Re: Eligibility of CC-BY-SA for documentation within a software project

[Anton S. Lytvynenko]

Dear Karl, dear Ineiev,

thank you very much for your answers.

First, to be more specific, I mean this project:

https://sites.google.com/site/mjollnirmagn/

https://bitbucket.org/featar/mjollnir/src/default/

It is devoted to the analysis of the experimental magnetochemical data
for paramagnetic coordination compounds. Briefly, it takes user-defined
model (number and kinds of paramagnetic ions as well as the interactions
between them) and adjusts the model parameters to fit the experimental
curve. The more or less usable releases were developed years ago and has
been utilizing since that times by a few scientific groups working in
this field. After that, some slow development together with checking of
various ideas has been performing, and now we are slowly approaching --
I hope! -- the next major release. The source code has been hosting on
Bitbucket, that is now shutting down Mercurial support. And, after that,
I am severely concerned by what they are going to shut down in the next
time. I fought against "humble advises" of some colleagues to put the
distribution of the software under heavy restrictions and wouldn't like
to loose the entire effort due to further "policy updates" of somebody...

Regarding the documentation license, it is not a problem to re-license.
However, in that case, to be honest, I would prefer, if it is possible,
keep both options to choose (dual GFDL+CC-BY-SA). There are a few
reasons for that, primarily due to a decent piece of the physical
background description incorporated into the document -- I believe that
at least that piece could be reusable in various teaching and/or
scientific works, and feel that GFDL imposes a bit rigid constraints
with notable overhead on that options. E.g. it is obscure to me to how
someone should borrow the set of the equations with the explanations for
the slides of a lecture taught to students on the computer offered by a
university (allegedly with some non-free requisites) obeying GFDL (I
don't say it is not possible or especially hard, but it is not easy to
comprehend the correct consequence of the steps).

There is another question emerging in that case -- am I allowed to clone
the entire repository with all previous history of commits (some of
which turned out to be inconsistent with the policies) to Savannah after
adjusting the current commits to the requirements?

In principle, as an option, I can keep developing only the code on
Savannah (which is under GPL from the beginning of the development) and
host the documentation elsewhere (especially bearing in mind that it is
a bit non-synchronized with the actual development code). However, it is
unlikely to remove a set of files from the entire history of the commits.

The examples of the config files are another concern -- technically,
they are artworks (description of the paramagnetic system and generated
"experimental" curves in a predefined format intended to be parsed by
the program), but are a part of the documentation as well (in
particular, some of them are included into the manual). The copyright
statement together with the entire GFDL looks a bit weird for these files.

Yours,

Anton

28.02.20 08:17, Ineiev пише:
> As far as I can recall, we are not allowed to accept gpl for manuals,
>>> although that seems draconian
> ...
>
> The GFDL is the license for GNU manuals; if some documentation
> is FDL-incompatible, GNU packages won't be able to use it, and
> it's expected that the GNU project should be able to copy
> from packages hosted on Savannah.