subsurface issue tracking

Dirk Hohndel dirk at hohndel.org
Fri Sep 7 15:06:43 PDT 2012


Pierre-Yves Chibon <pingou at pingoured.fr> writes:

> On Fri, 2012-09-07 at 16:09 +0200, Pierre-Yves Chibon wrote:
>> On Fri, 2012-09-07 at 15:54 +0200, Henrik Brautaset Aronsen wrote:
>> > Den 07.09.12 15:48, skrev Dirk Hohndel:
>> > > On Sep 7, 2012, at 6:25 AM, Lubomir I. Ivanov wrote:
>> > >
>> > >> hi,
>> > >>
>> > >> people with more than one project on their heads can often forget what
>> > >> has to be done...
>> > >>
>> > >> would it be a good idea to enable an "issue tracker" somewhere ?
>> > >> github has one, which looks modern and pretty, but i haven't tried it much.
>> > >> it probably would suffice.
>> > > I think Linus is not a huge fan of the way github does things.
>> > > I really don't want to implement a bugzilla instance, but let me look around if there is something nice and easy that I can do.
>> > > It's definitely a good idea.
>> > 
>> > How about we create a subsurface organization on Github: 
>> > https://github.com/account/organizations/new, and host the mainline code 
>> > & issues from there?  It's free for OS projects.
>> > 
>> > That way others could manage the issues as well.
>> 
>> Surely is one option but clearly not the most free (as in FOSS not $).
>
> One that looks *a lot* like github is gitlab: http://gitlabhq.com/
> It is written in ruby though, so according to the server my be
> annoying/a blocker.

I looked at this - it looks slick and very similar to github, but it
would be a pain to integrate into our workflow - all branches that
people would want Linus or me to pull from would have to live on our
server. The email integration is so-so. And the issue tracking is quite
lame.

> After there is always trac or roundup in python or mantis in php (I
> won't mention bugzilla (perl) since it has already been
> mentio^rejected).

I have looked at several systems over the past couple of years because
of other requirements outside of subsurface. They are either extremely
simplistic (like the one in gitlab) or they are so overblown and complex
that they seem completely overengineered for what we would need here.

So right now I think I'll fall back to "post issues to the mailing list"
and then I can easily search my list archives for things I still need to
work on :-)

/D


More information about the subsurface mailing list