[SciPy-dev] [Fwd: Re: license status of your code on netlib]

Benny Malengier benny.malengier@gmail....
Fri Sep 11 07:26:14 CDT 2009


This is actually the nice thing about the viral aspect of GPL. When
you start a new toolkit in your department by extending a GPL
component, you can tell the legal department that you don't have the
time to reimplement existing components (time pressure you know) so
you extend a GPL one, and now your component _must_ also be GPL. For
scipy this is more complicated, you must convince them you are better
of contributing to the project, and people who don't know the issue
have difficulty grasping the concept of this collaboration.

Hopefully you can later relax things to BSD, as within an academic
environment I see no reason why companies can not profit from the work
done. It is a reality however that when you are part of a project and
they see you will provide the modelling people are already asking if
it can 'exploited' later, and if so to add it to the part in the
project claiming what later can be done with the results.

It would be much easier if the European commission and other funding
organizations just require open source work under a set of approved
licences. After all, if it not distributed and just used internally by
the company the project is for (most likely) then GPL and BSD and
proprietary really don't matter as there is no distribution, but at
least a researcher has the right to take a copy of his code if he
obtains another position, or somebody else can grab it and do
something useful with it.

Benny

2009/9/11 David Cournapeau <cournape@gmail.com>:
> On Fri, Sep 11, 2009 at 7:47 PM, Robin <robince@gmail.com> wrote:
>> At the risk of deviating off topic - while we're talking about
>> licenses I was wondering if I could solicit some opinions.
>>
>> In my academic area there is a move towards groups releasing toolkits
>> and packages for use by the wider community, and most of these refer
>> to it as 'open source', but they are usually distributed under strange
>> licenses. Some examples:
>>
>> http://find.bccn.uni-freiburg.de/?n=Main.License
>> http://neurodatabase.org/src/license/
>> http://www.ibtb.org/Home/toolbox-downloads#currentversion
>> http://www.nest-initiative.org/index.php/Software:License
>>
>> I guess I wondered what people thought about this.
>
> This is bad in general, but people often have little choice, because
> those licenses often are imposed by the legal dpt or similar from the
> university/research institude in my experience.
>
>> I'm looking for
>> convincing arguments for people to stick to standard open source
>> licenses where possible. I think the general impression from the above
>> is that in most cases GPL would be more suitable than BSD - since they
>> specifically want to prevent commercial exploitation.
>
> There is this:
>
> http://google-opensource.blogspot.com/2008/05/standing-against-license-proliferation.html
>
> I think a quite strong argument, besides license compatibility, is
> that few licenses are actually tested in court. I trust the GPL quite
> a bit, since it has been written by people who knew about what they
> were doing, and the BSD is well trusted as well. If someone violates
> my license under the GPL, I can have some little hope that the FSF
> will help me.
>
> Writing your own license has nowhere near those guarantees.
>
>> But what do
>> people suggest about the citation clauses and things like that?
>
> I don't think it is very useful - I doubt that people who are not
> willing to cite your work will do so because you have a license in
> your software, and it is not like you will sue someone because they
> did not cite your work. I think it is a vain attempt at circumventing
> a basic lack of ethics.
>
> I also wonder whether a citation clause would not bring the same kind
> of trouble as the advertisement clause in the old BSD.
>
> cheers,
>
> David
> _______________________________________________
> Scipy-dev mailing list
> Scipy-dev@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-dev
>


More information about the Scipy-dev mailing list