--- Log opened Mon Feb 18 00:00:56 2013
20130218 00:10:31-!- cjhopman [~chris@wesnoth/developer/cjhopman] has quit [Read error: No route to host]
20130218 00:11:25-!- cjhopman [~chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20130218 00:12:25< shadowm> !log 56143
20130218 00:12:26< shikadibot> shadowm: Revision 56143 (jamit) on Thu Jan 31 20:58:05 2013:
20130218 00:12:26< shikadibot> shadowm: Clear haloes when extracting a unit.
20130218 00:12:26< shikadibot> shadowm:  
20130218 00:12:26< shikadibot> shadowm: This cleans up a display artifact at the beginning of Delfador's Memoirs (when
20130218 00:12:29< shikadibot> shadowm: (+1 discarded lines)
20130218 00:12:32< shikadibot> shadowm: Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=56143
20130218 00:13:32< shadowm> jamit: Note that this issue has also affected previous stable series, including 1.10.x.
20130218 00:14:18< shadowm> It's always annoyed me when it happens in my campaign because I happen to have halo-using units who get MUFed a lot. :/
20130218 00:15:06< shadowm> I don't think this kind of thing counts as something any WML author would consider a feature, so maybe this could be backported to 1.10? (With a proper changelog entry, of course.)
20130218 00:15:15< shadowm> If someone has any problems with it they can use #ifver, after all.
20130218 00:15:17-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 00:16:15< shadowm> The patch applies cleanly on latest 1.10 and seems to work fine.
20130218 00:17:55< jamit> shadowm: I think that did not get backported because either I was being cautious modifying trunk or I forgot. Probably the latter.
20130218 00:38:40< irker774> wesnoth: jamit * r56340 /branches/1.10/ (src/unit_map.cpp changelog): 
20130218 00:38:41< irker774> wesnoth: Clear halos when extracting a unit.
20130218 00:38:41< irker774> wesnoth: This cleans up a display artifact at the beginning of Delfador's Memoirs (when
20130218 00:38:42< irker774> wesnoth: Methor moves in the opening scenario).
20130218 00:38:42< irker774> wesnoth: Backport of r56143.
20130218 00:39:04< shadowm> \o/
20130218 00:39:19< esr> Interesting.   Am about 70% of the way through fixing up comments in the repo conversion, and commentinfg practice is noticeably better later in the history... more and better summary lines.  Maybe git was already having an influence?
20130218 00:39:44< shadowm> esr: Most likely, since I started using git-svn around late 2008.
20130218 00:40:05< shadowm> mordante started using it earlier the same year, I believe.
20130218 00:40:24< irker774> wesnoth: jamit * r56341 /trunk/changelog: 
20130218 00:40:24< irker774> wesnoth: Changelog entry for r56143.
20130218 00:40:37< shadowm> Other people joined in the git-svn bandwagon later on.
20130218 00:42:21< shadowm> I'm still not sure why you are going through the effort of fixing up comments when people will continue to screw that up after the migration is done, especially those who intend to use Github's SVN-to-git mechanism.
20130218 00:42:35< esr> shadowm: Been meaning to ask:  the shadowmaster_filesystem branch, is it obsolete?  It has a mid-branch delete, which is one of the few artifacts that can still confuse reposurgeon.  I couyld fix it up with hand surgery but it would simplify my life if I could just drio that branch.
20130218 00:43:08< esr> shadowm: To fix up the references, basically.
20130218 00:43:11< shadowm> Let me figure out a way to ask past me or check in my git-svn tree whether it exists.
20130218 00:43:38-!- mjs-de [~mjs-de@g224182062.adsl.alicedsl.de] has quit [Remote host closed the connection]
20130218 00:44:09-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20130218 00:44:13< shadowm> !log 24501
20130218 00:44:13< shikadibot> shadowm: Revision 24501 (shadowmaster) on Mon Mar 10 17:39:30 2008:
20130218 00:44:13< shikadibot> shadowm: * Apply patch #989 (TESTING)
20130218 00:44:13< shikadibot> shadowm: Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=24501
20130218 00:44:21< shadowm> wesbot: patch 989
20130218 00:44:22< wesbot> Patch #989  Assigned to: None  Status: Wont Do  Priority: 3 - Low
20130218 00:44:22< wesbot> Summary: Store userdata on Windows in "My Documents"
20130218 00:44:22< wesbot> Original submission: This patch changes the default location for userdata to "
20130218 00:44:25< wesbot> My documents/Wesnoth/VERSION". The old behaviour ("Wesnoth/userdata")is preserve
20130218 00:44:28< wesbot> URL: https://gna.org/patch/?989
20130218 00:44:30< wesbot> Attached file (1st): https://gna.org/patch/download.php?file_id=3867
20130218 00:45:16< shadowm> !log 27137
20130218 00:45:16< shikadibot> shadowm: Revision 27137 (shadowmaster) on Thu Jun 12 23:23:42 2008:
20130218 00:45:16< shikadibot> shadowm: * Apply patch #989 for production purposes on the Win32 platform
20130218 00:45:16< shikadibot> shadowm: Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=27137
20130218 00:46:13< shadowm> esr: My git-svn tree has exactly one commit (r24501) in that branch, which is superseded in trunk by r27137 which doesn't sound like a SVN branch merge operation.
20130218 00:46:19< shadowm> Does that help you?
20130218 00:46:37< esr> Yeah, that means it's dead.  Thabks.
20130218 00:47:02< esr> (I thought that would be the case after looking at it.)
20130218 00:48:07< shadowm> I vaguely recall that it didn't go anywhere.
20130218 00:48:41< shadowm> Pretty much like every other public branch I've ever made.
20130218 00:50:10-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 00:54:01< esr> shadowm: I'm considering proposing a topic-tagging system for commits, where people would be encouraged to lead their summary lines with tags from a semi-controlled vocabulary like [wml], [mp], [story], [art], etc.  I'sd prime the pump by tagging the last 500 or so commits in the history?  Do you think our devs would actually use it?
20130218 00:55:21< shadowm> Square brackets are a nuisance when using regular expressions to search through commit logs.
20130218 00:55:32< esr> Good point.
20130218 00:55:47< esr> I'm not attached to that syntax,
20130218 00:55:58< esr> Mayve {foo} would be better.
20130218 00:56:16< shadowm> I started making use of "topic: Subject line here" at some point myself.
20130218 00:56:38< shadowm> But I don't really think everyone would stick to any format unless forced to.
20130218 00:57:24< esr> I don't think we have any way of forcing people.  It has to be good example and moral suasion.
20130218 00:57:41< shadowm> Anyway, I'm neither the project manager nor I can speak for every other developer (in particular the svn-loving people), so that sounds more fit for a ML proposal.
20130218 00:59:04< esr> I thought I'd bounce it off you because well, you have an attitude that some would call "cynical" and others "realistic".  I figure that if you don't bardf on the idea it will be an easier pitch to most others.
20130218 01:00:13< shadowm> I for one would stick to what I already do (just a colon and potentially forward slashes to denote subtopics, no other fancy punctuation) given the choice.
20130218 01:04:13< shadowm> But yeah, I've grown accustomed to using Git or git-svn for everything, so I'm not really representative of the svn-loving population that does have a small yet vocal presence in this project.
20130218 01:19:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20130218 01:20:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20130218 01:37:42-!- Octalot [~noct@host86-148-68-137.range86-148.btcentralplus.com] has quit []
20130218 01:49:03-!- _8680_ [~quassel@mb92036d0.tmodns.net] has joined #wesnoth-dev
20130218 01:53:05-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum]
20130218 01:53:32-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130218 01:53:37-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Client Quit]
20130218 02:04:24-!- _8680_ [~quassel@mb92036d0.tmodns.net] has quit [Remote host closed the connection]
20130218 02:50:56-!- jamit [~jamit@wesnoth/developer/jamit] has quit [Quit: Leaving.]
20130218 02:56:43-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 03:07:10-!- Crendgrim_ [~crend@port-92-204-86-254.dynamic.qsc.de] has joined #wesnoth-dev
20130218 03:11:48-!- Crendgrim [~crend@port-92-204-92-117.dynamic.qsc.de] has quit [Ping timeout: 276 seconds]
20130218 03:24:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20130218 04:03:39-!- Ivanovic_ [~ivanovic@dtmd-4db27ac9.pool.mediaWays.net] has joined #wesnoth-dev
20130218 04:04:15-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Read error: Operation timed out]
20130218 04:05:34-!- Ivanovic_ is now known as Ivanovic
20130218 04:08:48-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 256 seconds]
20130218 04:10:24-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130218 04:14:14-!- Elvish_Pillager [~eli@dhip-029.rrw.residences.colby.edu] has quit [Ping timeout: 276 seconds]
20130218 04:27:09-!- Artemius [~yigit@139.179.207.74] has quit [Quit: Ex-Chat]
20130218 04:32:01-!- _8680_ [~quassel@S010620aa4b739215.vw.shawcable.net] has joined #wesnoth-dev
20130218 04:42:00-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20130218 04:57:02-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20130218 05:04:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 05:05:27-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has joined #wesnoth-dev
20130218 05:28:29-!- ancestral [~ancestral@75-168-48-246.mpls.qwest.net] has joined #wesnoth-dev
20130218 05:51:00-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 06:02:10< irker774> wesnoth: jetryl * r56342 /trunk/data/core/ (339 files in 3 dirs): 
20130218 06:02:10< irker774> wesnoth: Added in sleepwalker's new, complete set of animations for the Dwarven Fighter line, as well as two associated macros.
20130218 06:10:28-!- Gallaecio [~quassel@84.120.114.158.dyn.user.ono.com] has joined #wesnoth-dev
20130218 06:22:25-!- Gallaecio [~quassel@84.120.114.158.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20130218 06:23:39-!- Gallaecio [~quassel@84.120.114.158.dyn.user.ono.com] has joined #wesnoth-dev
20130218 06:36:52-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 06:46:30-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has quit [Quit: bye]
20130218 06:53:54-!- oldtopman [~oldtopman@unaffiliated/oldtopman] has quit [Quit: *pouf*]
20130218 07:41:29-!- ancestral [~ancestral@75-168-48-246.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20130218 07:41:40-!- Ivanovic [~ivanovic@dtmd-4db27ac9.pool.mediaWays.net] has quit [Changing host]
20130218 07:41:40-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20130218 08:20:06-!- _8680_ [~quassel@S010620aa4b739215.vw.shawcable.net] has quit []
20130218 08:30:08-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20130218 08:38:35-!- Gallaecio [~quassel@84.120.114.158.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20130218 08:46:03-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130218 09:18:23-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20130218 09:37:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 09:44:27-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 09:46:33-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20130218 09:50:03-!- exciton_ [chuck-the-@89.208.169.104] has quit [Read error: Connection reset by peer]
20130218 09:55:18-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev
20130218 10:14:08-!- stikonas [~gentoo@as932.girton.cam.ac.uk] has joined #wesnoth-dev
20130218 10:14:08-!- stikonas [~gentoo@as932.girton.cam.ac.uk] has quit [Changing host]
20130218 10:14:08-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20130218 10:32:52-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 248 seconds]
20130218 10:35:55-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130218 10:46:43-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20130218 11:49:44-!- LordNasty [~NaSTy@93-43-163-150.ip92.fastwebnet.it] has joined #wesnoth-dev
20130218 12:01:25-!- mjs-de [~mjs-de@g224180097.adsl.alicedsl.de] has joined #wesnoth-dev
20130218 12:08:08-!- skyfaller_ [~skyfaller@ool-43551edd.dyn.optonline.net] has joined #wesnoth-dev
20130218 12:08:11-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit [Read error: Connection reset by peer]
20130218 12:09:51-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-dev
20130218 12:09:51-!- skyfaller_ [~skyfaller@ool-43551edd.dyn.optonline.net] has quit [Read error: Connection reset by peer]
20130218 12:21:49-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 256 seconds]
20130218 12:39:19-!- horon [~horon@nttkyo320074.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has joined #wesnoth-dev
20130218 12:39:23-!- Elvish_Pillager [~eli@dhip-029.rrw.residences.colby.edu] has joined #wesnoth-dev
20130218 12:43:24-!- molgrum [~molgrum@h-234-238.a189.priv.bahnhof.se] has quit [Read error: Connection reset by peer]
20130218 12:43:49-!- molgrum [~molgrum@h-234-238.a189.priv.bahnhof.se] has joined #wesnoth-dev
20130218 12:59:09-!- loonybot [~loonybot@158.255.133.34] has joined #wesnoth-dev
20130218 12:59:09-!- loonybot [~loonybot@158.255.133.34] has quit [Changing host]
20130218 12:59:09-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20130218 13:40:48-!- ancestral [~ancestral@75-168-48-246.mpls.qwest.net] has joined #wesnoth-dev
20130218 14:38:51-!- irker774 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20130218 14:46:34-!- ancestral [~ancestral@75-168-48-246.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20130218 14:58:15-!- horon [~horon@nttkyo320074.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has quit [Quit: Leaving...]
20130218 15:00:44-!- exciton [chuck-the-@89.208.169.104] has quit [Read error: Connection reset by peer]
20130218 15:00:59-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev
20130218 15:07:59-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 15:12:39-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20130218 15:32:11-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Ping timeout: 260 seconds]
20130218 15:48:45-!- DCW [~Thunderbi@cpc1-finc14-2-0-cust12.4-2.cable.virginmedia.com] has joined #wesnoth-dev
20130218 15:52:17-!- Appleman1234 [~Appleman1@st0801.nas931.n-yokohama.nttpc.ne.jp] has quit [Ping timeout: 248 seconds]
20130218 16:01:57-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130218 16:15:52-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 16:26:55-!- mattsc [~mattsc@d154-20-32-241.bchsia.telus.net] has joined #wesnoth-dev
20130218 16:35:34-!- zormit_ is now known as zormit
20130218 16:56:12< zormit> hi
20130218 16:56:30< zormit> i just started looking around to find an easy coding task
20130218 16:56:54< zormit> it seem the EasyCoding wiki page is partially outdated, isnt it?
20130218 16:57:13< Soliton> quite likely.
20130218 16:58:30-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20130218 17:00:46< zormit> what's with the "More powerful village naming"? is this still to be done? Euschn seems to be offline, is he still active dev?
20130218 17:02:47< Soliton> no idea. should be fairly obvious after starting a game with a random map.
20130218 17:03:21< Soliton> Euschn isn't active afaik.
20130218 17:08:47-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has left #wesnoth-dev []
20130218 17:09:24-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 17:23:13-!- _8680_ [~quassel@S010620aa4b739215.vw.shawcable.net] has joined #wesnoth-dev
20130218 17:27:19-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20130218 17:28:27-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130218 17:49:27-!- DCW [~Thunderbi@cpc1-finc14-2-0-cust12.4-2.cable.virginmedia.com] has quit [Remote host closed the connection]
20130218 17:51:14-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20130218 18:21:57-!- nafac_ [nafac@koodaus.net] has quit [Quit: Lost terminal]
20130218 18:22:15-!- nafac [nafac@koodaus.net] has joined #wesnoth-dev
20130218 18:22:33< esr> Ivanovic: ping?
20130218 18:27:00-!- nafac [nafac@koodaus.net] has quit [Client Quit]
20130218 18:27:32< esr> fabi: ping?
20130218 18:30:41-!- nafac [nafac@koodaus.net] has joined #wesnoth-dev
20130218 18:31:09-!- Netsplit *.net <-> *.split quits: Elvish_Pillager, ToBeFree
20130218 18:36:13-!- Netsplit over, joins: Elvish_Pillager, ToBeFree
20130218 18:40:57-!- Gallaecio [~quassel@84.120.114.158.dyn.user.ono.com] has joined #wesnoth-dev
20130218 18:45:20< fabi> esr: pong
20130218 18:47:11< esr> fabi: I anm most of the way througj the repo conversion.  Because the repo is larger than github's recommended 1,4GB, I'm trying to remove obsolete branches.  What is the status of fendrin_ecitor and fendrin_pathfind?  Can they be removed?
20130218 18:47:11-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20130218 18:48:19< fabi> esr: yes, they can be removed.
20130218 18:48:32< esr> fabi: Thanks.
20130218 18:49:39< fabi> esr: I can't believe that the switch to git is really happening.
20130218 18:50:09< esr> fabi: BTW, the reason I was focusing on them (and shadowmaster_filename) is that they have mid-branch deletes. This is about the only repository feature that can confuse reposurgeon. 
20130218 18:52:37< fabi> esr: I would like to have several different checkouts around without the need to have a copy of the whole data directory tree for each, can git handle that? If not, can we have different repositories for src and data?
20130218 18:54:25< Soliton> just use branches properly?
20130218 18:55:05< ChrisOelmueller> that's really the thing git does, yes
20130218 18:55:17< esr> fabi: Dunno yet.  Many other people are strongly attached to one big repository if we can have it - I think you'd have a tough fight to convince the list that's a good idea, especially since many commits touch both source and data.  I will be asking about partitioning in by next dev list post
20130218 18:55:47-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 248 seconds]
20130218 18:56:08< fabi> esr: Maybe my request is just silly, I don't know much about git.
20130218 18:56:48< fabi> Soliton: ^
20130218 18:56:49< esr> I don't think branches are a good answer here. src and data are not separate lines of development and you want to be able to (for example) grep both at once.
20130218 18:57:56< esr> The two subtrees that probably *should* turn into branches are the website and the resource branch.
20130218 18:58:13< esr> Or maybe separate repositories.
20130218 18:59:11< fabi> If I understand the concept of branching right it sounds like separate repositories is the better alternative.
20130218 18:59:47< esr> Probably, yes, now that I think about it.
20130218 19:00:05< fabi> I thought git can only be fetched as a whole, is that still right?
20130218 19:00:21< fabi> I mean you can't checkout a subtree.
20130218 19:01:37< esr> fabi: Correct.  There's a limited exception for externals.  We could have a toplevel repo with externals referencin a couple of other repositories.  This would require extra checkout steps to keep everything in sync, though.
20130218 19:02:39< esr> An external doesn't so much point at a repo as it does as a *revision* of a repo. You have to run an update operation to change that revision.
20130218 19:02:57< fabi> I see.
20130218 19:03:09< esr> Trying to use this feature may be a bad idea, leading to OoS :-)
20130218 19:07:04< fabi> The current wesnoth svn is just huge. If I would need to fetch a git repository containing everything it would be really annoying.
20130218 19:07:37< esr> I'm not actually worried about that.  git clone is amazingly fast.
20130218 19:08:07< fabi> I complain because of disk space, not performance problems.
20130218 19:08:10< Soliton> fabi: if you want to work on different features separately you make a branch for each and switch between them, just checking out what's currently needed. that saves even more space and will still be quick.
20130218 19:08:21< esr> Even on repos this size.  Turns out the reason SVN checkout is so slow is its network layer sucks.
20130218 19:12:16< esr> And about disk space - whatSoliton said.  Your space cost won't go above 2GB.
20130218 19:12:17-!- negusnyul [~negusnyul@dsl4E5CC9A5.pool.t-online.hu] has joined #wesnoth-dev
20130218 19:13:30< fabi> esr: 2GB for everything that is currently in our repository?
20130218 19:13:53-!- negusnyul is now known as prkc_
20130218 19:13:59< esr> fabi: It should be well below that actually.
20130218 19:14:53< fabi> Cool, sounds like a major improvement. I had always several copies around wasting my valuable ssd storage space.
20130218 19:15:05< esr> fabi: AI0867 reported 1.4GB.
20130218 19:16:25< esr> Oh yes, it will eat far less space than a Subversion checkout.
20130218 19:16:29< Soliton> afair one svn checkout is about the size of a git checkout plus repo.
20130218 19:16:52< Soliton> that was with git-svn with a real git repo with branches removed etc that can only get better.
20130218 19:18:17< fabi> Soliton: How will we handle our stable release branches in the new git environment?
20130218 19:20:13< Soliton> as a.. branch?
20130218 19:21:24< esr> :-)
20130218 19:21:25< Soliton> if you want a dedicated checkout for such branches there is git new-workdir or some such to help save space.
20130218 19:25:34< fabi> Soliton: Sounds nice, I thing I will like git after some time of confusion.
20130218 19:25:53< fabi> esr: Don't you need the svn repository to be frozen for the conversion?
20130218 19:36:11-!- _8680_ [~quassel@S010620aa4b739215.vw.shawcable.net] has quit [Ping timeout: 256 seconds]
20130218 19:36:11-!- mjs-de [~mjs-de@g224180097.adsl.alicedsl.de] has quit [Remote host closed the connection]
20130218 19:38:48-!- isaac [~isaac@87-198-135-206.static.ptr.magnet.ie] has quit [Read error: Connection reset by peer]
20130218 19:39:16-!- isaac [~isaac@87-198-135-206.static.ptr.magnet.ie] has joined #wesnoth-dev
20130218 19:39:39< Ivanovic> esr: what's up? (not in a status of full internet yet, the provider seems to have some problems with the line, still using 3G as temp solution)
20130218 19:40:55< esr> Ivanovic: This is a heads-up that I'm about three days from a full repo conversion, I think.
20130218 19:41:08< Ivanovic> esr: cool
20130218 19:41:16< Ivanovic> (by then i might have working internet...)
20130218 19:41:35< Ivanovic> esr: question: how do we handle the migration of devs/accounts from gna to github?
20130218 19:41:50< esr> Ivanovic: That's with all references lifted, cvs2svn and SVN artifacts cleaned out, everything rwally polushed.
20130218 19:42:54< esr> Ivanovic: I think Crab is working on pulling our other metadata - tracker state and so forth.  Committers will have to get github accounts.
20130218 19:43:35< Ivanovic> esr: i got a github account but it is not identical to my gna account name (since 'ivanovic' was already taken there...)
20130218 19:44:16< esr> Ivanovic: Yes, annoyingly esr is taken as well.
20130218 19:44:59< esr> Between namespace clashes and the disk quota, perhaps we should be looking elsewhere?
20130218 19:47:55< esr> Ivanovic: One possibility is that we simply move sideways to git on Gna.  I know there are longer-term worries about Gna's viability, but it might be intelligent to tret that as a separate problem from git migration.
20130218 19:48:58< Ivanovic> as far as i was aware there is simply no git support at gna.org
20130218 19:49:37< esr> Right, foo, I was thinking of Savannah.
20130218 19:50:33< skyfaller> I'd just like to say that I've had very good experiences with Github for the Wesnoth add-on that I've been contributing to. But I'm not a core Wesnoth developer...
20130218 19:51:27< esr> skyfaller: Do you know about the disk quota issue?  Wesnoth is bigger than they want to host?
20130218 19:51:47< Ivanovic> has someone asked them if they make an exception for wesnoth?
20130218 19:51:55< esr> skyfaller: Also, I admit to being a bit unhappy that the site is closed source.
20130218 19:52:55< skyfaller> well, I know their website says that there is no hard disk quota, they just prefer repos to be under 1GB... I'm not sure how well projects do with circumventing that rule of thumb
20130218 19:53:30< esr> skyfaller: We don't know either.  Which makes us nervous.
20130218 19:54:28< esr> I'm going tp do some things to slim down the repo but I don't think I'm going t get it below 1GB.
20130218 19:55:28< skyfaller> I know that there are plenty of projects to replicate Github in an open source fashion, like Gitorious or Gitlab, but even projects dedicated to open source are using Github for userfriendliness, e.g. OpenHatch
20130218 19:55:42< Ivanovic> esr: as already asked 5min ago: what about just asking them for a wesnoth exception?
20130218 19:56:00< esr> Ivanovic: Fine, you do it :-)
20130218 19:56:17< Ivanovic> sorry, no time these days
20130218 19:56:45 * esr never heard of gitla before.  .me googles
20130218 19:56:51< ChrisOelmueller> Ivanovic: should be aware by now that open source folks only like talking about stuff, not actually doing stuff
20130218 19:57:18< ChrisOelmueller> heck how hard can it even be to write a single mail for people discussing this on the ML for weeks now, virtually
20130218 19:57:25-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20130218 19:57:25-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20130218 19:57:25-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20130218 19:57:26< Ivanovic> ChrisOelmueller: that is why i tend to just do the stuff myself which is only possible when i have some time (not the case right now)
20130218 19:57:33< ChrisOelmueller> yeah that was totally not about you
20130218 19:57:38< mordante> servus
20130218 19:57:46< Ivanovic> hi mordante
20130218 19:57:52< mordante> hi Ivanovic 
20130218 19:57:54-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20130218 19:58:14< mordante> Ivanovic, moved all your stuff to your new apartment?
20130218 19:58:18< skyfaller> I would be happy to ask Github about an exception, but I'm just a Wesnoth user
20130218 19:58:43< ChrisOelmueller> same
20130218 19:58:51< Ivanovic> move all stuff but internet is not working via vdsl yet (using a 3g temp solution until the 'deutsche telekom' fixes their lines...)
20130218 19:59:10< skyfaller> I'm certain that e.g. esr would carry more moral authority / respect ;-)
20130218 19:59:12< mordante> esr, any news on the final size of the git repo?
20130218 20:00:14< esr> mordante: AI0867 reported that after he repacked and GCed it was 1.4GB.  I think I can slim it down some, but probably not below 1GB.
20130218 20:01:27< mordante> esr, ok I already feared that, but you sounded hopeful at the start
20130218 20:01:29-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130218 20:02:34< esr> mordante: It all depends on how big the resources branch is that and the website branch are strong candidates to be kicked out into separate repos.
20130218 20:02:45-!- _8680_ [~quassel@S0106002564d593e5.vw.shawcable.net] has joined #wesnoth-dev
20130218 20:03:08< skyfaller> I would definitely put the website into a separate repo, I don't know enough about resources to speak
20130218 20:03:10< Ivanovic> website can easily be a 2nd repo
20130218 20:03:19< Ivanovic> but the website part is tiny
20130218 20:03:30< esr> Ivanovic: I lnow.
20130218 20:03:57< esr> Ivanovic: That's why I say it all depends on the size of the resources branch.
20130218 20:04:12< Ivanovic> might be 100MB, probably less for the website
20130218 20:04:21< Ivanovic> (svn checkout size is 50MB)
20130218 20:04:35< esr> We could break out the music subtree.
20130218 20:04:47< shadowm> Le sigh.
20130218 20:05:04< mordante> music is about 100 MB not sure how bit the history would be
20130218 20:05:22< esr> shadowm: I don't want to do that, but we might be forced to it.
20130218 20:05:39< ChrisOelmueller> why?
20130218 20:05:41< esr> mordante: The history is the issue.
20130218 20:05:43< shadowm> How about you ask Github first?
20130218 20:05:57< mordante> shadowm, I already did, about to mail about it
20130218 20:06:13< mordante> they like their 1 GB limit
20130218 20:06:18< shadowm> mordante: "already"? You mean you are about to do?
20130218 20:06:29< esr> shit
20130218 20:06:43< mordante> shadowm, no I did contact them as I said I would on the dev-ml
20130218 20:06:58< skyfaller> why don't we want to make resources a separate repo? Sorry if this is a silly question
20130218 20:07:15< mordante> esr, I see another option http://sourceforge.net/apps/trac/sourceforge/wiki/Disk%20quotas
20130218 20:07:18< shadowm> skyfaller: The resources branch, or the in-game resources?
20130218 20:07:36< esr> skyfaller: resoiurces as a separate repo is an easy call, it's splittung out music that would be a PITA.
20130218 20:07:37< mordante> but once I catch up with the log of today I'll compose an email
20130218 20:07:43< skyfaller> I guess the resources branch? Isn't that what we're discussing?
20130218 20:07:43< shadowm> Yes, there's a resources branch that has got nothing to do with the resources Wesnoth uses other than serving as "source code" for a few of the largest assets.
20130218 20:09:08< skyfaller> mordante: Github doesn't have a fixed limit either, they just defined "reasonable" as "under 1 GB", I don't know what Sourceforge considers "reasonable"
20130218 20:09:42< esr> skyfaller: I don't believe SF has a quota at all.
20130218 20:09:55< mordante> skyfaller, GitHub lists the 1 GB as limit
20130218 20:10:21< skyfaller> no, it's not a hard quota, it's a rule of thumb
20130218 20:10:30< esr> I wasn't around then.  Anybody know why we *left* SF?
20130218 20:10:43< mordante> and I don't (want to) know how many GB of diskspace our binaries are
20130218 20:11:16< shadowm> Binary content?
20130218 20:11:16< mordante> esr I'm not sure whether our sources were ever hosted at SF
20130218 20:11:31< skyfaller> yes, Github doesn't want to serve up binaries anymore I understand, which I think is a silly decision but I understand why they did that
20130218 20:11:55< shadowm> https://github.com/blog/1302-goodbye-uploads
20130218 20:12:06< esr> Hmmmmm.....moving back to SourceForge might not be the worst plan in the world.
20130218 20:12:06< mordante> skyfaller, I had contact with GitHub about it, they don't seem to be happy about larger repos
20130218 20:12:48< shadowm> esr: Depends on how easily we can add hooks to the upstream repository.
20130218 20:12:58< shadowm> At the very minimum we need a mailing list hook.
20130218 20:13:11-!- halfdan- [~halfdan@thor.geekproject.eu] has joined #wesnoth-dev
20130218 20:13:14< halfdan-> hi guys
20130218 20:13:16< skyfaller> frankly I think the user-friendliness win from Github and encouraging contributors is worth a lot of pain... I would be happy to help move to Github in any way I can
20130218 20:13:16< shadowm> _That_ is how our current irker setup works, using my ML pipeline.
20130218 20:13:44< esr> shadowm: I think we can install our own git hooks on SF.
20130218 20:13:57< halfdan-> esr: why SF?
20130218 20:14:01< mordante> skyfaller, I think it is most important it works well for the _current_ contributors instead of hoping for new ones
20130218 20:14:12< halfdan-> skyfaller: what exactly is the issue with 1.4G git repo?
20130218 20:14:36< skyfaller> halfdan-: you mean aside from Github wanting repos to be under 1GB?
20130218 20:14:49< shadowm> skyfaller: Frankly, I think the user-friendliness from Github might encourage sloppy pull requests. :)
20130218 20:14:53< esr> halfdan-: Absence of disk quota, solid git support, we already own the project name there, and we can install our own git hooks.
20130218 20:15:00< halfdan-> shadowm: that is untrue
20130218 20:15:18< halfdan-> untrue as in nonsense
20130218 20:15:41< skyfaller> shadowm: It's possible, but the Wesnoth add-on I currently help out with hasn't had that problem so far
20130218 20:15:53< halfdan-> we recently moved piwik from svn to git(hub) and for one we have an increase in visibility and more contributors who know what they are doing
20130218 20:16:10< halfdan-> skyfaller: you could use gitmodules
20130218 20:16:16< shadowm> halfdan-: Well, yeah, I suppose the repository size will be a large deterrent in our case.
20130218 20:16:47< halfdan-> is the 1.4G all history or do you just have a lot of binary data?
20130218 20:17:04< mordante> both
20130218 20:17:27< skyfaller> we're currently discussing breaking out the website and the resources branch into separate repos, but I understand that would only save us about 100MB?
20130218 20:17:28< mordante> lot of history translations are also big
20130218 20:17:38< shadowm> esr: As the irker maintainer I suppose you've heard success stories from people using it upstream on SF.net's git hosting facilities?
20130218 20:17:47< shadowm> Or at least the hook component.
20130218 20:18:24< shadowm> halfdan-: Still, I would not instantly shoot down that statement as 'nonsense'. I'd say it vastly depends on your target audience.
20130218 20:18:53< shadowm> Wesnoth doesn't contain _just_ C and C++ code.
20130218 20:19:40< skyfaller> shadowm: isn't this something like arguing for security through obscurity? :P The ability to navigate an inconvenient contribution process doesn't necessarily translate into better pull requests.
20130218 20:20:30< shadowm> skyfaller: Of course not, but they are rare atm.
20130218 20:20:42< shadowm> I have had to reject a couple of sloppy patches in the past myself.
20130218 20:20:53< skyfaller> esr: what exactly would be required to break out the music? I assume we're talking about uncompressed source files?
20130218 20:21:34< esr> shadowm: Actually I have no such stories.  I see the irker activity on #conmits buy nobody bothers to tell me where it's running :-)
20130218 20:22:00< skyfaller> and would breaking out the music into a separate repo save us enough space to get under 1GB?
20130218 20:22:03< esr> skyfaller: Trivial reposugen operation - expunge on that branch 
20130218 20:22:16< esr> skyfaller: I don't know yet.
20130218 20:22:52< skyfaller> halfdan-: how would gitmodules help?
20130218 20:23:40< halfdan-> skyfaller: you can extract stuff out into other repositories and just use git submodule init/update to get to the same local clone
20130218 20:24:14-!- milki [~milki@milki.ircmylife.com] has joined #wesnoth-dev
20130218 20:24:23< skyfaller> I suppose that might be useful for e.g. the music source files?
20130218 20:24:23< milki> git?
20130218 20:25:11< shadowm> skyfaller: The music "source files" are the final Ogg Vorbis tracks used in-game.
20130218 20:25:51< skyfaller> ok, I wasn't sure what we were discussing in terms of music... the uncompressed stuff is in the resources branch then?
20130218 20:25:55< Soliton> esr: you can take breaking out the website and resource branch into their own repos as a given. i think useful discussion can commence once we know where exactly we stand then in regards to repo size.
20130218 20:26:04< shadowm> skyfaller: No music lives in the resources branch.
20130218 20:26:33< skyfaller> ... so there are no uncompressed music files anywhere?
20130218 20:26:33< shadowm> shadowm@nanacore:~$ ls src/wesnoth-resources 
20130218 20:26:33< shadowm> attacks  campaigns  cartography-tools  effects  image-localization  images  portraits  sounds-wav  story_images
20130218 20:27:51< shadowm> There are some large PNG files that are used in-game in JPEG format, as well as XCF sources of large in-game PNGs with a fair amount of story.
20130218 20:28:07< skyfaller> milki: we're discussing how to get a 1.4GB repo under 1GB, and whether Github will let it slide if we can't make it under 1GB
20130218 20:28:16< shadowm> But it isn't a lot of data.
20130218 20:28:37< halfdan-> skyfaller: I think you could just ask the Github staff about it, they respond quite fast
20130218 20:28:45< skyfaller> mordante indicated that Github was unhappy with the idea of larger repos
20130218 20:28:49< milki> skyfaller: ya, im trying to think of how projects deal with their binary resources
20130218 20:29:27< shadowm> https://pastebin.com/raZeZTbz   Given git's deltification performance so far I'd say history for these files could be even smaller than their check out size.
20130218 20:29:35< skyfaller> mordante: any chance you could pastebin your conversation with Github? I'm curious exactly what they said
20130218 20:30:07< Ivanovic> skyfaller: no idea how much we might save for branches/resources/ but for website *alone* it will be <100MB
20130218 20:30:12< Ivanovic> though it makes sense to split it out
20130218 20:31:09< halfdan-> Ivanovic: website should always be a seperate repo
20130218 20:31:45< shadowm> Sigh. I thought we already agreed that those two particular things would be separate repositories daaays ago.
20130218 20:31:46< Ivanovic> halfdan-: of course the website should not be part of the 'main' git repo at any site
20130218 20:31:56< Ivanovic> halfdan-: this is *NORMAL*
20130218 20:32:03< shadowm> And yet that seems to continue being discussed.
20130218 20:32:09< Ivanovic> halfdan-: the only question is how much this might save us of the 1.4GB
20130218 20:32:42< ChrisOelmueller> discuss for the sake of discussion, not results
20130218 20:32:52< skyfaller> Maybe we should table the conversation until we've moved resources and music and see if we're close to 1GB?
20130218 20:33:14< Ivanovic> skyfaller: wrong move
20130218 20:33:22< Ivanovic> skyfaller: why mix things?
20130218 20:33:29< Ivanovic> skyfaller: this is a *BAD* idea
20130218 20:33:44< Ivanovic> there is branches/resources, which will not be part of the git repo
20130218 20:33:51< Ivanovic> and there is trunk/data/core/music
20130218 20:34:01< Ivanovic> seperating this one out is a later option
20130218 20:34:10< shadowm> And branches/X.Y/data/core/music.
20130218 20:34:21< Ivanovic> jepp, and that
20130218 20:34:29< mordante> halfdan-, the music still has some ties in the main repository with some WML
20130218 20:34:41< Ivanovic> it has *lots* of ties
20130218 20:34:49< esr> Agreed.
20130218 20:34:53< halfdan-> did you guys try to push it once? I recently imported a large SVN repo into Git and the folder itself was about 400M, after the commit it went down to about 80M..
20130218 20:34:56< mordante> btw I send an email to the dev-ml regarding GitHub
20130218 20:35:10< halfdan-> mordante: support@github.com?
20130218 20:35:11< esr> Splitting out the music should be a last resort.
20130218 20:35:12< shadowm> At last.
20130218 20:35:14< Ivanovic> IMO the first try should be to bundle everything required to create a release bundle inside the git
20130218 20:35:31< halfdan-> s/commit/push
20130218 20:35:44< shadowm> mordante: You didn't quote their exact words from the email, though.
20130218 20:36:57< mordante> skyfaller, I'm not going to publicly post conversations with people who haven't been told the upfront
20130218 20:37:31< shadowm> I am shadowm, not skyfaller, but okay.
20130218 20:38:04< shadowm> Still doesn't seem like we learned anything new with that email.
20130218 20:39:29< skyfaller> are there any other moves that are "low-hanging fruit" for reducing repo size, once we've separated the website and resources into separate repos? If moving music out is a last resort...
20130218 20:39:37< mordante> shadowm, that reply was regarding skyfaller's request
20130218 20:39:43< halfdan-> REPEAT: did you guys try to push it once? I recently imported a large SVN repo into Git and the folder itself was about 400M, after the commit it went down to about 80M..
20130218 20:40:54< mordante> halfdan-, no we didn't but AI0867 did some optimisations on it, it was much larger before (several GB's forgot the exact number)
20130218 20:41:32< shadowm> I mentioned once that just git gc --aggressive was enough to bring down my git-svn tree from 2.4 GiB or so to 1.5. GiB.
20130218 20:41:34-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130218 20:42:25-!- ancestral [~ancestral@50-78-227-230-static.hfc.comcastbusiness.net] has joined #wesnoth-dev
20130218 20:43:32< shadowm> 2.2 GiB to 1.6 GiB.
20130218 20:47:41< Soliton> ufoai's git repo seems to be about 2.7GB and is hosted on SF.
20130218 20:48:22< halfdan-> Soliton: link?
20130218 20:49:22< skyfaller> http://sourceforge.net/p/ufoai/code/ci/7cce42fe769e1a6e406a44dfb2aa0dc7f498cb8d/tree/  ?
20130218 20:50:09< skyfaller> I don't see anywhere listing the size of the repo
20130218 20:50:35< Soliton> halfdan-: over there on my hard drive.
20130218 20:56:48< mordante> the GitHub communication for the developers interested http://forum.wesnoth.org/viewtopic.php?f=16&t=38361
20130218 20:58:15-!- ancestral [~ancestral@50-78-227-230-static.hfc.comcastbusiness.net] has quit [Quit: i go nstuf kthxbai]
20130218 20:58:42< halfdan-> "The board requires you to be registered and logged in to view this forum." - seriously?
20130218 20:58:50< mordante> halfdan-, yes
20130218 20:59:12< mordante> and it requires you to be a developer
20130218 20:59:29< skyfaller> I'm registered and logged in and not authorized to view the forum, b/c I am not a developer :P
20130218 20:59:55< halfdan-> mordante: sorry, but that's stupid if you want people to discuss this topic
20130218 20:59:55< mordante> yes as said before I don't want to make the private conversation public
20130218 21:00:39< skyfaller> I understand, it just makes it difficult for us non-developers to discuss the issue or help out
20130218 21:01:12< mordante> why I just tells you what I wrote in the public mailing list
20130218 21:01:35< mordante> https://mail.gna.org/public/wesnoth-dev/2013-02/msg00078.html
20130218 21:01:36< Soliton> there is not a lot more to discuss in regards to github until we know at what repo size we'll end up.
20130218 21:03:44< halfdan-> noone seems to be willing to "just try it"
20130218 21:03:53< skyfaller> Well, I would be happy to help in any way I can. Sadly I am a git noob and not a Wesnoth developer, so I might be useless.
20130218 21:04:06< timotei> halfdan-: probably because you should research before doing such thing? I mean, maybe we'll get banned or so :P
20130218 21:04:29< timotei> Plus, it's a nice thing to tell GitHub before breaking their limit ;)
20130218 21:04:30< Soliton> the repo isn't ready yet...
20130218 21:04:34< halfdan-> timotei: you can try it locally and just push / clone to and from another folder
20130218 21:05:00< Soliton> and you can repack. all already done.
20130218 21:05:10< skyfaller> I think people have been saying that AI0867 already did that, but he is not here
20130218 21:05:25< halfdan-> well it would be nice to know what exactly he tried
20130218 21:05:40< halfdan-> there are lots of branches hanging around that probably can be removed
20130218 21:05:43< skyfaller> maybe if we say his name enough times he will appear ;-)
20130218 21:06:12< Soliton> and esr is dealing with cleaning up the repo which includes removing unnecessary branches.
20130218 21:06:44-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has joined #wesnoth-dev
20130218 21:07:13< esr> Soliton: I am, but I'm highly doutbtful that will pare off 400MB of stuff.
20130218 21:07:19< halfdan-> then there is no need to discuss this any further at this point. just wait for the people to finish their research
20130218 21:07:48< halfdan-> esr: can you just try and git clone /path/to/repo
20130218 21:08:01< halfdan-> only clone the master branch
20130218 21:08:37< Soliton> esr: *shrug* we'll know when you're done.
20130218 21:10:40< Soliton> anyway, there are alternatives... another being setting up our own git hosting, especially if we're also going to setup our own tracker/ticket system.
20130218 21:11:32< esr> Soliton: Not a *crazy* ideam but what recommends it over (say) SourceForge?
20130218 21:12:30< Soliton> independence, better integration with said tracker, i'd think.
20130218 21:12:55< Soliton> but SF is a very good alternative indeed.
20130218 21:13:07< halfdan-> you guys could use GitlabHQ
20130218 21:13:17< halfdan-> which is a self-hosted github
20130218 21:13:23< skyfaller> yes, I mentioned that earlier
20130218 21:13:32< halfdan-> skyfaller: ok, didn't see it
20130218 21:15:01< skyfaller> is GitLab better than Gitorious? I've never seriously used either, GitLab does look prettier
20130218 21:15:37< skyfaller> Personally I prefer Github for ease of use and network effects, but it would be nice to have something that is completely FOSS, of course
20130218 21:15:49< halfdan-> skyfaller: same here
20130218 21:16:17< halfdan-> not sure what you mean by better, Gitorious is just an alternative to Github AFAIK, while you can install Gitlab on your own server
20130218 21:16:42< halfdan-> functionality-wise I think that gitlab made some really good improvents in the last few months
20130218 21:16:53-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130218 21:16:59< halfdan-> gitlab will add group support in the next release IIRC
20130218 21:17:04< skyfaller> well, Gitorious is Affero GPL, it is open source, so you can install it yourself if you want
20130218 21:17:13< ChrisOelmueller> (you don't want)
20130218 21:17:19< skyfaller> hahaha
20130218 21:17:20< esr> I am mildly opposed to self-hosting, on the grounds that I like having someone else do the grunt sysadminning. But only mildly - if someone here thinks that's their holy mission I won't stop 'em.
20130218 21:19:18 * Soliton agrees.
20130218 21:20:28< ChrisOelmueller> i'm gonna leave this discussion for a few days, with the final remark that picking sourceforge as platform now would be missing a huge opportunity and i hope you're not walking into that
20130218 21:20:48-!- ChrisOelmueller [~eoc@2a01:238:42b6:4400:f140:fff4:e413:e393] has left #wesnoth-dev ["Where's those banana cream pies, eh?  He dangles two pillows and watches her."]
20130218 21:20:58< halfdan-> can't agree more
20130218 21:21:30< esr> Um, what opportunity is that?
20130218 21:22:07< skyfaller> He's probably referring to the opportunity to move to Github and gain ease of use and network effects.
20130218 21:22:57< halfdan-> correct
20130218 21:23:11< esr> Maybe.  SourceForge is still the 800-pound gorilla of network effects, though, even if it's kinda old school in some ways.
20130218 21:24:23< shadowm> (And occasionally screws up.)
20130218 21:24:29< skyfaller> My feeling is that Github encourages collaboration in a qualitatively different way from any other existing option
20130218 21:25:15< esr> The new hotness, yes.
20130218 21:25:24< skyfaller> nobody else is investing as much time and energy into that, and sadly alternatives like Gitlab are playing catchup with significantly less resources
20130218 21:26:18-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130218 21:28:50-!- Octalot [~noct@host86-148-68-137.range86-148.btcentralplus.com] has joined #wesnoth-dev
20130218 21:29:47< skyfaller> personally I think it is vitally important to encourage users to engage with and contribute to the software they use, and I think Github is currently the best way to do that, it helps fulfill the promise of open source
20130218 21:30:04< esr> skyfaller: It's closed source.
20130218 21:30:35< shadowm> skyfaller: Personally, I think we do a piss poor job at that and it's not only because of the obsolete and unmaintained forge we use.
20130218 21:31:07< halfdan-> esr: what is closed source?
20130218 21:31:13< skyfaller> Github is closed source but it is currently the easiest way for humans to contribute.
20130218 21:31:32< skyfaller> shadowm: what else is the problem besides the forge?
20130218 21:31:42< esr> I don't trust closed source for the long term.  Ever.
20130218 21:32:07< shadowm> skyfaller: Our website and wiki structure doesn't lend itself to attract mainline coders, only user-made content creators.
20130218 21:32:16< halfdan-> k, and open source would make github better in what way?
20130218 21:32:31< shadowm> IMHO there's also not enough interaction between the development team and the community at large.
20130218 21:32:42< skyfaller> yes, I must admit it reminds me a little bit of the Bitkeeper controversy :P
20130218 21:33:16< halfdan-> skyfaller: that's a completely different thing, they decided to commercialise the complete system 
20130218 21:33:36< esr> halfdan-: Sustainability.  When a closed-source product EOLs uou're fucked.  Auditability.  Customizability. Transparency of process.  I could go on...
20130218 21:34:08< skyfaller> shadowm: I guess maybe we should consider how the project could be made more open, aside from the service hosting the source code itself
20130218 21:34:13< halfdan-> esr: well the great thing with git is that you always have your project backed up
20130218 21:34:14< shadowm> I personally don't have a concern with Github's closed-sourcedness since we aren't going to use their services aside from repository hosting anyway, and since Git is a DVCS, well.
20130218 21:34:29< shadowm> I don't think I need to preach the advantages of DVCSes in that regard.
20130218 21:34:40< shadowm> Not to _this_ audience, anyway.
20130218 21:34:45< esr> shadowm: Yeah, but there's other project metadata.
20130218 21:34:47< halfdan-> and if github goes down one day you can just fire up a gitlab installation
20130218 21:35:18< shadowm> skyfaller: I've been thinking about the issue for years and I haven't yet found a solution aside from a better website structure and documentation in the wiki.
20130218 21:35:20< halfdan-> err btw, is sourceforge open source?
20130218 21:35:32< shadowm> skyfaller: Which are problems that need to be solved step-by-step.
20130218 21:35:39< esr> halfdan-: Yes. Look up Allura
20130218 21:36:02< shadowm> I kind of started to work on that, but haven't had time to progress further.
20130218 21:36:11< shadowm> e.g. Compare http://wiki.wesnoth.org/PatchSubmissionGuidelines with http://wiki.wesnoth.org/User:Shadowmaster/PatchSubmissionGuidelines
20130218 21:36:23< skyfaller> yeah, SourceForge released its source in June 2012 according to Wikipedia
20130218 21:36:26< halfdan-> esr: k, what's the big difference with gitlab then? if SF were to suddenly go down you'd still've lost all metadata..
20130218 21:37:42< halfdan-> and just as with github where you can fire up a gitlab installation you could do the same with allura when sf goes down. sure github != gitlab in some ways, but it comes close
20130218 21:37:45< esr> I don't have a position against gitlab, except that we'd have to self-host it.
20130218 21:38:40< mordante> I'm off night
20130218 21:39:40-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20130218 21:40:35-!- Crendgrim_ is now known as Crendgrim
20130218 21:41:08< esr> I agree that DVCSes reduce the risk of a Bitkeeper-like fiasco, but when you add the fact that they *don't want us*...
20130218 21:41:26< shadowm> esr: Did you read the post in the forums already by the way?
20130218 21:42:01< shadowm> Hm, of course you did not, because mordante mentioned he was going to post it in another channel.
20130218 21:42:21< esr> No, I don't seem to be rehistered on that forum.  I sent a password reminder request and nothing has happened yet.
20130218 21:42:26< shadowm> esr: http://forums.wesnoth.org/viewtopic.php?f=16&t=38361#p548059 (And no, you are not authorized to repaste that elsewhere. I'm speaking as the law here.)
20130218 21:42:33< shadowm> esr: What, forums.wesnoth.org?
20130218 21:42:37< shadowm> You _are_ registered, don't be silly.
20130218 21:42:47-!- scientes [~scientes@unaffiliated/scientes] has joined #wesnoth-dev
20130218 21:43:23< esr> Ah, OK, this time my password autocompletw worked.  Looking...
20130218 21:43:52< esr> OK, I've read it.
20130218 21:47:08< esr> shadowm: sounds to me like they're actually worried about bandwidth hogs ("user experience").  If that's the problem, cutting our repo into chunks is in no way a sure path to solving it.
20130218 21:48:12< shadowm> TBH any Git repository above 10 MiB ruins my user experience, at the moment of cloning. ;)
20130218 21:48:28< shadowm> Same goes with every other "thing" I cannot wget -c.
20130218 21:49:01< esr> Fair point.
20130218 22:05:47-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 256 seconds]
20130218 22:07:15-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130218 22:09:53-!- scientes [~scientes@unaffiliated/scientes] has quit [Remote host closed the connection]
20130218 22:12:33-!- _8680_ [~quassel@S0106002564d593e5.vw.shawcable.net] has quit [Ping timeout: 248 seconds]
20130218 22:18:32-!- prkc_ [~negusnyul@dsl4E5CC9A5.pool.t-online.hu] has quit [Quit: Konversation terminated!]
20130218 22:24:20-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130218 22:26:07< shadowm> anonymissimus: A real concern I have with unstore_unit.check_passability's default of yes is that it leads to this kind of misleading messages when storing and unstoring units who for whatever reason are not allowed to move on the hexes they've got assigned to them:
20130218 22:26:14< shadowm> 20130218 04:57:20 error engine: Trying to add  - Leech-1056 at an invalid location; Discarding.
20130218 22:26:51-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 276 seconds]
20130218 22:27:12< shadowm> It took me ages to finally understand that I was getting these errors not because there was something wrong with the stored units, but simply because they cannot move on the locations where they are originally created (these units are intended to be completely immobile).
20130218 22:27:45< shadowm> And it was all because I was unaware of this unstore_unit.check_passability = yes default. Using check_passability=no in the [unstore_unit] sequence gets rid of that problem.
20130218 22:27:50< shadowm> I'm on 1.10, by the way.
20130218 22:29:48< shadowm> The method originating the error message is unit_map::insert(), but maybe once control is returned to the caller, it could determine why the insertion failed and then warn us that check_passability's current value might need to be adjusted?
20130218 22:31:00< shadowm> To make things worse, this unstore_unit sequence succeeded a [replace_map] operation, so I spent a long time believing it was a bad interaction with [replace_map] (due to varying original and new map dimensions) when that was never the issue in the first place.
20130218 22:31:07-!- LordNasty [~NaSTy@93-43-163-150.ip92.fastwebnet.it] has quit [Read error: Connection reset by peer]
20130218 22:31:15< shadowm> My expression when I finally realized the cause: :/
20130218 22:36:43-!- Upthorn [~ogmar@69.62.144.56] has quit [Ping timeout: 246 seconds]
20130218 22:37:15-!- _8680_ [~quassel@S010620aa4b739215.vw.shawcable.net] has joined #wesnoth-dev
20130218 23:06:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 244 seconds]
20130218 23:34:16-!- csarmi [~csarmi@195-38-105-97.pool.digikabel.hu] has joined #wesnoth-dev
20130218 23:37:58-!- Gambit [~gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20130218 23:54:46-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Remote host closed the connection]
20130218 23:56:16-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
--- Log closed Tue Feb 19 00:00:16 2013