Tuncer Ayaz
On Thu, Aug 6, 2015 at 4:51 PM, Brandon Allbery wrote:
So are we going to do anything about Google Code Issues going away in a little over two weeks?
First question is whether existing tickets have to be preserved. If so, then where should the migration lead to?
Second question is whether Darcs a strict requirement. If so, hub.darcs.net could work, as darcs support on haskell.org seems to be deprecated.
Either way, wasn't the plan to move to haskell.org's infrastructure and leverage Phabricator (with hg or git)? For a feature overview, see http://phabricator.org/comparison/. That seems like a good "future-proof" path to take, if you ask me.
Alternatively, Bitbucket, Gitlab, or Github could make sense. Gitlab has its own CI tool, so that's a plus, but other than that (for XMonad's case) there's no clear winner between the three.
Also, like the pending release, this is another administrative topic for which there seems to be nobody responsible who also happens to be blessed with enough free time. Therefore, it's important to assign the tasks, once the migration path has been decided.
We all have strong opinions about which tools and hosting providers we should be using. That seems to have stalled any movement on this front. We're not all going to agree on this so we should just accept that all of the options are workable and pick the one with the least amount of resistance. I volunteer to be the release manager for the project if we switch to Git and host the code and tickets on Github. It's not a perfect solution but it does have everything we need and there's almost an expectation these days that open source projects are hosted on it. This might have the side effect of increasing the number of contributions to xmonad. At the very least it gives the project more visibility. -- Peter Jones, Founder, Devalot.com Defending the honor of good code