[Pkg-mailman-hackers] [Mailman-cabal] Re: removal from Debian

[Pierre-Elliott Bécue]

Hi,

Sorry for taking so much time to answer.

Le mercredi 02 octobre 2019 à 10:37:00+0000, bendikker at vfemail.net a écrit :
> No, that was not a joke,
> 
> I do understand what you as mailman want to do. Im just surprised
> that mailman is a gnu package.
> You as a GNU package could at least inform GNU. That is why
> I now also cc maintainers at gnu.org .

I think you miss some intel regarding the current situation.

Being part of GNU only means two things : being a free software and
being part of a GNU/Linux system. It doesn't bind any component one to
another. In that sense, there is nobody to warn, should mailman3 have
been depending on libmatheval and willing to change such a fact.

BUT, it's not the case. Mailman3 *transitively* depends on libmatheval
*in Debian*, because as the maintainer of Mailman3 suite in Debian, I've
decided to rely on uwsgi to provide the web frontend, and uwsgi does
depend on libmatheval.

So here the Mailman3 developers are not making any choice. I, as a
Debian Developer, am making the choice to ask the *Debian* uwsgi
maintainers if they could drop libmatheval dependency so that uwsgi
doesn't get expelled from Debian, and hence, mailman3 doesn't get
expelled along with uwsgi.

There are many reasons why this path is the best one:

 1. libmatheval is no longer maintained
 2. In Debian, no package currently relies on the features the math plugin
    provided by uwsgi (this plugin being the only reason why uwsgi depends
    on libmatheval)
 3. Keeping libmatheval into Debian would mean making it guile-2.2
    compatible, which is a lot of work that should be done upstream, and
    that no developer is willing to do
 4. It's better to have some software out of a distro than having it
    ill-maintained with a stick for the sake of keeping something in the
    archive.

> Ok, Im not a programmer, but earlier I thought it is unsure you manage to
> find a solution.
> Maybe then, I thought, GNU itself might be able to manage libmatheval (which
> I myself strongly doubt).

This is not the first, and won't be the last GNU project that gets
unmaintained and takes the dust on a shelf. It's the principle of having
people working on their personal time on projects. Some live, other die.
It's a classic thing.

> Sorry, this is my last post on this subject.

I hope this helps you to understand my choice.

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/pkg-mailman-hackers/attachments/20191013/1a5310a0/attachment.sig>