Re: Working towards a new implementation of Debbugs?

[Glenn Morris]

Hi,

Thanks for writing.

The things you describe about providing a nicer frontend (web-based
submission, etc) all sound like good ideas. Though I don't know whether
they should exist as alternatives to the current one, or just replace it.

Ricardo Wurmus wrote:

> Last I heard the GNU instance of Debbugs was looking for someone to pick
> up maintainenance, to "despecialize" it with regard to the upstream
> version used by Debian.

There are two aspects that really need attention.

1) The system is running on a by now very dated Debian installation, in
a fairly hand-patched way. It really needs to be updated to current
Debian stable, in a more organised fashion.

2) The debbugs instance is locally patched, in-place, from an older
upstream debbugs. I think these local changes are good ones, but in some
cases upstream has since addressed the same issues.
Noam Postavsky did some excellent work in integrating these local
patches into a git repository based on a more recent upstream debbugs,
but it hasn't been deployed on the actual system.

There is a newer version of debbugs.gnu.org that is based on a newer
Debian and an updated debbugs, but plans to deploy it in place of the
existing one have been stalled for a long time. 
Some other folks started helping out a while ago, but I dropped the ball
on helping them switch things over (I really have little enthusiasm for
working on debbugs.gnu.org any more).

> I wonder if there might be value in switching to a new implementation
> of a Debbugs-like system written in Guile by extending Mumi. 

I don't know. It's less clear to me that there is value in
reimplementing the backend. I guess if it makes things easier to
develop, or perform better, or attracts more contributors, then maybe.
(I have no idea how active upstream debbugs is these days.)


But mainly, I think the future of debbugs.gnu.org is down to whoever wants
to do the work. That isn't me (and hasn't been for a long time).
I certainly wouldn't stand in the way of anyone who wanted to develop
things.


> PS: to avoid hitting the debbugs.gnu.org web interface too hard to fetch
> mbox files, would it be possible to get rsync access to the file system
> holding the raw mbox files?

Hah, I just saw this now. :)