taskSavannah Administration - Tasks: task #15348, Submission of Free Molecular...

 
 

task #15348: Submission of Free Molecular Dynamics

Submitter:  Arham Amouei <a_amouei>
Submitted:  Mon 29 Jul 2019 03:02:31 PM UTC
   
 
Should Start On:  Mon 29 Jul 2019 12:00:00 AM UTC Should be Finished on:  Thu 08 Aug 2019 12:00:00 AM UTC
Category:  Project Approval Priority:  * 5 - Normal
Status:  Cancelled Privacy:  Public
Assigned to:  ineiev Open/Closed:  Closed
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Tue 20 Aug 2019 10:39:34 AM UTC, comment #19: 

Mature people accept their failures and see the reality as it is, not as one wishes it to be. The GNU project, though very helpful, was only partially successful. Now there is no OS that can be called GNU.

Thanks for the review.

Arham Amouei <a_amouei>
Tue 20 Aug 2019 10:22:23 AM UTC, comment #18: 


> Respect doesn't justify using misnomers and seeing the reality in a distorted way. I also have respect for the developers of LibreOffice, X11, Firefox, etc. But should I include these in the name of the whole system? As you know, a distribution like Ubuntu or CentOS includes software from different origins.
>
> Informed people use the name "Linux distribution" for these Desktop systems not because Linux makes up the biggest part of the system, but because it is the heart of the system, the most vital single module. If you are worried that people may think that the whole system is originated from Mr Torvalds, I will explain the meaning of the term "Linux distribution" in the documentation, which is not written yet.


All these points are addressed in the GNU/Linux FAQ.

> You are right that the phrase "Operating Systems with Linux kernel" includes Android which is not the target of this package. Adding the modifier "Desktop" at the beginning may solve it but makes the phrase a little too long!


This also reminds me the argument where the OP said, "a two-legged animal without feathers and with broad flat nails."

I believe the further discussion is pointless.  We've clearly established that you don't agree with Savannah hosting requirements.

Ineiev <ineiev>
Site Administrator
Tue 20 Aug 2019 09:21:47 AM UTC, comment #17: 

Thanks for the reply.

Okay. So the only remaining point is the system name.

>  "GNU+Linux" will do; if your respect is so great, you could call it just "GNU".


Respect doesn't justify using misnomers and seeing the reality in a distorted way. I also have respect for the developers of LibreOffice, X11, Firefox, etc. But should I include these in the name of the whole system? As you know, a distribution like Ubuntu or CentOS includes software from different origins.

Informed people use the name "Linux distribution" for these Desktop systems not because Linux makes up the biggest part of the system, but because it is the heart of the system, the most vital single module. If you are worried that people may think that the whole system is originated from Mr Torvalds, I will explain the meaning of the term "Linux distribution" in the documentation, which is not written yet.

You are right that the phrase "Operating Systems with Linux kernel" includes Android which is not the target of this package. Adding the modifier "Desktop" at the beginning may solve it but makes the phrase a little too long!

Arham Amouei <a_amouei>
Tue 20 Aug 2019 07:20:11 AM UTC, comment #16: 


> Actually I'm paying great attention to the notices, thanks to what I've learned from people like you. But README and the example files do not constitute the main part of this project. It's very unlikely that one day we face a dispute over who their exact copyright holders are


Lightning strikes are also very unlikely; however, in certain cases lightning rods are mandatory, even though they don't always protect from lightning.

> (even if that happens the records in the git repository will settle that).


The records in the git repository typically don't contain the names of copyright holders.  They record committers and authors (if the committer cares to write them down), no copyright holders.  Also, the records of the repository won't help at the point where the repository isn't available.

> I won't insist on using the phrase "The authors of FMD" in the notices of those files, but it would look awkward if, for example, the names of 10 people are listed in a short readme file because each of them added one paragraph or so.


If people made a small addition, copyright notices needn't mention them ("In that case, you should always include a copyright notice in the name of main copyright holder of the file."); moreover, contributions of 5-6 lines are not copyrightable at all.

> Two final remarks:
> 1. Despite my really great respect and debt to people who have worked on GNU, I don't like the term GNU/Linux.


"GNU+Linux" will do; if your respect is so great, you could call it just "GNU".

> I'm going to use phrases like "Operating Systems with Linux kernel"


Does your package really depend on Linux more than on GNU?  I guess, the users would have less issues with running it on top of GNU/kFreeBSD or maybe even GNU/Hurd than on top of Android, wouldn't they?

> 2. We may have more than one host for this project (2 hosts are enough). I hope this isn't against your rules. The git repositories on all hosts will always be up-to-date.


Yes, using additional external facilities is up to you.

Ineiev <ineiev>
Site Administrator
Sun 18 Aug 2019 12:54:46 PM UTC, comment #15: 

Thanks for the reply.

> What actually matters is whether the maintainers of the package basically understand where these parts of copyright notices come from, and whether they have a commitment to maintain the notices.


Actually I'm paying great attention to the notices, thanks to what I've learned from people like you. But README and the example files do not constitute the main part of this project. It's very unlikely that one day we face a dispute over who their exact copyright holders are (even if that happens the records in the git repository will settle that). I won't insist on using the phrase "The authors of FMD" in the notices of those files, but it would look awkward if, for example, the names of 10 people are listed in a short readme file because each of them added one paragraph or so. Please inform me of your final judgment about the usage of the phrase "The authors of FMD" in the mentioned files. I'll prepare the files according to that.

Two final remarks:
1. Despite my really great respect and debt to people who have worked on GNU, I don't like the term GNU/Linux. I'm going to use phrases like "Operating Systems with Linux kernel" or, more generally, "Unix-like systems" instead.
2. We may have more than one host for this project (2 hosts are enough). I hope this isn't against your rules. The git repositories on all hosts will always be up-to-date.

Arham Amouei <a_amouei>
Sun 18 Aug 2019 11:16:05 AM UTC, comment #14: 


> But if you are not familiar with the legal aspects enough, how can you decide if the copyright notices in a package are "valid" or not?


The copyright notices should have the established form and list the right years and copyright holders.  Of course, in most cases I hardly can tell who is the real copyright holder of the files I wasn't involved in writing (as opposed to the files I write myself).  However, this isn't essential.

What actually matters is whether the maintainers of the package basically understand where these parts of copyright notices come from, and whether they have a commitment to maintain the notices; so far, I'm getting an impression you haven't.

> Some projects use...


It also doesn't matter what other people do.  You and me are not responsible for their actions.

> And I see that even many GNU projects does NOT abide by all the things you say. Just for an example, the files ChangeLog and THANKS of the latest gzip torball doesn't have notices.


Thank you!  Please file a bug report against gzip and make sure the issue is fixed.

Ineiev <ineiev>
Site Administrator
Fri 16 Aug 2019 10:01:15 AM UTC, comment #13: 

Thanks for the reply. But if you are not familiar with the legal aspects enough, how can you decide if the copyright notices in a package are "valid" or not? Some projects use collective copyright notices like

Copyright © 2016-2018 The TokTok team.
Copyright © 2013 Tox project.

(URL: https://github.com/TokTok/c-toxcore)

Maybe you should consult with a lawyer in FSF about this because of your job and responsibility.

And I see that even many GNU projects does NOT abide by all the things you say. Just for an example, the files ChangeLog and THANKS of the latest gzip torball doesn't have notices.

Arham Amouei <a_amouei>
Thu 15 Aug 2019 09:13:16 AM UTC, comment #12: 


> Is there a way that we declare all the contributors (including
> those in the future) as writers/copyright holders of some files?


I'm not a lawyer, but as far as I know, assigning copyright needs some kind of paperwork from the current copyright holders.  Every time you add a contributor, you would need signed papers from every previous contributor.

> It is not important to us who the exact authors of some files are.


Having valid copyright notices is important for the users of your package.  If you have no real commitment to maintain these records carefully, let us not register your package on Savannah.

Ineiev <ineiev>
Site Administrator
Tue 13 Aug 2019 11:41:08 AM UTC, comment #11: 

I'm learning from you and I'm thankful for that.

> In README.txt, the copyright note refers to AUTHORS; this
> is too easy to get wrong when AUTHORS is edited.  I'd
> suggest writing down the actual authors instead.


Is there a way that we declare all the contributors (including
those in the future) as writers/copyright holders of some files?
It is not important to us who the exact authors of some files are.

We may add new names to AUTHORS but no name will be removed.

Arham Amouei <a_amouei>
Tue 13 Aug 2019 09:32:52 AM UTC, comment #10: 

Thank you!

In README.txt, the copyright note refers to AUTHORS; this
is too easy to get wrong when AUTHORS is edited.  I'd
suggest writing down the actual authors instead.

Likewise, Python files in examples/ refer to AUTHORS
instead of explicitly listing them.  These files also
say that a copy of CC0 should be provided with them,
but no copy of CC0 is included.

(If you add a copy of CC0, please note that the text
of it has no copyright and license notices, you'll
need to add them like in task #15279.)

Files like examples/log.lammps.argon.morse have no notices.

Please check other files as well.

Ineiev <ineiev>
Site Administrator
Sun 11 Aug 2019 01:42:37 PM UTC, comment #9: 

To facilitate the process, I have also made the following changes to the package:

  • added copyright and license notices to src/makefile
  • added copyright and license notices to setup.py
  • added copyright and license notices to tests/test_molframe.py
  • changed README.md to README.txt and added copyright and license notices (GNU All-Permissive License)
  • deleted the file inside potentials directory, because wasn't sure about its license (I've written to its author to see if it can be distributed with a GPL-licensed software -- but it's not an essential part of the package)
  • dedicated examples/*.c & examples/*.py & examples/in.lammps.argon to public domain by including related notices


The two files examples/log.lammps.argon.lj and examples/log.lammps.argon.morse are log files produced by the free software LAMMPS, when it processes the input script file examples/in.lammps.argon. We have included them so that the user can easily compare the results of our software with the output of the famous LAMMPS.

Please see the updated tarball.

Thanks,
Arham

Arham Amouei <a_amouei>
Sat 10 Aug 2019 07:34:58 PM UTC, comment #8: 

Now I understand free licenses and the meaning of compatibility with GPL much better.

Here is an updated list of the dependencies:

GNU Scientific Library:

License: GPL v3 or later
URL: https://savannah.gnu.org/projects/gsl

Message Passing Interface (MPI)

There are different implementations of MPI, some of which are free software. For example, Open MPI uses "3-clause BSD license" (Modified BSD license).
URL: https://www.open-mpi.org/community/license.php

NumPy

License: 3-clause BSD license (Modified BSD license)
URL: https://numpy.org/index.html

ASE

License: LGPL v2.1 or later
URL: https://wiki.fysik.dtu.dk/ase/about.html

periodictable

License: Public Domain
URL: https://pypi.org/project/periodictable/

mpi4py

License: FreeBSD license (2-clause BSD license)
URL: https://pypi.org/project/mpi4py/

Arham Amouei <a_amouei>
Tue 06 Aug 2019 11:47:37 AM UTC, comment #7: 

I see. I think within a week I can learn what I need to know about free licenses.

Arham Amouei <a_amouei>
Tue 06 Aug 2019 09:15:37 AM UTC, comment #6: 

We have no resources to continually check all hosted packages for compliance. The GNU project does provide a set of licensing-related pages like https://www.gnu.org/licenses/license-list.html https://www.gnu.org/licenses/gpl-faq.html https://www.gnu.org/licenses/gpl-howto.html, but in general, we have to rely on the ability of package maintainers to tell whether licenses are compatible and whether they are free.

Ineiev <ineiev>
Site Administrator
Thu 01 Aug 2019 03:26:56 PM UTC, comment #5: 

I am sorry for my mistake about the license of ASE. In the following updated list, I have used the license names as mentioned by the authors/maintainers themselves. Honestly, I do not have detailed knowledge about free licenses available, so I should leave it to you to judge if the licenses of NumPy and mpi4py are GPL-compatible.

The dependencies include:

GNU Scientific Library

License: GPL
URL: https://www.gnu.org/software/gsl

Message Passing Interface (MPI)

There are different implementations of MPI, some of which are free software. For example, Open MPI uses "3-clause BSD license".
URL: https://www.open-mpi.org/community/license.php

NumPy

License: BSD
URL: https://numpy.org/index.html

ASE

License: LGPL 2.1 or any later version
URL: https://wiki.fysik.dtu.dk/ase/about.html

periodictable

License: Public Domain
URL: https://pypi.org/project/periodictable/

mpi4py

License: BSD
URL: https://pypi.org/project/mpi4py/

Arham Amouei <a_amouei>
Thu 01 Aug 2019 09:49:00 AM UTC, comment #4: 

https://wiki.fysik.dtu.dk/ase/about.html says "LGPL 2.1 or later", not "GPL"; also, the wording "BSD-like" is too vague to define a license.  It includes at least one GPL-incompatible license, which is unacceptable for Savannah.

Could you rewrite this list more carefully?

Ineiev <ineiev>
Site Administrator
Wed 31 Jul 2019 03:44:14 PM UTC, comment #3: 

The dependencies include:

  • GNU Scientific Library

License: GPL
URL: https://www.gnu.org/software/gsl

  • Message Passing Interface (MPI)

There are different implementations of MPI, some of which are free software. For example MPICH, a very widely used one, uses a BSD-like license.
URL: https://www.mpich.org/downloads/

  • NumPy

License: BSD-like
URL: https://numpy.org/license.html

  • ASE

License: GPL
URL: https://wiki.fysik.dtu.dk/ase/about.html

  • periodictable

License: Public Domain
URL: https://pypi.org/project/periodictable/

  • mpi4py

License: BSD
URL: https://pypi.org/project/mpi4py/

One part of the code needs Python to run.

We like the name "fmd", but if you insist "fmdlib" is also acceptable to us. :)

I also attached a tarball. Please let me know if anything else should be done as well. Thank you.

(file #47280)

Arham Amouei <a_amouei>
Wed 31 Jul 2019 02:28:53 PM UTC, comment #2: 


comment #1:

> Please list all your dependencies with links to their licensing terms; also, please provide a tarball of your package.
>
> Also, it would be nice if the System Name were longer and more descriptive; could you suggest a new one?

Arham Amouei <a_amouei>
Tue 30 Jul 2019 10:26:39 AM UTC, comment #1: 

Please list all your dependencies with links to their licensing terms; also, please provide a tarball of your package.

Also, it would be nice if the System Name were longer and more descriptive; could you suggest a new one?

Ineiev <ineiev>
Site Administrator
Mon 29 Jul 2019 03:02:31 PM UTC, original submission:  

A new project has been registered at Savannah
This project account will remain inactive until a site admin approves
or discards the registration.


Registration Administration


While this item will be useful to track the registration process,
approving or discarding the registration must be done using the specific Group Administration page, accessible only to site administrators,
effectively logged as site administrators (superuser):



Registration Details


  • Name: Free Molecular Dynamics
  • System Name:  fmd
  • Type: non-GNU software and documentation
  • License: GNU General Public License v3 or later





Description:

Free Molecular Dynamics (FMD) is a library for performing classical molecular dynamics (MD) simulations on different kinds of computers, from laptops to computer clusters. Its core part, which particularly executes all computationally expensive operations, is written in C for efficiency. Another part of FMD, written in Python and named PyFMD, adds features which are essential for many, if not most, simulations. PyFMD also provides Python programmers with an object-oriented interface for interacting with the core part of FMD within Python programs.


Other Software Required:

GNU Scientific Library
Message-Passing Interface (MPI)


Tarball URL:

https://github.com/a-amouei/fmd


Arham Amouei <a_amouei>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

Attached Files
file #47323:  fmd-2019-aug-11.tar.gz added by a_amouei (78KiB - application/gzip)
file #47280:  fmd.tar.gz added by a_amouei (553KiB - application/gzip - tarball)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by ineiev (Posted a comment)
  • -email is unavailable- added by a_amouei (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only logged-in users can vote.

     

    Follow 10 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2019-08-20 ineiev StatusIn Progress Cancelled
        Open/ClosedOpen Closed
    2019-08-11 a_amouei Attached File- Added fmd-2019-aug-11.tar.gz, #47323
    2019-08-06 ineiev Should Start On2019-08-01 2019-07-29
        Should be Finished on2019-08-01 2019-08-08
    2019-08-01 ineiev Should Start On2019-07-29 2019-08-01
        Should be Finished on2019-08-08 2019-08-01
    2019-07-31 a_amouei Attached File- Added fmd.tar.gz, #47280
    2019-07-30 ineiev StatusNone In Progress
        Assigned toNone ineiev

    Back to the top

    Powered by Savane 3.13-4448.
    Corresponding source code