[Exherbo-dev] On burying ::haskell

Arthur Nascimento tureba at gmail.com
Sat Aug 18 13:57:43 UTC 2018


Hi all,

On Sat, 18 Aug 2018 at 07:59, Mykola Orliuk <virkony at gmail.com> wrote:
> I'd like to vot against burying ::haskell.
> I don't see how this solves anything except of freeing some time for people who are dedicated to maintain ::haskell. If you'll move it to ::haskell-unofficial then you can achieve same, I guess.

Agreed. Removing ::haskell solves nothing. And shoving more unrelated
stuff into ::arbor is always a -1 for me too. Besides, if somebody
tried to hack ghc into their own repo, it's easy to just switch
::haskell for it, but it's much more work dealing with package
conflicts in ::arbor, masking packages and setting proper importance
properties on their repos...


> Yes. And reason for that is Exherbo policy. You have issues - you fix it.
> For long response on reviews - not enough people for official repos, I guess. Sometimes I consider making own fork of ::haskell because of that.

Sadly, the intersection of people with time, skill and interest in
some subjects in Exherbo is dimming every day. Considering that in the
past 6+ months this conversation has come up in irc quite a few times
and only the ~four of us ever demonstrated interest, I also came to
the conclusion that we should clone the repo in one of our users,
grant push permissions for the others, use it as a new blessed repo
source and go wild fixing and breaking things to our liking. The
'official' repos are only as official as the consensus of the users
go. If, in the future, a commiter gets interested again, it's as easy
as pie to merge a fast-forward MR back into ::haskell, with no harm
done.


> I think most of the issues and reasons can be applied to Exherbo itself. Isn't it? For me most of the problems you mentioned are not specific to ::haskell (except of slow reviews).

Indeed. The only difference in this subject is that is has a more
vocal person spearheading the discussion. :-)


> For me it feels like Paludis abandoned for years already. I tried to start working on it to get pluggable repo formats, but I quickly gave up because of specific codebase (as for me) and a bit unpleasant contribution experience.

Strongly agree in the experience. We all try; we all give up. Most
other projects would recognize a deficit in community management of
their part in this situation. But I won't comment any further on this.


In my view, part of our trouble is having too much rotting stuff. I
see that ::haskell shows 517 packages here, but our users only care
about gch, pandoc and git-annex (I also use postgrest, but we never
packaged it and I'm not in a hurry). I'd start by burying everything
that's not related to them, because while they're there, any changes
to exlibs will break hundreds of packages nobody cares about and
fixing the slotting or anything else that's important to us, will do
so also. Nobody wants to propose even a simple change that touches all
that; and nobody wants to review such change either! So there's a high
maintenance cost to having dead packages anywhere. (That is also my
view for all other repos, packages and even paludis code itself.)

It's only after burying these undead that we can start rewriting
exlibs and exheres and maybe designing a cabal repo format, all geared
toward satisfying the user's use cases raised in this thread.



Tureba - Arthur Nascimento



More information about the Exherbo-dev mailing list