| *** jinty has quit IRC | 01:29 | |
| *** filip101 has joined #schooltool | 01:57 | |
| *** wrobel has quit IRC | 03:25 | |
| *** lisppaste5 has quit IRC | 04:55 | |
| *** lisppaste5 has joined #schooltool | 05:03 | |
| *** ignas has quit IRC | 05:23 | |
| *** alga has quit IRC | 05:54 | |
| *** th1a has joined #schooltool | 06:09 | |
| Lumiere | hi th1a | 06:12 |
|---|---|---|
| th1a | Lumiere: Hi. | 06:12 |
| th1a | How's it going? | 06:12 |
| Lumiere | not too bad | 06:13 |
| Lumiere | I've got your developers busy | 06:13 |
| Lumiere | ^^ | 06:13 |
| Lumiere | pcardune and I put together an ftest for security on resources | 06:14 |
| Lumiere | it's not complete quite yet, but we'll clean it up over the next week as we get to it | 06:14 |
| Lumiere | and I spent a little bit cleaning up cando's current set of specs | 06:15 |
| Lumiere | wrote the event-ui spec for schooltool too | 06:15 |
| Lumiere | (there's really nothing like looking at the blueprints front page of launchpad and seeing the top of both the registered and completed lists being your specs | 06:16 |
| *** pcardune_away has quit IRC | 06:30 | |
| *** pcardune_away has joined #schooltool | 06:40 | |
| Lumiere | bbl | 06:45 |
| Lumiere | like... tomorrow | 06:45 |
| *** pcardune_away is now known as pcardune | 06:54 | |
| *** lisppaste5 has quit IRC | 07:33 | |
| *** lisppaste5 has joined #schooltool | 07:44 | |
| *** pcardune is now known as pcardune_sleep | 07:47 | |
| *** pcardune_sleep has quit IRC | 07:59 | |
| *** jinty has joined #schooltool | 08:57 | |
| jinty | hoi Lumiere | 08:58 |
| *** wrobel has joined #schooltool | 09:02 | |
| *** alga has joined #SchoolTool | 09:37 | |
| *** alga has quit IRC | 10:03 | |
| *** thisfred has joined #schooltool | 10:58 | |
| *** thisfre1 has joined #schooltool | 11:37 | |
| *** thisfre1 has left #schooltool | 11:37 | |
| *** ignas has joined #schooltool | 11:40 | |
| *** jfroche has joined #schooltool | 12:54 | |
| *** ignas has quit IRC | 12:56 | |
| *** JohnnyST has quit IRC | 14:00 | |
| *** Fujitsu has quit IRC | 14:07 | |
| *** ignas has joined #schooltool | 14:22 | |
| jfroche | ignas: thanks for your reply to my commits | 14:31 |
| jfroche | ignas: i get a UnpickleableError: Cannot pickle <type 'zope.security._proxy._Proxy'> objects | 15:07 |
| jfroche | any idea ? | 15:07 |
| ignas | removeSecurityProxy for objects that will be stored in the database | 15:07 |
| jfroche | ok | 15:08 |
| jfroche | where does that come from ? | 15:08 |
| ignas | wht? | 15:09 |
| jfroche | i try to understand why | 15:10 |
| ignas | oh - well you can't put security wrapped objects in database | 15:19 |
| ignas | so if you get something out of let's say person container | 15:19 |
| ignas | you should removeSecurityProxy from it before storing it in for example person.advisor attribute of another person | 15:19 |
| *** pcardune_sleep has joined #schooltool | 15:30 | |
| *** pcardune_sleep has quit IRC | 15:47 | |
| * th1a shuffles some papers around. | 16:27 | |
| ignas | hi | 16:27 |
| th1a | Hi ignas, jfroche, jinty, Lumiere... | 16:27 |
| jinty | hi | 16:28 |
| th1a | OK... first off, should we switch to 1330 UTC next week. | 16:29 |
| th1a | I don't need the incentive to sleep in. | 16:29 |
| jfroche | hello th1a | 16:29 |
| th1a | Hi jfroche. | 16:29 |
| th1a | Any objections to 1330? | 16:31 |
| ignas | nope | 16:31 |
| jinty | nope | 16:31 |
| jfroche | no problem | 16:32 |
| th1a | OK. Who knows how to switch the IRC header thingy? | 16:32 |
| *** jfroche changes topic to "SchoolTool development | IRC logs at http://source.schooltool.org/irclogs/ | Dev meetings Mon, 13:30 UTC (15:30 EET) | Use http://paste.lisp.org/new/schooltool for pasting | pcardune's Zope class, Saturdays at noon US/Eastern (GMT-5) (no class today)" | 16:33 | |
| th1a | Thanks. | 16:33 |
| th1a | Lumiere: You around? | 16:33 |
| th1a | ignas: What have you been up to? | 16:34 |
| ignas | lyceum, and lyceum, and thinking of how to pull off the release without having some serious technical problems after that | 16:35 |
| th1a | How are things with the Lyceum? | 16:36 |
| ignas | hmm, most of the new functionality works | 16:37 |
| ignas | i must make some tweaks and test it properly | 16:37 |
| ignas | because it's all a bit of a spike to see if it's possible | 16:38 |
| th1a | The new functionality you've been working on for Lyceum, you mean. | 16:38 |
| ignas | timetables for persons, timetables for groups, calendar for groups | 16:39 |
| ignas | aggregate calendar | 16:39 |
| ignas | that is | 16:39 |
| th1a | Right. | 16:39 |
| ignas | the security functional test for can do is passing now | 16:40 |
| th1a | Ah, good. Did you do that today? | 16:41 |
| ignas | nah, yesterday | 16:41 |
| th1a | Excellent. | 16:41 |
| th1a | OK. jfroche, what have you been up to? | 16:42 |
| th1a | When is that Plone sprint? | 16:42 |
| jfroche | i worked on import for the teachers, courses and sections | 16:42 |
| jfroche | sprint is next week from Sunday 25 til Monday 02 | 16:43 |
| jfroche | but i think i will be able to attend the meeting | 16:43 |
| th1a | OK. So you're really shooting for getting your import done this week. | 16:43 |
| jfroche | right | 16:44 |
| jfroche | the calendaring for the courses won't be done this week as i need to meet with the teacher in charge for the school timetable | 16:44 |
| th1a | OK. | 16:44 |
| jfroche | but i prefer to start with the student notes asap | 16:45 |
| th1a | OK. | 16:45 |
| th1a | jinty: Did you get my email about the meeting in Seville? | 16:47 |
| jinty | hmm, not yet | 16:47 |
| th1a | Hm. | 16:47 |
| * jinty investigates | 16:48 | |
| th1a | Well, anyhow, I'm probably going to an Edubuntu planning meeting of some sort in Spain on May 5-6-ish. | 16:48 |
| th1a | It should be more definite in a few days. | 16:48 |
| th1a | I'd like jinty to be there too if possible, so we can discuss packaging in more detail with the Edubuntu team. | 16:49 |
| jinty | yeah, that would be pretty cool | 16:49 |
| th1a | jinty: You're in Barcelona, right? | 16:49 |
| jinty | yep, Seville is pretty far away | 16:50 |
| jinty | reasonably hot in may as well | 16:50 |
| th1a | I'm sure there'll be a tree for us to sit under. | 16:50 |
| * jinty checks how long a train ride would take | 16:50 | |
| jfroche | ignas: in module header, do we need to update the copyright year ? | 16:51 |
| ignas | i do most of the time | 16:51 |
| ignas | unless i forget | 16:51 |
| th1a | jinty: I could probably fly you. | 16:51 |
| th1a | I'll let you know once things are certain. | 16:51 |
| th1a | OK. Lets return to discussing, in broad terms, what a feisty+1 release would do. | 16:52 |
| th1a | 1) This is a base for a CanDo release. | 16:52 |
| jinty | th1a: there seems to be overnight trains, also I would have to clear this with the guys I work with | 16:53 |
| th1a | 2) It does not have to be particularly featureful. | 16:53 |
| th1a | jinty: Give me a couple days to make sure of what's going on. | 16:54 |
| jinty | th1a: yeah, np | 16:54 |
| th1a | That is, this release should do: | 16:54 |
| th1a | a) calendaring & resource booking; | 16:54 |
| th1a | b) probably some demographics. | 16:55 |
| Lumiere | here now | 16:55 |
| th1a | Hi Lumiere | 16:55 |
| th1a | But Ubuntu users will be looking more for an update of SchoolTool as calendar than screaming for an attendance component. | 16:56 |
| th1a | 3) We want to do something that is easily forward-compatible. | 16:56 |
| ignas | hmm, so timetables, courses, sections, attendance, gradebook, notes can be left out? | 16:56 |
| th1a | ignas: Not that much! | 16:57 |
| ignas | how much then ? | 16:57 |
| Lumiere | I would leave timetables, courses, sections | 16:57 |
| th1a | You need timetables, courses, sections. | 16:57 |
| ignas | so calendaring, resource booking, courses, sections, timetables, terms | 16:57 |
| ignas | for a) | 16:57 |
| Lumiere | th1a: I don't start work until 14:00-14:15... so I'd be even later then I was today if it was moved... | 16:58 |
| ignas | so we can only disable(not enable) gradebook, levels and maybe notes, leaving the rest in the default configuration? | 16:58 |
| th1a | Exactly what the demographics will look like is something we will have to discuss later. | 16:59 |
| th1a | Lumiere: I'm not sure if 1430 is inconveniently late for our European colleagues. | 16:59 |
| Lumiere | it's ok | 16:59 |
| Lumiere | I'll just backrea | 17:00 |
| ignas | 14:30 is good enough for me | 17:00 |
| Lumiere | at some point, it would be good to setup a base package and some choices... so schooltool-resources or schooltool-timetables | 17:00 |
| th1a | ignas: I think so, although there is probably some stuff that can be removed and not just disabled. | 17:00 |
| Lumiere | where timetables installs the idea of courses sections and terms as well | 17:01 |
| th1a | Lumiere: I think we could do that eventually, but it isn't really necessary now. | 17:01 |
| ignas | Lumiere: the problem with the timetables being built into this release | 17:01 |
| ignas | is that we will have problems with making it optional in the future | 17:02 |
| ignas | as in - if you have a database with timetables, you won't be able to apt-get install schooltool without apt-get installing schooltool-timetables | 17:02 |
| ignas | or your database will break | 17:02 |
| th1a | I just don't see it as a big priority either way. | 17:02 |
| th1a | This is the kind of issue we can spend a lot of time on to no real advantage. | 17:03 |
| th1a | Anyhow, regarding customizations and Debian packages. | 17:04 |
| th1a | Is it fair to say that it is not a goal that you can apt-get schooltool, modify it, apt-get update a new version of schooltool, and expect it to not break? | 17:04 |
| th1a | I mean, that's just not a reasonable requirement? | 17:05 |
| ignas | emm - i am not afraid of modifications | 17:05 |
| ignas | if you are modifying - you know what you are doing | 17:05 |
| ignas | but of double click users who would find eggs too difficult | 17:05 |
| ignas | because these will not be able to cope with anything less than an automatic flawless upgrade | 17:05 |
| th1a | ignas: Yes, but if you are modifying, aren't you using a tarball instead of deb? | 17:05 |
| ignas | th1a: in the perfect way - no, you can easily use the deb and add your modifications (if jinty will allow custom site.zcml's and multiple schooltool server instances on the same machine) | 17:06 |
| jinty | th1a: if you apt-get install schooltool, then apt-get install schooltool-cando, in an ideal world you would modify the currently running schooltool | 17:07 |
| ignas | and it's not a problem, you will just update configuration when schooltool gets separated into multiple packages | 17:07 |
| ignas | and get it working | 17:07 |
| ignas | the problem is the automatic installed schooltools | 17:07 |
| jinty | ignas: they are both are allowed, but you've gotta do it yourself | 17:07 |
| ignas | jinty will have a lot of fun writing configuration migration from schooltool to schooltool-timetable + schooltool | 17:08 |
| th1a | ignas: That's one reason I don't want to mess with that. | 17:08 |
| ignas | with what? | 17:08 |
| th1a | Splitting timetable into a new deb package. | 17:08 |
| ignas | if I had enough time to do it for this release it would save us a lot of effort in our next releases | 17:09 |
| * jinty thinks we need to define a policy for add-ons, like: * can be removed without breaking the database, * can be added without breaking the database | 17:09 | |
| Lumiere | I agree with jinty | 17:10 |
| ignas | jinty: all addons can be added (i got the idea from j1m on how to do that) | 17:10 |
| jinty | the basic minimum level that everyone agrees on | 17:10 |
| ignas | the problem is "evolution" from "monolith" to "set of parts" without breaking compatibility | 17:11 |
| th1a | jinty: +1. | 17:11 |
| Lumiere | my issue is, I don't want to get to a point where we have half of it in the schooltool package | 17:11 |
| Lumiere | and other pieces seperate | 17:12 |
| th1a | ignas: Does that problem == lots of tricky evolution scripts and not much else? | 17:12 |
| ignas | th1a: it is probably solvable by lots of tricky evolution scripts + magic in debian post install and pre install scripts | 17:13 |
| ignas | but yes it is doable | 17:13 |
| ignas | even if we will have to inspect internal ZODB evolution status | 17:14 |
| th1a | Like, "doable at a sprint" doable? | 17:14 |
| ignas | no, as in if i would do it before the sprint (no way too much lyceum work on my hands) it would be like 2-3 weeks, if it will will release it will be 2-3 weeks to make it work + 2-3 weeks to make it evolve | 17:15 |
| ignas | s/it will will/we will/ | 17:16 |
| ignas | sprint will be dedicated to packaging | 17:16 |
| ignas | so except for this problem (solvable by sheer programming power some time later) | 17:17 |
| ignas | another one is that I didn't anticipate any releases when comming up with my plan | 17:17 |
| th1a | So are we back around to the fact that there isn't actually time to do this reorganization before a release? | 17:18 |
| ignas | yes | 17:18 |
| th1a | OK. Which is why I was against it at the beginning. | 17:18 |
| ignas | i am just warning about the increase in cost if we release it in the summer, that's all | 17:19 |
| th1a | Yes, this is what I'm trying to sort out. | 17:19 |
| ignas | before this week i though that migration was not solvable at all | 17:19 |
| th1a | What changed your mind? | 17:20 |
| ignas | discussions with alga and justas | 17:20 |
| th1a | Technically? | 17:21 |
| ignas | ? | 17:21 |
| th1a | I mean, was their some kind of technical revelation that made you see that it was possible? | 17:22 |
| ignas | yes | 17:22 |
| th1a | Oh, these laconic Lithuanians... | 17:23 |
| ignas | and the managerial side of it - i mean all the time i spend on release, resource booking and integration is something that is either comming from my free time or from my commitment to lyceum | 17:23 |
| th1a | ignas: Yes. I am very aware of that problem. | 17:24 |
| Lumiere | and I thank you for it | 17:24 |
| th1a | I'm trying to figure out what kind of release, if any, we can do without messing you up. | 17:24 |
| th1a | Getting SchoolTool deployed in Lyceum is the highest priority for both of us. | 17:24 |
| ignas | th1a: last meeting you said that CanDo was the highest priority for us ;) | 17:25 |
| th1a | I didn't mean the highest priority over all. | 17:25 |
| th1a | I meant the highest priority for the release, I think. | 17:26 |
| th1a | Although you can check the log ;-) | 17:26 |
| ignas | you see - splitting it into parts is something that aligns with my needs for lyceum, making it releasable and then evolving from that - is something completely orthogonal to lyceum requirements :/ | 17:27 |
| ignas | i have reused quite some code from the resource booking sprint while doing composite calendars for lyceum ;) | 17:27 |
| th1a | But yes, the entire question is not "is it possible for us to package a release," but "is it possible for us to package a release without messing up our other priorities?" | 17:27 |
| ignas | possible - yes, will cost us after the release - yes, as long as you know that, it's fine with me | 17:29 |
| th1a | I'm willing to incurr some technical debt to get out a simple feisty+1 release that works. | 17:29 |
| ignas | jinty: when you will have time we should talk about the ways to install schooltool to have less problems in the future | 17:29 |
| ignas | as we will have no good configuration files in this release site.zcml management and Data.fs locations will be the important thing | 17:30 |
| th1a | Before we go, everyone: Are we making progress in this conversation or just going in circles? | 17:30 |
| ignas | made some progress, now i know that you know what i know, and what i think | 17:31 |
| ignas | one more question about the release: | 17:31 |
| th1a | jinty, jfroche, Lumiere? | 17:31 |
| ignas | CanDo install will be used by people, and cando guys will be maintaining it (looking at bug reports, explaining people how to use it etc.) | 17:32 |
| ignas | do we want schooltool that will be released used by non technically minded people or not? | 17:32 |
| jinty | ignas: I'm pretty confident in the way that the site.zcml is currently managed in the debian packages, perhaps we should talk about why the way i'm doing it is bad | 17:33 |
| ignas | if not - we should make up a scary name for the package "lib-schooltool" | 17:33 |
| jfroche | th1a: if i can help for this tell me, would be happy to help with migration scripts & configuration | 17:33 |
| ignas | if we want users - we should brace ourselves for support work on our new release | 17:33 |
| jinty | python-schooltool ;) | 17:33 |
| jinty | it already exists... | 17:34 |
| ignas | and really commiting to maintaining it | 17:34 |
| th1a | ignas: I think we'll be more able to handle the support work because we don't have plans to write a bunch of new functionality immediately after the release. | 17:34 |
| ignas | because i am way demotivated by yelling "we know it's a bug, we will not fix it for the next year, because we don't have resources" | 17:34 |
| th1a | And we have more consistent access to you and jfroche. | 17:35 |
| ignas | th1a: i have plans to write quite a lot of new functionality for lyceum | 17:35 |
| * jinty agrees totally with ignas | 17:35 | |
| ignas | as in - i will be as busy as i am now | 17:35 |
| Lumiere | I am hoping that we're going to be able to throw some of the cando interns | 17:36 |
| Lumiere | at the schooltool side of bugs etc | 17:36 |
| th1a | Lumiere: Yes... we just have to do our best. | 17:36 |
| ignas | so we are going for the full "YAY schooltool was released, come and use it, and give us a lot of feedback" | 17:37 |
| ignas | which we will appreciate, and implement | 17:37 |
| th1a | ignas: It will be the half-YAY, rather than the full-YAY. | 17:38 |
| Lumiere | lol | 17:38 |
| ignas | so "yay" not "YAY" | 17:38 |
| Lumiere | yea | 17:38 |
| Lumiere | it's a yea not a YAY | 17:38 |
| th1a | And I'm not going to worry about feature requests. | 17:39 |
| *** mgedmin has joined #schooltool | 17:39 | |
| Lumiere | but I think we need to sit down with welsh and elkner | 17:39 |
| Lumiere | and get them to agree to some of the intern hours to schooltool | 17:39 |
| th1a | This will not be SCHOOLTOOL 1.0 RELEASED!!! | 17:40 |
| Lumiere | yea | 17:40 |
| Lumiere | maybe .1 | 17:40 |
| th1a | Any last words? | 17:40 |
| Lumiere | I'm going to have welsh's attention starting now | 17:41 |
| Lumiere | anything I really need to drag him through the mud for? | 17:41 |
| ignas | jinty: have you tried packaging CanDo? | 17:41 |
| th1a | Lumiere: Not that I can think of. | 17:42 |
| Lumiere | we definately could use a release of the branch same as the last ones. | 17:42 |
| Lumiere | we had a critical bug get fixed last night | 17:42 |
| Lumiere | or 2 nights ago now | 17:42 |
| jinty | ignas: yes, both as an "include the whole of schooltool and zope3 codebase" monolith | 17:42 |
| Lumiere | which is what it does now (the cando-06 pacakge in the schooltool repository is monolithic) | 17:43 |
| jinty | ignas: and as a small and light "just depend on the schooltool package and don't duplicate much" | 17:43 |
| ignas | jinty: hmm, any way to make some "fake" debs with current CanDo and schooltool so i could look at them and meditate about configuration management? | 17:43 |
| jinty | and that's also why cando-06 will probably never make it into a real distribution | 17:44 |
| Lumiere | jinty: which we're ok with | 17:44 |
| Lumiere | we made a decision early on this year that we weren't going to worry about it | 17:44 |
| jinty | Lumiere, yes, glad the message got through;) | 17:44 |
| Lumiere | which is also why we wanted cando-06 | 17:45 |
| Lumiere | instead of cando | 17:45 |
| Lumiere | once we move to a cando package or a schooltool-cando package | 17:45 |
| jinty | ignas, I think there are some debs of the alpha schooltool release lying aroud somewhere | 17:45 |
| Lumiere | we'll be saying that we're going to have a sane update path | 17:45 |
| Lumiere | right now... I can't ensure that | 17:45 |
| ignas | jinty: hmm, i am thinking of "schooltool-trunk" nightly .deb addable to apt-source list | 17:46 |
| ignas | jinty: but that might take too much of your time | 17:46 |
| * th1a drops the bag of gravel. Talk among yourselves... | 17:47 | |
| jinty | ignas: would be nice, but have to get the nightly tarball working again first;) | 17:47 |
| Lumiere | was that the bag of gravel or the bag of gavel? | 17:47 |
| ignas | jinty: anything i can do to help you with that? | 17:48 |
| th1a | Lumiere: The gavel long ago mutated into gravel. | 17:48 |
| Lumiere | ah | 17:48 |
| Lumiere | jinty: I am looking at setting up a buildbot sometime in the near future | 17:49 |
| Lumiere | even if I have to setup cron to get code | 17:49 |
| jinty | ignas: the trunk of this repo was where I last stopped http://svn.debian.org/wsvn/pkg-zope/schooltool | 17:50 |
| jinty | ignas: I just scped a schooltool and python-schooltool deb made from the above source to my home directory on schooltool.org | 17:53 |
| * ignas would like to see the release machinery working until we actually start working on the release | 17:54 | |
| ignas | so that in the sprint we could easily build debs and testinstall them on random ubuntu machines | 17:54 |
| jinty | ignas: +100, but we are in limbo right now... | 17:54 |
| ignas | but yes, it all depends on how Zope3 will be installed | 17:55 |
| jinty | yes | 17:55 |
| ignas | and i have no idea about what we will do with alpha releases of eggs we depend on | 17:55 |
| jinty | they need to be packaged, or we shouldn't be depending on them | 17:56 |
| jinty | in general we shouldn't depend on alpha software anyway... | 17:56 |
| ignas | if maintainers will not package them - we will have to | 17:56 |
| jinty | yes | 17:57 |
| th1a | Perhaps we can get help from Canonical on that. | 17:57 |
| jinty | it's not difficult to package a single egg | 17:57 |
| ignas | th1a: canonical can't do much if the package is maintained by J1m ... | 17:57 |
| ignas | or someone else from Zope Corp | 17:57 |
| jinty | (package as in egg, or deb?) | 17:57 |
| ignas | egg | 17:58 |
| th1a | They can't make a deb of the version we require? | 17:58 |
| ignas | but still - r6789.egg does not seem reassuring | 17:58 |
| ignas | and you can't make someone get the egg into stable releasable state | 17:58 |
| *** Aiste has quit IRC | 18:35 | |
| *** thisfred has quit IRC | 19:24 | |
| jfroche | ignas: person container view takes around 30s to be generated with my 1062 persons. Should i play with catalog there ? | 19:47 |
| ignas | you might try | 19:47 |
| ignas | maybe it will work | 19:47 |
| ignas | but pay attention to the i18n issues when sorting by name | 19:48 |
| jfroche | mean the special character ? | 19:48 |
| ignas | yes | 19:51 |
| jfroche | ok i ll look. thanks | 19:52 |
| Lumiere | ignas: is there a way in ftests to set times | 20:27 |
| ignas | set times? | 20:27 |
| Lumiere | and track how long things are 'allowed' to take... or how long a test takes | 20:27 |
| ignas | want to set time limits? | 20:27 |
| Lumiere | yea | 20:27 |
| Lumiere | because I plan on writing a stress test ftest for cando | 20:28 |
| ignas | problem is that these are very machine specific | 20:28 |
| ignas | as in - my computer might be way slower or a lot faster than yours | 20:28 |
| Lumiere | even if it just reports back how long it takes | 20:28 |
| ignas | you might want to create a file "time_report.txt" | 20:29 |
| ignas | and write times there manualy | 20:29 |
| ignas | by using normal python timing utilities, and data writing operations | 20:29 |
| ignas | or use python logging framework maybe | 20:30 |
| Lumiere | k | 20:32 |
| *** mgedmin has quit IRC | 21:06 | |
| *** Aiste has joined #schooltool | 21:10 | |
| *** jinty has quit IRC | 21:11 | |
| Lumiere | I'm heading home bbiab | 21:30 |
| *** ignas has quit IRC | 21:30 | |
| *** Aiste has quit IRC | 22:19 | |
| *** wjohnsto has quit IRC | 23:06 | |
| *** didymo has quit IRC | 23:40 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!