Monday, 2010-12-06

*** phurl has quit IRC00:02
*** erlehmann has quit IRC00:08
*** kreynen_ has quit IRC00:54
*** bassel has quit IRC01:03
*** bassel has joined #cc02:07
*** mrb__ has quit IRC02:36
*** mrb__ has joined #cc02:37
*** mrb__ has quit IRC02:42
*** kreynen has joined #cc03:50
*** NameLess-js has joined #cc03:51
NameLess-jshi i got question03:52
NameLess-jswhich license is best for source codes?03:52
greg-gNameLess-js: I would recommend either a GPL/AGPL or Apache/MIT depending on the style of license you want (copyleft or permissive)03:56
NameLess-jsi have no clue about licenses :)03:57
greg-gNameLess-js: *none* of the CC licenses are for source code, however. They were not designed/written for the world of software03:57
NameLess-jsso i should use spl03:57
NameLess-jsand how do i state it?03:58
NameLess-js'This work is licensed under the GPL License Agreement'?03:58
NameLess-jsgreg-g: the statement seems to be quite long, how can i just use i sentence?04:01
greg-git tells you how you should do it04:02
NameLess-jsi don't c04:03
greg-gNameLess-js: just follow the instructions in that guide. I'm sorry, I really need to go to bed now04:07
NameLess-jsok, sorry to disturb :)04:08
greg-gno problem. have a good <whatever time of day it is for you> :)04:08
*** NameLess-js has quit IRC04:17
*** wormsxulla has quit IRC04:21
*** wormsxulla has joined #cc04:35
*** mrb has joined #cc04:38
*** kreynen has quit IRC05:26
*** phurl has joined #cc07:04
*** pmiller has joined #cc08:44
*** pmiller has left #cc09:06
*** erlehmann has joined #cc09:11
*** Kaetemi has quit IRC09:22
*** erlehmann has quit IRC09:31
*** erlehmann has joined #cc09:31
*** JED3 has quit IRC09:56
*** erlehmann has quit IRC10:02
*** erlehmann has joined #cc10:02
*** bassel has quit IRC11:12
*** bassel has joined #cc11:12
*** mrb__ has joined #cc11:12
*** mrb has quit IRC11:15
*** erlehmann has quit IRC11:35
*** akila87 has joined #cc13:11
*** kreynen has joined #cc13:14
*** karlcow has quit IRC13:25
*** karlcow has joined #cc13:26
*** bassel has quit IRC13:34
*** erlehmann has joined #cc13:36
*** akila871 has joined #cc13:36
*** akila871 has left #cc13:36
*** erlehmann has quit IRC13:38
*** akila87 has quit IRC13:38
*** bassel has joined #cc13:45
*** erlehmann has joined #cc13:57
*** nkinkade has joined #cc14:44
*** akozak has joined #cc16:55
*** erlehmann has quit IRC17:03
*** wormsxulla has quit IRC17:09
*** wormsxulla has joined #cc17:24
*** erlehmann has joined #cc17:44
*** JED3 has joined #cc18:03
*** JED3 has quit IRC18:10
*** JED3 has joined #cc18:46
JED3ping nyergler19:26
nyerglerJED3: pong19:27
* paroneayea puts himself in the ping queue after JED3 ;)19:27
JED3re: you specify that morePermURL's must not contain more than 1 reference, does that adhere to whats spec.d in CC+?19:28
nyergler"spec.d in CC+" is a hilarious phrase ;)19:29
JED3hehe yeah looking at it now19:30
nyerglerit's not specified right now19:30
nyergleri don't know of anyone doing more than one at the moment (and don't know what we'd even do with it)19:30
JED3i thought it was more explicit :)19:30
nyerglerthat'd be great if it were, but it's not at the moment.19:30
nyergleri suppose you can debate whether or not it makes sense to include that predicate in the list that triggers no-parse19:30
nyerglerer, no-mark19:31
nyerglerdoes that answer the question? maybe omit it from the list in 740 at the moment?19:31
nyergleri'm happy to annotate the ticket as such19:31
JED3yes that answers the question, I just wanted to make sure it wasn't an oversight19:31
nyerglerparoneayea: your turn19:32
paroneayeanyergler: cool, so19:32
paroneayealooking at how to do the transifex client in cc.i18n19:32
paroneayeait's kind of annoying......19:32
*** erlehmann has quit IRC19:32
paroneayealooking at the egg installed by buildout19:33
nyerglerthe tx egg?19:33
paroneayeait doesn't look like it has an entry point to the "tx" script19:33
paroneayeathe transifex-client egg19:33
paroneayeawhich is a script19:33
nyerglerso i wonder if they're using the old scripts syntax in setup.py19:34
nyergleruh, does scripts/tx just call into the tx package?19:34
paroneayeareload(sys) # WTF? Otherwise setdefaultencoding doesn't work19:35
nyerglerof COURSE setdefaultencoding doesn't work! it's deleted by site.py19:35
* nyergler looks in the python docs so he can be pedantic19:36
nyerglerThis function is only intended to be used by the site module implementation and, where needed, by sitecustomize. Once used by the site module, it is removed from the sys module’s namespace.19:36
paroneayeayeah I know :)19:36
nyerglerright, sorry, i'm over my mini-tantrum19:36
paroneayeaalthough, that looks like maybe it's some way to set setdefaultencoding after deletes it19:36
paroneayeadidn't know that trick19:37
nyergleryeah, it's pretty evil though -- i'm betting it wouldn't be thread safe19:37
paroneayeayes, pretty evil19:37
paroneayeaguessing their code doesn't explicitly decode to utf-8 everywhere19:38
nyerglerone second, i want to find the source tree for the tx client, see where that script comes from19:38
paroneayeasounds like other codebases I've seen somewhere! ;)19:38
nyergler(i suppose it makes sense that they need to do different encodings since they're doing translation)19:38
nyerglerthe evil part? ;)19:38
nyerglerparoneayea: what's the pypi name for the client?19:39
nyerglerparoneayea: ugh19:42
nyerglerthat is totally disgusting19:42
nyerglernot totally, sorry, i saw Burlesque this weekend, brought out the drama queen in me19:42
nyerglerCher does that19:42
nyerglerit does look like they're using the old syntax for the script, and keeping the code outside the package19:43
paroneayeanyergler: well that's easy enough to fix.  I could write a patch for it and see if they respond within the week19:43
nyerglerparoneayea: yeah, i think the patch does the following:19:44
nyergler1) moves the main function into the txclient package (maybe a module?)19:44
nyergler2) keeps tx in the tree but just calls into the package in __main__19:44
nyergler3) defines the console-script entry point19:44
nyerglerso that has both definitions19:44
paroneayeasounds good, and pretty simple19:45
nyerglerright, 30 minutes?19:45
nyergleri'd just for on bitbucket, then make a merge request19:45
paroneayea+ time for me to stumble my way through hg, at least ;)19:45
nyerglermake it super easy for them, and in the meantime use our forked repo to make an egg19:45
CC_HudsonYippie, build fixed!19:53
CC_HudsonProject deedscraper build (62): FIXED in 2 min 39 sec:
CC_HudsonJED3: Forcing metadata uniqueness before displaying info19:53
*** Kaetemi has joined #cc20:11
*** bassel has quit IRC21:30
nkinkadeparoneayea: So I think I've got the tx CLI worked out for now.21:46
paroneayeankinkade: oh yeah?21:46
nkinkadeBasically, I don't thing there is anything you need to do.21:46
paroneayeawunderbar news21:47
paroneayeankinkade: so by you got it worked out21:48
paroneayeawhat did you do / have you done?21:48
nkinkadeparoneayea: The only thing we'll need to coordinate is how we push new stuff up to
*** alecstory has left #cc21:50
paroneayeadid you put it on a cronjob somewhere?21:51
nkinkadeparoneayea: I haven't set up a cron job yet.21:52
nkinkadeIt's on a7 /home/cronuser/transifex.net_i18n_checkout21:52
nkinkadeThat's a clone of git/i18n, and there is a tx CLI config file that seems pull from just fine.21:53
nkinkadeI haven't tried a push.21:53
nkinkadeparoneayea: Really, there is no reason that you have to use that clone to do anything.21:54
nkinkadeI suppose you're free to push new strings to from your own checkout somewhere else.21:54
*** bassel has joined #cc21:55
nkinkadeparoneayea: I suppose the best bet for us will be to use that single clone if we can, just for the sake of consistency.21:55
paroneayeankinkade: I was wondering how you set it up when we couldn't get tx as a script to work with buildout yet21:55
nkinkadeAnd to pull from via cron job, but to push manually.21:55
paroneayeayou just easy_installed to the global site-packages ;)21:56
nkinkadeparoneayea: Why do you need buildout to work with tx?21:56
paroneayeankinkade: well theoretically we want the scripts that work with our packages to be accessible from a self-contained buildout...21:56
nkinkadeThis clone of i18n will be used to pull down translations from and push them to our repo.21:56
nkinkadeSo can't your buildout process just continue to work directly from our git repo and forget about
nkinkadeparoneayea: Am I wrong?21:57
nkinkadeI feel like nothing will need to change at all about how buildout works with regard to i18n.21:57
paroneayeankinkade: I mean, the transifex-client package21:57
paroneayeaI just meant the tx script21:58
paroneayeait's not a big deal21:58
nkinkadeWhat about it?21:58
paroneayeawell normally we'd access it from21:58
paroneayeais how we normally do things21:58
paroneayea(anyway, that's why I made those patches to transifex-client)21:59
nkinkadeparoneayea: Okay, that's fine and I understand that, but I'm confused why you're wanting/needing to involve tx in the buildout at all?  You mentioned on #transifex that you wrote some patches even.21:59
nkinkadeAs far as I know, your processes go on just like before, not needing to know anything about or the tx CLI.21:59
paroneayeabecause presumably the tx script is populated via buildout21:59
paroneayeaI didn't mean the buildout that's set up currently22:00
paroneayeaI just meant22:00
paroneayeainstead of having the tx cli available globally ohn the system, it would be part of the scripts buildout sets up in bin/22:00
paroneayeaso that's why I made the patches22:00
paroneayeabecause their setup didn't allow for it22:00
nkinkadeparoneayea: But my question is why you'd even want to include in i18n/bin in the first place?22:01
paroneayeankinkade: well because I assumed that's how we'd do it, because normally we want a package and all its tools included in the buildout22:01
paroneayeankinkade: but anyway, it doesn't have to happen that way22:01
paroneayeadoing it system-wide is fine22:01
nkinkadeparoneayea: I think we've got a disconnect here.22:02
paroneayeahaha, probably22:02
nkinkadeLet me clarify, maybe.  In what way does your buildout script need to have any awareness of Transifex, whether it be, the CLI, or otherwise?22:03
paroneayeaI'm having a hard time explaining this in a way that's not what I said22:03
paroneayeankinkade: because I assumed that we *wanted* it that way.  It's not critical to the package, but normally we package tools *like* tx with the buildout22:03
paroneayeait keeps things self-contained to a buildout22:03
paroneayeabut it's not *necessary* if you're okay with doing system packages, but as a rule in general we tend to avoid system packages for python things22:04
paroneayeaI think if I say anything else I'm going to run back in a circle22:04
paroneayea(I already have I think)22:04
paroneayeaanyway it doesn't hurt to help them have those packages, so they can use the more modern setup of using entry point based scripts22:05
paroneayeaer, those patches, not packages22:05
paroneayeathat's all I have to say.22:05
nkinkadeOkay, so your buildout won't change at all and nothing in the i18n package will communicate with or care at all about Transifex, right?22:05
paroneayeankinkade: for now :)22:05
paroneayeaeventually I might make tx available in bin/22:05
paroneayeain case they accept my patches and also it ends up being useful for me to be able to push things from my end22:06
paroneayeapush new translations I mean22:06
nkinkadeOkay, cool.22:06
paroneayeayay, conversation resolved ;)22:06
nkinkadeI can see the utility of being able to push new strings from any checkout.22:06
akozakwhat the heck is hal doing at the beginning of this video tility of being able to push new strings from any checkout.22:23
paroneayeaakozak: you sound like a markov bot22:24
akozakodd paste problem22:24
akozakno really though, the beginning of that video is odd22:24
nyerglerit was odd to be there, too22:32
*** erlehmann has joined #cc22:34
*** kreynen has quit IRC22:40
*** luisv has joined #cc22:41
paroneayeayo luisv22:44
luisvAm I making this up, or did some of the CC 3 licenses have "also compatible with reciprocal licenses" clauses?22:44
paroneayealuisv: I'm not sure what you mean by that22:47
luisvI may be making it up :)22:48
* paroneayea summons greg-g 22:48
luisvbut I seem to recall that one of the SA licenses, at one point (maybe 2.0 and removed in 3.0?) had language that said something to the effect of "you can use this content under the CC license it was published under, or under other licenses if they also allow compatibility with CC"22:48
paroneayealuisv: do you mean 4b?22:50
paroneayeaI think it's standard copyleft in that the end-result derivative work ends up having to be able to be by-sa 3.0 basically22:51
luisvI mean22:52
luisvand 1(c)22:52
luisvwhich I was not finding in my copy of CC-SA22:52
luisvbecause I was looking at CC-BY :(22:52
akozakluisv, yea it only appears in by-sa22:52
akozakalso see 1.c.22:52
luisvperfect, thanks.22:53
paroneayeaI like 4b iv's phrasing of Creative Commons Compatible License22:53
paroneayeait sounds like22:53
paroneayeathe CC Compatible License, 3.022:53
luisvthat could be confusing22:53
* paroneayea shouldn't spread confusions probably :D22:54
* luisv idly wonders if MPL would qualify22:54
paroneayealuisv: MPL is still a per-file copyleft isn't it?22:55
paroneayeathat'd be interesting but it sounds a lot to me like that recent conversation on how-feasible-would-it-be-to-make-cc-by-sa-and-gpl-compatible22:55
paroneayea(that's a lot of hyphens)22:55
*** bassel has quit IRC22:56
paroneayeawhich bkuhn seems to think at least that there's no way to do that with two copylefts without explicitly stating "these two copylefts are compatible, I have decreed it" in the license22:56
* paroneayea looks up that conversation again22:57
paroneayealuisv: if you haven't read it, waste some time productively: :)22:58
paroneayeamonster thread22:58
luisvI had started it ages ago23:00
luisvdid not realize how long it got23:00
paroneayeathere we go23:00
luisvyes, useful, thanks23:01
paroneayealuisv: actually GPL and MPL compatibility come up in there23:02
paroneayeaMPL comes up a few times, text search for it!23:03
luisvI had seen some of those23:04
luisvhadn't realized they were all technically part of the same thread23:05
luisvbrad lost me when he claimed it would be trivial to state MPL as a set of exceptions to GPL23:05
luisvI don't think anyone at FSF realizes that LGPL 3 is unreadable23:05
paroneayeabest part of that whole thing is how it started as a joke23:05
paroneayealuisv: lol23:05
nyergler(btw, iirc the compatibility in 3.0 requires us to evaluate another license and say, "yes, this is compatible")23:05
luisv(and of course, he's not the only one to say 'just write it as GPL exceptions- it'll be easy')23:06
luisv(somehow none of those people has gone and done it)23:06
luisvnyergler: yes23:06
luisvnyergler: which no one has done23:06
nyergler(sorry if that came up already, i didn't read the entire scrollback)23:06
nyerglernkinkade: is upgrading WP on labs on your task list?23:06
nkinkadenyergler: Funny you should ask.  I just logged into a7 to upgrade labs not 30 seconds ago.23:07
nkinkadeWeird timing.23:07
luisv"cross-copyleft compatibility is area of active research by best minds in # thought;Answer's unknown,but clearly non-trivial"23:07
luisvnyergler: the passage was referred to23:07
paroneayeaactive research by world's leading copyleft scientists23:07
* luisv puts on white labcoat, pulls out flasks full of bubbling liquids23:08
luisvnyergler: some folks at OSI have suggested we put similar clauses in MPL 223:08
paroneayea     ////23:08
paroneayea    /o-o-\  - We need to keep the23:08
paroneayea    \ =  /    accelerator at full capacity,23:08
paroneayea    /|-/| \   or else we'll all perish in23:08
luisvnyergler: but the face that no one has taken advantage of it in the CC case makes me leery23:08
paroneayea   |\| |/| |  the resulting fallout!23:08
nyerglerluisv: don't cross the streams23:09
nyerglerluisv: i suppose the thought experiment is, has no one done it for BY-SA because we have a dominant position in non-software public licenses?23:10
nyergleror because no one cares in any context?23:10
nyergler(ie, would people be more anxious with software since there's more competition)23:10
luisvnyergler: I think my intuition is the latter, but if anyone in your org thinks it is the former, or has any other intuitions on the matter, I'd love to hear it23:10
paroneayeawell the foss gaming community cares, but nobody cares about the foss gaming community :(23:12
* paroneayea stares grumpily at the schrodinger's box of game code/data23:12
nyerglerluisv: just asked mlinksva23:13
nyerglerhe pointed out the motivation behind that clause was getting to GFDL compatibility23:13
nyerglerwhich we got in other ways23:13
greg-gluisv: I assume the reason I was summoned before is clear now (the reciprocal license thing)23:13
luisvgreg-g: yeah23:14
luisvgreg-g: thanks anyway :)23:14
luisvnyergler: ah, indeed23:15
paroneayeagreg-g: I also summoned you to test my latest summoning spell23:15
paroneayeathat you did not appear as a half-toad is evidence of its success23:15
paroneayea   \'.23:15
paroneayea    \ '.23:15
paroneayea     \  \23:15
paroneayea     _L__L_23:15
paroneayea     //~ ~23:16
paroneayea     G   >23:16
paroneayea     / \\\\-------.__  * * * *23:16
paroneayea    / ' \\\______ |T  * MAGIC *23:16
paroneayea   |   \ \\      \|    * * * *23:16
* paroneayea thinks it is soon time to stop working on CC stuff and time to start working on making a snowman23:16
greg-gparoneayea: snowman! jealous! we barely got a dusting here today23:16
greg-gparoneayea: and thanks for not making me a half-toad23:17
paroneayeano problem23:17
* luisv finally groks the Collection/Adaptation distinction23:17
*** knknk has joined #cc23:23
*** erlehmann has quit IRC23:28
*** Kaetemi has quit IRC23:56

Generated by 2.6 by Marius Gedminas - find it at!