[License-discuss] license for code used for scientific results?

Kevin Hunter hunteke at earlham.edu
Mon Apr 30 15:36:57 UTC 2012


Hullo List,

For a scientific computing project, I'd like to encourage redistribution 
of software upon publication of research results (e.g. academic 
journals) by third parties using our code.  I'm not currently aware of 
any license geared toward this scenario.  (There are lots of examples of 
academic journals requiring this, but I'm having a difficult time 
tracking down a license for /code/ that is generally used to inform an 
academic publication.  More specifically, the sub-sub discipline for 
which the code I have written will potentially be used currently does 
not run rich with a FOSS ethos.)

As an example, if we license our code under the GPL, then folks who 
publish papers with results from use of our software are never 
redistributing any /software/, only publishing a paper, and are thus not 
required to share code.  However, we feel from a scientific standpoint 
that it's crucial that the code (*and* data) used for publication 
results be shared.

The closest I've been able to find is the EPPA license, which has this 
as provision 6 (currently available from: 
http://globalchange.mit.edu/research/IGSM/eppadl_form ):

-----
6. Any party which publishes results from a modification of the Software 
Model must also publish the source code of their modifications, in the 
same form as the Software Model is here released, and under the same 
license terms.  Said party must cause the modified files to carry 
prominent notices stating that the files were changed and the date of 
the change.
-----

Unfortunately, there are other parts of the license which are not 
acceptable to us (e.g. we do not want to limit non-academic use)

So, are there other licenses geared toward code (and data) used for 
academic publications?

Also, in the case that there are no other licenses acceptable/available 
to us, a second, meta-license question: How are the texts of the various 
OSI approved licenses themselves licensed?  Are they free for folks to 
modify?  (Perhaps to be described as a 'modified-OSI license' when used 
...?)

Also, please, I'm not looking for responses along the lines of "you 
can't enforce it so ignore it."  I'm very specifically focused on the 
licensing aspect.  (As they say, "One problem at a time!")

Thank you,

Kevin



More information about the License-discuss mailing list