Perl 6 - the future is here, just unevenly distributed

IRC log for #darcs, 2014-03-05

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:07 mizu_no_oto joined #darcs
00:39 edwardk joined #darcs
01:36 mizu_no_oto joined #darcs
01:57 mizu_no_oto joined #darcs
02:00 dolio joined #darcs
02:10 stepcut joined #darcs
02:54 gh__ joined #darcs
03:03 alexei_ joined #darcs
03:32 stepkut joined #darcs
03:49 alexsuraci joined #darcs
03:49 heath joined #darcs
03:49 heath joined #darcs
05:04 favonia joined #darcs
06:42 lelit joined #darcs
07:58 raichoo joined #darcs
08:06 sm_ joined #darcs
09:00 heath joined #darcs
09:00 lelit joined #darcs
09:05 alexsuraci__ joined #darcs
10:35 gal_bolle joined #darcs
11:49 stepkut` joined #darcs
11:52 alexsuraci___ joined #darcs
11:53 florent__ joined #darcs
13:56 mizu_no_oto joined #darcs
14:15 mizu_no_oto joined #darcs
14:49 alexei_ joined #darcs
15:20 nomeata joined #darcs
15:38 mizu_no_oto joined #darcs
16:53 mizu_no_oto joined #darcs
16:59 gh__ joined #darcs
17:49 alexei_ joined #darcs
18:03 mizu_no_oto joined #darcs
18:09 edwardk joined #darcs
19:03 materia joined #darcs
19:04 kowey joined #darcs
19:08 alexei_ joined #darcs
19:28 alexei___ joined #darcs
20:00 alexei___ joined #darcs
20:03 alexei___ joined #darcs
20:58 kowey sm, alexsuraci: so… darcsden and hub.darcs.net…
20:58 kowey should they merge? or is there reason to hold off?
20:59 sm hi kowey
21:00 sm it's up to alex. darcsden.com is down right now
21:00 kowey hello! if I understand correctly, you both seem ok with it in principle
21:00 kowey partly about whether it would be wise, whether it would solve any actual problems
21:00 sm I wasn't happy with darcsden.com, hub.darcs.net is my replacement
21:01 kowey (eg. redundancy/confusion, repos going dead)
21:01 sm but as I tweeted, the immediate problem is the out of date package description
21:01 kowey would it be technically difficult to import darcsden users and have everything transparently migrate?
21:01 kowey sm: right, so in my vision, it'd be something that worked behind the scenes via cunningly arranged redirects
21:02 kowey or not-so-cunny
21:02 kowey cunning
21:02 kowey the main point being that darcs shouldn't be the cause of yet-another-yak-shave for somebody like Paul
21:02 kowey ie. somebody who'd like to do some Haskell stuff, but
21:03 kowey (context for folks on the channel having no idea what started this conversation, https://twitter.com/extempore2/status/441277552825360384 )
21:03 kowey interestingly the cabal file for ghc-mtl is up to date
21:03 kowey must just be the hackage pkg
21:05 sm I don't know how big a problem this is, questions include are there still many packages referencing darcsden.com ? How long has it been offline ? Does it have many active users ?
21:06 kowey i wonder how easy it'd be to search hackage
21:06 sm https://www.google.com/search?hl=en&q=site%3Ahackage.haskell.org%20%22darcsden.com%22 says 237
21:07 sm https://www.google.com/search?hl=en&q=site%3Ahackage.haskell.org%20%22darcsden.com%22#hl=en&q=site:hackage.haskell.org+%22hub.darcs.net%22 says  13300.. uh that can't be right
21:07 kowey different versions of same pkg?
21:08 sm ah right
21:08 kowey ok that's encouraging
21:08 kowey so another solution is to scrape the 237 results for authors and email them?
21:09 sm good idea
21:12 sm alexsuraci created darcsden and I'm "competing" with his site so I don't feel right telling him to shut it down, even if that's the right thing for the Darcs web presence
21:13 kowey yeah, it's up to him to determine if maintaining it brings any benefit to the community or to him (eg. as something to continue working on for the joy of working on something)
21:13 kowey but it's good to have your opinions about what is right for the community all other things being equal
21:14 * kowey apologises for only randomly jumping in on HCAR days or when somebody tweets something interesting about darcs :-)
21:14 kowey I've been getting a lot more perspective on Git branching lately
21:14 kowey and in a way am glad we never got around to figuring branches out
21:14 kowey I hope we do somebody, but holy cow, navigating smart-busy-people through Git remote tracking branches…
21:15 n-dolio I've become convinced that git branches aren't good.
21:15 kowey but so convenient!
21:16 kowey the scenario that brought all this up
21:16 n-dolio Many times they're not convenient.
21:16 kowey user clones my repo
21:16 kowey then starts hacking on it (they don't have a GitHub account)
21:16 kowey now they want to get the changes back to me, and likely continue hacking
21:16 kowey in darcsland, we say, ok, do a darcs send
21:16 * sm liked the look of gitless, a front end that simplifies some of git's staging & branching stuff
21:17 n-dolio A guy at the office was just talking to this guy on a scala list, complaining that from-scratch build times are slow.
21:17 kowey here I try to get them to set up their own GitHub repo, and change their remote aliases around, and point their local branch to their GitHub repo instead of mine
21:17 kowey yaks were shorn
21:17 n-dolio Because he has 3 git branches for different versions, and switching between them kills all build system info, so he has to clean and build from scratch.
21:17 kowey users were frustrated and confused…
21:18 n-dolio So my coworker says, "just use three directories, one for each branch."
21:18 kowey n-dolio: hmm! I thought the whole appeal was the shared-working-dir thing?
21:18 kowey hah!
21:18 n-dolio And the guy is too mentally blinded by git branches to understand that solution.
21:18 n-dolio Even after he's told the solution.
21:18 kowey :-(
21:18 kowey even though it's also quite technically feasible with git
21:18 n-dolio Yes, you can of course do it.
21:19 kowey just that local branch is the hammer for which everything appears to be a nail
21:19 n-dolio We use mercurial here, and I have one directory per branch, for that reason.
21:19 kowey it's conceptually easier
21:19 kowey people don't like the creating dirs aspects (having to do more filesystem org)
21:20 kowey and they don't like losing non-checked-in artefacts along the way
21:20 kowey (having to rebuild stuff, eg)
21:20 kowey but it's way way way simpler conceptually not to have the distinction
21:26 sm yup
21:30 kowey if all darcs repos were hosted --no-working-dir
21:30 kowey we could have a convention of branches being referred to by subdirs
21:31 kowey or if we had something like a .darcs
21:31 kowey darcs get http://hub.darcs.net/kowey/my-repo.darcs/new-features
21:31 kowey buh too late to think
21:32 kowey anyway, thanks, sm!
22:03 stepcut joined #darcs
22:05 sm night kowey!
22:07 colDrMcBeardman joined #darcs
22:55 edwardk joined #darcs
22:59 favonia joined #darcs
23:24 alexei___ joined #darcs
23:57 edwardk joined #darcs
23:57 favonia joined #darcs

| Channels | #darcs index | Today | | Search | Google Search | Plain-Text | summary