Any interest in moving GHC issues / wiki to github?

We have a repeated problem with the GHC wiki and issue tracker getting filled with spam, and the server is currently taking about a minute to respond to each attempt to load a web page. I wonder if anyone would be open to us migrating the content to github instead. (Actually, while I'm at it, cloning from github is much faster than from darcs.haskell.org, but that's a fight for another day.)

Hi, Am Freitag, den 05.07.2013, 14:36 -0700 schrieb Bryan O'Sullivan:
We have a repeated problem with the GHC wiki and issue tracker getting filled with spam, and the server is currently taking about a minute to respond to each attempt to load a web page.
I wonder if anyone would be open to us migrating the content to github instead. (Actually, while I'm at it, cloning from github is much faster than from darcs.haskell.org, but that's a fight for another day.)
just joined this list, but this will not prevent me from adding my 2¢ :-). While github has become a great platform for sharing code, I find its issue tracker quite simplistic; good and simple for small projects with few bugs (<50 open bugs, maybe), but insufficient for projects with custom issue workflow, things like milestones and issue owners etc. In that respect, trac is pretty good, and so (with my bug reporter hat on) I’d prefer trace, at least as long as it is sustainable to maintain it. Greetings, Joachim -- Joachim “nomeata” Breitner mail@joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata@joachim-breitner.de • GPG-Key: 0x4743206C Debian Developer: nomeata@debian.org

Cloud Haskell is using a free open source licence donated by atlassian, for their hosted Jira platform. It requires login to submit issues and has a rich permissions scheme. All I had to do was register as an open source project and they contacted me via email to set everything up. They also offer the confluence wiki (which I'm not usin since it doesn't format Haskell code properly) and hosted bamboo ci service free to open source projects. In general, the infrastructure is really reliable and fairly easy to set up and use.
Jira (and the other products) provides a rich API that can be utilised to automate migration from other platforms, though there was some manual work involved when we moved from GitHub issues to Jira.
Cheers,
Tim
On 5 Jul 2013, at 22:52, Joachim Breitner
Hi,
Am Freitag, den 05.07.2013, 14:36 -0700 schrieb Bryan O'Sullivan:
We have a repeated problem with the GHC wiki and issue tracker getting filled with spam, and the server is currently taking about a minute to respond to each attempt to load a web page.
I wonder if anyone would be open to us migrating the content to github instead. (Actually, while I'm at it, cloning from github is much faster than from darcs.haskell.org, but that's a fight for another day.)
just joined this list, but this will not prevent me from adding my 2¢ :-).
While github has become a great platform for sharing code, I find its issue tracker quite simplistic; good and simple for small projects with few bugs (<50 open bugs, maybe), but insufficient for projects with custom issue workflow, things like milestones and issue owners etc. In that respect, trac is pretty good, and so (with my bug reporter hat on) I’d prefer trace, at least as long as it is sustainable to maintain it.
Greetings, Joachim
-- Joachim “nomeata” Breitner mail@joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata@joachim-breitner.de • GPG-Key: 0x4743206C Debian Developer: nomeata@debian.org _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

I believe some of those issues will be resolved with the pending move of a
lot of haskell.org infrastructure to a better dedicated box that gershom
and others have started undertaking. Not sure what the precise timeline is,
but they're actually working on spending volunteer time making it so, and
it will be quite I believe.
migrating all the tickets and such would be a pretty colossal undertaking,
and the ROI of that vs just getting the dedicated hosting upgraded doesn't
seem to be there, and would take longer.
I might be missing some pointed, but all the trac stuff and friends will be
getting a nice update soon once the hosting infrastructure is updated. (the
ghc trac is hosted with the other haskell.org software right?)
On Fri, Jul 5, 2013 at 5:36 PM, Bryan O'Sullivan
We have a repeated problem with the GHC wiki and issue tracker getting filled with spam, and the server is currently taking about a minute to respond to each attempt to load a web page.
I wonder if anyone would be open to us migrating the content to github instead. (Actually, while I'm at it, cloning from github is much faster than from darcs.haskell.org, but that's a fight for another day.)
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

On 05/07/13 22:36, Bryan O'Sullivan wrote:
We have a repeated problem with the GHC wiki and issue tracker getting filled with spam, and the server is currently taking about a minute to respond to each attempt to load a web page.
I wonder if anyone would be open to us migrating the content to github instead. (Actually, while I'm at it, cloning from github is much faster than from darcs.haskell.org http://darcs.haskell.org, but that's a fight for another day.)
I don't see any spam - where are you looking? Also the Trac is loading in a second or two for me. We do occasionally have to revert some spam, but I like the fact that with Trac we're in control of our own infrastructure. This has turned out to be useful several times in the past - when we want some extra functionality, we go find a Trac plugin to do it, or we implement the change ourselves (or rather we ask Ian nicely to do it :-). We have a lot more functionality in our issue tracker than github does. There are a couple of pressing issues with Trac. One, searching is crap. Using Google with "site:hackage.haskell.org/trac/ghc" is much better (you can set up a shortcut in your browser to do this, there are some instructions on the wiki somewhere). Two, repo browsing was too slow, so we had to turn it off. This will be fixed by upgrading Trac, which I believe Ian is planning to do as part of the server migration. Cheers, Simon
participants (5)
-
Bryan O'Sullivan
-
Carter Schonwald
-
Joachim Breitner
-
Simon Marlow
-
Tim Watson