.html)
I suggest you submit your project to Savannah (savannah.nongnu.org).
Savannah is a central point for development, distribution and
maintenance of Free Software that runs on free operating systems. And
Savannah hacker do really respect computing freedom. :-)
> Since OSI doesn't recognize the Boost license as open source, are all we
> who have simply downloaded and installed the Boost library in source
> code form, not paying a penny, in violation of Boost's license?
I think they are plain wrong--the mere fact that license does not
mention 'source code' does not make it non-free. In this particular
case it seems like source code is the only way of redistributing boost
libraries (might not be valid for derived works, tough), so probably
license authors didn't bother.
> Or should we not use the Boost license for open source C++ code?
I guess so :-) Why people tend to invent a new license for just about
every single library anyway?.. Wouldn't it be easier to adopt GPL,
BSD and MIT? (Even these three is far more than enough, as for me.)
This would certainly save a _lot_ of time and effort.
Kind regards,
Alex
PS: some useful (hopefully) links :-)
http://www.fsf.org/ -- Free Software Foundation
http://www.gnu.org/ -- The GNU Project
http://www.gnu.org/philosophy/philosophy.html -- Philosophy of the
GNU Project
http://www.gnu.org/philosophy/open-source-misses-the-point.html
--
LR
> As a hobby project, I've been fiddling with a ref-counted C++ string
> "mostly constant time" class with Boost license, <url:
> home.no.net/alfps/cpp/lib/alfs_v06.zip>; Boost license because I gather
> that most C++ practitioners are familiar with Boost and can feel safe
> with that license -- no more hidden legal traps than we so far have
> had the impression that using Boost has.
IANAL. In my experience I do not understand the law. I particularly do
not understand the law the as I think a lawyer might. (As an exercise,
to help me understand why this might be, I've tried to imagine how a
lawyer might write C++ code.)
> Now I thought I'd put that project on SourceForge, mainly to gain some
> experience with open source development. Also, perhaps others could
> then participate. But it was rejected due to the Boost license, which
> apparently does not permit distribution of the source code!
That doesn't seem to be the actual complaint of the person who rejected
your project.
Rather the rejection seems to be because distribution of the source code
is not explicit in the boost license.
[snip lead in to quote of rejection]
> This project is being rejected at this time as the selected license is
> not OSI compliant. The main point here is that the license you selected
> does not explicitly state that that the source code may also be freely
> distributed.
[snip rest of quote]
> Checking... Sure enough, there's no Boost license in the OSI list, at
> <url: http://www.opensource.org/licenses/alphabetical>. And this
> license is a few years old. If it was OK as open source, presumably it
> would be in the OSI list by now?
I'm not really qualified to say. I think that certain recent
developments (if my understanding is correct) may require the removal of
the GPL if this is their criteria. Well, perhaps only in the USA.
> Since OSI doesn't recognize the Boost license as open source, are all we
> who have simply downloaded and installed the Boost library in source
> code form, not paying a penny, in violation of Boost's license?
Again, IANAL, but no, I wouldn't think so. Or at least probably not.
AFAIK courts, no matter their jurisdiction, are *not* required to use
the same criteria for interpreting a license as SF uses to approve it
for use on their site. Just as SF is not required, absent an enforceable
order to the contrary, to use a court's interpretation to select
licenses they permit for use on their site.
> Or should we not use the Boost license for open source C++ code?
I am not a judge either. And the answers to this rather complicated
question may, probably will, vary from jurisdiction to jurisdiction.
I think that if you have any doubts about this you may benefit from
consulting an attorney or solicitor who is versed in IP law. Perhaps
even if you yourself are a lawyer.
<very short rant>
The GPL and it's ilk might be seen as an attempt to impose morality on
the law. I often think that these kinds of attempts, not just those
initiated by programmers, tend to founder on the jagged rocks hidden
just under the surface of the placid sea of good intentions.
<\very short rant>
My free (and worthless) advice is that The Law is not Computer
Programming Code and vice-versa and that licenses, contracts and indeed
the law itself, will not always be interpreted by people who agree with
you or share your desires or have your best interests or at heart. In
fact the contrary is very likely. Please act accordingly.
LR
--
Nemanja Trifunovic
>
> Quoting the SourceForge project rejection:
>
> <quote>
> Current status: Rejected by SF.net
> Review comment: Greetings,
>
> This project is being rejected at this time as the selected license is
> not OSI compliant. The main point here is that the license you selected
> does not explicitly state that that the source code may also be freely
> distributed.
>
> I recommend that you select a different license, the BSD license may be
> more along the lines you seek, and resubmit this project request. We
> will approve it at that time if you select that license or other similar
> such OSI compliant license.
>
> Best regards,
>
> Daniel Hinojosa - Sr. Manager Support, SourceForge.net
> </quote>
You should point out to Mr Hinojasa that there are other projects
hosted at SourceForge with the same license. An example:
https://sourceforge.net/projects/utfcpp
--
Alexander Terekhov
"Alf P. Steinbach" wrote:
[...]
> Checking... Sure enough, there's no Boost license in the OSI list, at
> <url: http://www.opensource.org/licenses/alphabetical>. And this
> license is a few years old. If it was OK as open source, presumably it
> would be in the OSI list by now?
http://www.crynwr.com/cgi-bin/ezmlm-cgi?3:mss:14120
(Subject: For Approval: Boost Software License - Version 1.0 - August
17th, 2003)
regards,
alexander.
--
Alexander Terekhov
Alex Shulgin wrote:
[...]
> license. Also, please note that gnu.org lists it not just as a free
> software license, but even GPL-compatible one! (see
> http://www.gnu.org/licenses/license-list.html)
I just wonder in what sense is it "GPL-compatible one". According to
RMS it means permission to "relicense"*** under the GPL terms, but
being a non-exclusive licensee that doesn't explicitly grant the right
to sublicense it just can't be that permissive.
http://www.eapdlaw.com/newsstand/detail.aspx?news=9
"Licensees who are granted non-exclusive licenses generally do not
receive the right to grant sublicenses, in part because the potential
sublicensee can obtain a direct license from the licensor. In fact, it
is generally held that a non-exclusive patent licensee cannot grant
sublicenses unless it is expressly granted such right. However,
exclusive licenses commonly include a right to sublicense, at least
with the consent of the licensor."
http://www.coolcopyright.com/cases/fulltext/bridgeportdimensiontext.htm
"See Leicester v. Warner Bros., 47 U.S.P.Q.2d 1501, 1998 U.S. Dist.
LEXIS 8366 (C.D. Cal. 1998), aff'd, 232 F.3d 1212 (2d Cir. 2000). In
Leicester, [**18] a real estate developer employed an artist to
create sculptural elements for inclusion in the courtyard of a
building under construction in Los Angeles. The artist granted the
owner the exclusive right to make three-dimensional copies of the
work, and a non-exclusive right to make two-dimensional or pictorial
copies. The developer allowed a motion picture company to film the
sculptural elements as part of a movie. The artist sued the motion
picture company, claiming infringement, on the grounds that the
developer did not have the right to sub-license his non-exclusive
right to make two-dimensional or pictorial copies. During the course
of the litigation, the developer was granted a "sub-license" by the
building's architect, who the court found to be a co-owner with the
artist of some of the elements. The court found that the architect
could not grant a sub-license to the developer because a non-exclusive
license could not be sub-licensed. Id. at *17, 1998 U.S. Dist. LEXIS
8366. "
regards,
alexander.
***) http://fsfeurope.org/projects/gplv3/barcelona-rms-transcript.en.html
"The idea is that there are some other Free Software licences which
are compatible with the GPL meaning that if a program is released
under one of those licences, that licence gives, effectively,
permission to relicence under the GPL. There are two ways that can
happen. Some licences explicitly say "you can also use this program
under the GNU GPL". In other cases, it's because the licence is so
permissive that to relicence it under the GNU GPL is permitted."
Rene Rivera
<http://boost.org/more/license_info.html>.
Alexander Terekhov wrote:
> "The idea is that there are some other Free Software licences which
> are compatible with the GPL meaning that if a program is released
> under one of those licences, that licence gives, effectively,
> permission to relicence under the GPL. There are two ways that can
> happen. Some licences explicitly say "you can also use this program
> under the GNU GPL". In other cases, it's because the licence is so
> permissive that to relicence it under the GNU GPL is permitted."
Note, RMS is not a lawyer ;-)
--
-- Grafik - Don't Assume Anything
-- Redshift Software, Inc. - http://redshift-software.com
-- rrivera/acm.org - grafik/redshift-software.com
-- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
Alex Shulgin
> Alex Shulgin wrote:
>
> [...]
>
> > license. Also, please note that gnu.org lists it not just as a free
> > software license, but even GPL-compatible one! (see
> >http://www.gnu.org/licenses/license-list.html)
>
> I just wonder in what sense is it "GPL-compatible one". According to
> RMS it means permission to "relicense"*** under the GPL terms, but
> being a non-exclusive licensee that doesn't explicitly grant the right
> to sublicense it just can't be that permissive.
IANAL, however, I think it means you have the right to combine
(#include) boost and some GPL-ed code to form "a derived work" and
then distribute it as a whole on the terms of GPL. I do not see any
problems with boost being used in GPL-ed projects so I would basically
agree with GNU classification of boost license.
Cheers,
Alex
PS: hope this didn't went too much off the topic :)
David Abrahams
> Quoting the SourceForge project rejection:
>
> <quote>
> Current status: Rejected by SF.net
> Review comment: Greetings,
>
> This project is being rejected at this time as the selected license is
> not OSI compliant. The main point here is that the license you selected
> does not explicitly state that that the source code may also be freely
> distributed.
When boost.org actually had a lawyer, we had him trying to get us OSI
certification. He wrote, of that experience:
They failed to respond. I believe I cc'ed you on all of my
submissions,
so you should have copies. I harassed them for a response, but
after an
initial "we'll take a look", I heard nothing. In fact, I think they
are
actively trying to limit the number of OSI-approved licenses. So
it's a
bit of an up-hill battle.
My personal impression is that OSI is more a hobby for a few people
than a
structured organization, and that they feel no obligation to respond
to
requests (especially since they don't ask for $).
--
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com
The Astoria Seminar ==> http://www.astoriaseminar.com
--
Alex
wrote:...
> You should point out to Mr Hinojasa that there are other projects
> hosted at SourceForge with the same license. An example:https://sourceforge.net/projects/utfcpp
Here's our $.02 for Boost license ;-)
http://sourceforge.net/projects/poco
Alex
David Abrahams
on Fri Sep 14 2007, "Alf P. Steinbach" <alfps-AT-start.no> wrote:
> Now I thought I'd put that project on SourceForge, mainly to gain some
> experience with open source development. Also, perhaps others could
> then participate. But it was rejected due to the Boost license, which
> apparently does not permit distribution of the source code!
>
> Quoting the SourceForge project rejection:
>
> <quote>
> Current status: Rejected by SF.net
> Review comment: Greetings,
>
> This project is being rejected at this time as the selected license is
> not OSI compliant.
FYI, we've re-submitted our application for OSI approval and have
received the board's promise that they will finally consider it.
http://www.crynwr.com/cgi-bin/ezmlm-cgi?3:sss:14482:200710:bakbhhjpafpgoioklknn#b
--
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com
[ See http://www.gotw.ca/resources/clcm.htm for info about ]