[Debichem-devel] Looking for a mentor -- multiscale molecular modeling package MMB (and molmodel)

[Samuel Flores]

On Tue, Mar 3, 2020 at 5:50 PM Andrius Merkys <andrius.merkys at gmail.com>
wrote:

> Dear Samuel,
>
> Embedded 3rd party sources will most likely cause trouble in getting MMB
> into Debian. Please see the recent mail thread on packaging tensorflow
> [1] to get the feel.


I guess i am not on the right list, anyway i could not find this thread.


> If OpenMM people do not want their software in
> Debian,


They seem to have no objection, in fact they were flattered and
appreciative of the thought. They just didn't want to commit to supporting
this distribution method.


> maybe an alternative to their neighbourlisting function could be
> found?
>
>
Ah yes. Such a big package to include, for only one function. You are in
principle completely right, the issue is that most likely it would take me
forever to get around to that since it has no payoff other than this. The
other thing is that my collaborators just added the ccp4 package in the
same way (they checked and the license allows this), and are soon to add
another for mmCIF. So if 3rd party packages being included is a
dealbreaker, then I guess I am lost anyway.

I have little knowledge about Docker. However, I usually look inside the
> Dockerfile for the clues on building the package.
>
> Best wishes,
> Andrius
>
> [1]
>
> https://alioth-lists.debian.net/pipermail/debian-science-maintainers/2020-March/079054.html
>
> On 2020-02-25 16:14, Samuel Flores wrote:
> > Hi Andreas and Andrius,
> >
> > I guess we got stalled when the OpenMM guys declined to support debian
> > packages. At the time it was against my religion to fork supported
> > packages, but I have had a change of heart. I really only use the
> > neighborlisting function of OpenMM, and anyway I can always update the
> > local fork. So what I decided to do is just add the entire OpenMM
> > repository to the 3rdparty directory of my MMB git repo.  Also I put
> > MMB on the MIT license to comply with Debian's guidelines. Were these
> > the main stumbling blocks?
> >
> > In the many months since we last spoke, I have also created a docker
> > container (samuelflores/mmb-ubuntu). I just set this  to compile MMB
> > from the 3rdparty directory -- a simple change which hopefully will
> > build soon on dockerhub. Assuming that goes well the Dockerfile should
> > be a good model for compiling with apt-get. There are a couple of
> > people in Prague contributing new code, and I have also taken on a
> > couple of MS students so there should be advances this summer on the
> > code. So in short -- the project has advanced substantially in the
> > past few months and I think it is in a good place for packaging.
> >
> > I wonder if we can pick up the packaging project again?  Hopefully
> > this will go smoothly now that things have been worked on for a while.
> >
> > With kind regards,
> >
> > Sam
>
> --
> Andrius Merkys
> Vilnius University Institute of Biotechnology, Saulėtekio al. 7, room V325
> LT-10257 Vilnius, Lithuania
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debichem-devel/attachments/20200303/7bf4f20e/attachment.html>