IRC log of #schooltool for Wednesday, 2011-11-30

*** menesis has quit IRC00:43
*** replaceafill has quit IRC01:22
*** aks has joined #schooltool05:12
*** aks has quit IRC05:36
*** aks has joined #schooltool05:53
*** aks has joined #schooltool05:53
*** yvl has quit IRC09:41
*** yvl has joined #schooltool09:48
*** menesis has joined #schooltool11:22
*** ignas has joined #schooltool13:16
*** aks has quit IRC13:18
*** menesis has quit IRC16:18
*** replaceafill has joined #schooltool16:22
*** menesis has joined #schooltool16:27
th1aHi menesis, replaceafill, aelkner, yvl.16:30
yvlgood morning16:30
replaceafillgood morning/afternoon16:30
th1aLet's have quick updates and then more discussion of the next release.16:32
aelkneri started working on the courseinfo system16:32
aelkneri'm able to see david's bugs on his demo instance16:33
aelkneras i try to reproduce them on an empty db, i'm finding others bugs16:33
aelknerand i'm trying to figure out how the code works16:34
th1aDon't be afraid to ask filip.16:34
aelkneralso, i'm waiting to talk to david (hopefully today)16:34
aelkneryes, i can always ask him if i can't figure something out, but mostly i need david's intentions16:35
aelkneri'm not sure how much he has actually used this16:35
th1aAt least he generally knows what he wants.16:36
aelkneranyway, that's it for me16:36
replaceafillon monday, david reported an issue with the competency xml importer16:38
replaceafillbasically, they were starting with a fresh database and the importer took about an hour to finish16:38
replaceafillthe new VERSO XML file contains about 41k competencies16:38
replaceafillso it's big16:38
replaceafilli profiled the importer16:39
replaceafilland noticed that cando sets a catalog index for competencies16:39
replaceafilli think every time a competency is created/updated, the catalog index subscribers are called16:39
replaceafillalso, the competency tree is designed in a recursive way16:40
replaceafillthat makes it very hard to update16:40
replaceafillso i tested that, including the lucid packaged version they're using16:41
replaceafilldavid replied today saying that the main issue is not know how long it will take to import16:41
replaceafillthat the user doesn't know if the process is working or not16:41
replaceafillkind of the same problem we have with the XLS importing16:42
replaceafilli also started the spanish version of the book16:42
replaceafillfound a couple of unicode issues with sphinx16:43
replaceafillbut was able to fix them thanks to their mailing list16:43
replaceafillfinally, schwa reported that the book was not showing the ubuntu font16:43
replaceafillwe were missing the <link /> to get the ubuntu web font in the book16:44
replaceafillso i fixed that too16:44
*** jelkner has joined #schooltool16:44
replaceafillth1a, i was wondering if i should continue with the cando issue16:44
th1aCan you just add a javascript spinner?16:45
th1aAnd leave it at that?16:45
replaceafilli could, but that's just an "illusion"16:45
replaceafillto give the user some "hope" that the process is going on16:45
th1aIs there any reason to think that it is not?16:46
th1aTiming out is the biggest problem, which a little javascript would fix, right?16:46
th1aAnd then they do get a confirmation in the end?16:46
replaceafillth1a, yes, they do16:46
replaceafilli don't know how a little js would fix it16:46
th1aJust do that and we'll completely re-write it next year.16:46
replaceafillah ok16:46
replaceafillok, i'll do that16:47
th1aIf it doesn't time out and it works, we're good.16:47
replaceafillyeah, Matt told me that there may be issues with proxies, but they haven't had any so far16:47
replaceafilli think not many users are importing the new xml anyway16:47
replaceafillok, that's it from me16:48
th1aThis seems like the kind of thing more easily re-written from scratch than untangled.16:48
menesisNeed to save the file and import in separate thread. Also show status somewhere16:48
th1aGood... I 'm definitely in favor of just a clean reimplementation.16:49
menesisDon't ask how to do it XD16:49
th1aWe don't need to do it right at this point.16:49
th1aIt would be better to just redirect them to a page that said "OK, come back tomorrow morning, thanks!"16:50
replaceafillbackground processing is not for 2012, right?16:50
th1aWell... it is not a top priority.16:50
* replaceafill done16:51
th1aMaybe starting new threads is easier in Pyramid.  ;-)16:51
th1aThanks replaceafill.16:52
yvlyesterday I was working on batching here and there16:52
yvlalso added usernames to tables that list persons16:52
yvlusers, to be precise16:53
yvltoday I worked on ajaxification of relationships16:53
yvlit is quite exciting, actually16:53
replaceafillyvl, what's the meaning of: Pyatiletka? :)16:54
replaceafilljust curious16:54
yvlwikipedia it ;)16:54
yvlI think we'll need to ajaxify more things in ST16:55
yvlso I'm also adding some infrastructure to help with that16:55
yvlbeing quite conservative not to waste too much time on it though16:56
yvlI think I'll be done Fri/Mon16:56
yvl(with ajaxy tables)16:56
* yvl done16:56
th1ayvl can anticipate my concerns now...16:57
th1anot TOO much infrastructure wanking...16:57
th1aThanks yvl.16:57
menesisBackported 2.0.0 packages to lucid. Copied to maverick. Checked that all dependencies are satisfied on lucid. Copied two missing ones from maverick.16:58
menesisUploaded 2.0 to precise. Link to correct branches from ubuntu packages on launchpad.16:59
menesisReviewed many more bugs. Sorry for the spam.16:59
th1aOK, so at this point the packages in Precise might be reviewed by someone else?17:00
menesisWill file more bugs for myself17:00
menesisYes if someone dares to test precise17:00
menesisIt is there17:01
menesisAlso finished dh-python2 transition17:01
menesisUploaded zope packages17:02
menesisMade all new upstream releases17:02
menesisUpdated ztk17:03
th1aSo... if nothing else happens, 2.0 is in the precise release, and somebody might raise some concerns at some point.17:03
menesisUpdated versionator17:03
menesisIs all.17:03
th1aThanks menesis.17:03
th1aSo I'm thinking that focusing the goals for new features in Precise on transcripts might give us a good focus.17:04
yvlit is long overdue17:04
th1aBasically, to have transcripts, you need to have the school business logic pretty fully fleshed out.17:04
th1a(which is why we don't have it)17:05
th1aOK, so the transcript is, of course, focused on the student.17:05
th1aIt is organized by school year.17:05
th1a(let me know if this is wrong for your experience)17:05
th1aSo you need to know which years the student was actively enrolled.17:06
th1aI guess technically we could do this by membership in the student group.17:06
th1aBut enrollment statuses would be more explicit.17:06
replaceafillth1a, a transcript is like a grades record?17:07
replaceafillah ok17:07
th1aThat's what we're going over now... so we know what we're talking about.17:07
th1aThe point is though, that in thinking about enrollment statuses, we want to think about it working from the point of view of the final transcript.17:08
th1aSo we want to make sure that in the end you can say "this student was enrolled at this date, completed at this date"  Was a student all these years.17:09
th1aExactly how much detail we need to track about crazy enroll/unenroll scenarios, we'll have to think about at bit.17:10
th1aQuestions about that/17:10
th1a(we're not designing this now, I'm just trying to give you the big picture)17:10
yvlbtw, some examples of transcripts would be nice17:11
yvljust to get a sense of how many "other" info may be there17:11
yvllike... I dunno... Maybe some comments,17:11
aelknerif enrollment status is a transitory thing, i'm not sure it can be linked to a static transcript17:11
yvlgraduated cum laude17:12
yvland whatnot17:12
aelkneri think using the prsense in the student group is more static17:12
th1aBasically, I'd be googling them.17:12
yvlI understand, th1a17:12
th1aThe "other" stuff I17:12
th1aam not so much worried about.17:12
yvlit's just that you can probably google better examples17:12
th1aaelkner:  I don't see what you mean.17:13
*** menesis has quit IRC17:13
yvlor maybe it's actually this simple17:13
*** menesis has joined #schooltool17:13
th1aWhat does "static" have to do with anything?17:13
aelkneri'm not sure how enrollment status looks in the db, is it a setting for a student for each year?17:13
th1aIt does not exist.17:14
aelkneri mean in your conception, i know it's not in the app17:14
replaceafillaccording to the original design, it was just an attribute on person17:14
th1aIt can't just be one setting for the year.17:14
th1aBut perhaps it just needs a list of changes.17:14
*** menesis has joined #schooltool17:14
aelkneryes, if it's a history list, then that could be used to determine the transcript17:15
th1aIt doesn't really need to live in the year even since you can figure out what changes happened in which year based on the dates.17:15
aelknerbecause the presence of an entry for a given year would be enough to know to print grades for that year17:15
th1aSo basically, you need to do the same kind of thing with levels, and it probably should have the same implementation.17:16
yvlI think we may end up with additional info on relationships17:16
th1aaelkner was enrolled on August 1, 1979 as a 9th grader.17:16
th1aAdvanced to 10th grade August 1, 1980, etc.17:17
th1aThen we need to know what classes the student enrolled in, which ones he completed, and what the grades were.17:18
yvlalso, maybe: Tim enrolled Computer Science Advanced on June, but dropped that class two months later17:18
th1aAnd calculate a cumulative average.17:18
th1ayes, yvl.17:18
th1aSo actually, these are three instances of the same problem.17:19
* yvl agrees17:20
th1aThat is, tracking changes in status by date.17:20
yvland maybe some additional "status" attribute17:20
yvldropped out; completed; etc.17:20
yvlby date, also17:20
yvlright, that's what you said17:21
* yvl is loosing concentration :)17:21
th1aSo I guess some investment in infrastructure for solving these the "right way" would be justified.17:21
th1aIf we decide there's a "right way"17:21
th1aActually, this is very relevant to schooltool.competency too, as it will track changes in scores over time.17:22
* yvl would say "a way that does not slow down the system tremendously or slows down development speed tremendously"17:22
th1atbh, just keeping a list of changes might be best.17:22
yvlyes, but there's a bit more to that17:23
th1aWell, that might describe any possible solution.  :-D17:23
yvlyes, it's not a matter of if, it's a matter of how and where :)17:24
th1aMight be a matter of just making sure the indexing/cataloging gets done right.17:25
yvl"at least gets done" would also be an improvement :)17:25
th1aActually the main thing might be just making sure it isn't implemented four different ways.  ;-)17:26
yvl(for section membership)17:26
yvlthat too17:26
th1aCome to think of it, we'll need this for excusing absences too...17:27
yvlactually, implementing the base of enrollment does not seem to me like a huge chunk of work to me -- a bit tricky though17:27
th1aSo basically, the next thing you can work on is this pattern, yvl.17:27
* yvl really wanted to do that17:27
th1areplaceafill has most of the basic enrollment statuses done, btw.17:27
yvlbranch, please ;)17:28
replaceafillhaven't been able to find that branch, sorry17:28
replaceafilli'm still looking through my old backups though17:28
* th1a sighs.17:28
th1aDoes this make sense, aelkner?17:29
aelknerhow do you mean?17:30
aelknerbtw, i see two separate issues here17:31
aelknerenrollment status is one thing that we don't have and need to define17:31
aelknercando competency grade history is something completely different17:31
aelknerwe can add the history to the existing data model17:31
aelknerwe just need to change a score value to a list and have a method for getting current score17:32
th1a"how do we track changes in data over time" is a generic question.17:32
th1a(not necessarily a hard one)17:32
aelkneri would say a list is the answer17:32
aelknersorted by date with method for latest17:33
*** jelkner has quit IRC17:33
aelknerand evolution script to turn atomic value into list of one17:33
th1aI think we're all more or less on the same page.17:33
th1aSo I can send them in with the report/proposal.17:34
th1aThat is all.17:34
th1aHave a great rest of the week and weekend gentlemen!17:35
aelknercan i wait until monday so that i don't need to break up a week?17:35
* th1a drops the bag of gravel.17:35
th1aNo aelkner.17:35
aelknerso just don't include this week17:35
th1aGive me whatever you want tomorrow aelkner.17:35
yvloh, I remembered a small thing17:35
yvlresource attributes17:35
aelkneri can include this week in december17:36
yvland their evolution17:36
yvlis it done / needs bug report / needs to go on the blueprint?17:36
yvlthose bookable things17:37
yvlsome attributes were removed this summer17:37
yvlreplaceafill, do you remember status of this?17:38
replaceafillyvl, i think it's not done17:38
replaceafillat least by me17:38
yvlI vaguely recall that was "to be done" a month ago17:38
replaceafillbut i remember i needed the new description field for it17:38
replaceafillbecause of the notes field17:38
yvlcan you please file a bug report about that?17:39
yvl(and put some small note what needs to be done)17:39
replaceafilli vaguely recalls too yvl but i'll try17:39
* yvl would hate to have poop left in old databases17:40
yvl(unevolved attributes, that is)17:40
replaceafillWrite evolution script for resources, make the "type" attribute a selection list demographics field and notes a description demographics field17:40
replaceafillthat's in my notes17:40
yvlit would be good to have at least something there17:42
yvlwe can flesh out that area later this year17:42
yvl(there -> Launchpad)17:42
yvlok, thanks guys!17:42
yvlhappy coding17:42
yvl(and happy report writing, th1a :) )17:43
yvl( it should be quite fun this time :) )17:43
th1aWell... I'm finding it a bit less so than I would have thought.17:43
th1aWe set the stage for future growth but didn't actually grow much.17:44
th1aSo it isn't a complete slam dunk.17:44
yvlstill - a good place to be17:45
th1aMuch better than before.17:45
th1aGetting CL's transition to 2.0 right is huge.17:45
th1aYou and I might be taking a trip to Portugal for a few days next year.17:45
* yvl doesn't mind going to Portugal :)17:49
yvlI'd like to take a week of mid January17:49
yvlor more likely somewhere around the end of January17:50
yvl* off17:50
replaceafillth1a, could yvl work on this soon?
yvljust wanted to check if we don't have anything important in Jan17:50
replaceafillsorry about being nagging about this, but i think we really need it17:50
yvllike a sprint or such17:50
th1aThere will be a sprint around there.17:51
yvlhmm, right, replaceafill17:51
th1aWe'll plan it as soon as we get approved for next year.17:51
yvlI'm kind of buying vacation plane tickets soon17:51
yvlso it would be good not to clash17:51
th1ayvl:  You could do that bug next.17:51
th1aWhen do you want to buy them?17:52
yvlafter ajaxy happy stuff, I promise, replaceafill!17:52
*** ignas has quit IRC17:52
replaceafillyvl, thanks!17:52
yvlth1a, somewhere Jan 15th - Feb 8 th17:52
th1aI mean, how soon do you need to book it?17:53
yvlthe sooner the better, but it can wait a week or two17:53
th1aOK.  I'll give you some dates soon.17:53
th1aI suspect we aren't going to do the Lvov thing.17:54
yvl(wasn't it Lodz?)17:54
yvlI'd be okay with that :)17:54
th1aOh, Lodz.17:55
th1aSomething like that.  ;-)17:55
yvlallright then17:55
yvlsee you soon, guys17:55
th1aLater yvl.17:57
* replaceafill is turning his flourish notes into bugs, so excuse the spam :)18:05
th1aThanks replaceafill.18:05
replaceafillThere are some that i don't understand now like: "Remove ubuntu font link to googleapis" :|18:05
replaceafillfor some reason i wanted to remove it ;)18:06
replaceafillth1a, what's the main reason for "Remove ethnicity field from default demographics fields"?18:07
replaceafillso i can put it in the bug notes18:07
replaceafill"because it's offensive"18:07
th1aPeople are offended by it.18:07
replaceafillah ok18:07
*** ignas has joined #schooltool19:24
th1aIs that what I should be using?19:35
menesisyou are welcome :)19:37
th1aseries flourish?19:38
menesisto match other projects, yes19:39
menesisi have also published the old book to
th1aYes, thanks.19:40
replaceafillmenesis, nice!19:41
replaceafillnow i can add bugs for the book ;)19:41
menesisyes :)19:41
replaceafillmenesis, are you doing one for the website?19:41 or lp:schooltool-org?19:41
replaceafillyour call :D19:42
menesiswill do later, with the dot19:42
* menesis has to go home19:42
replaceafillthanks menesis19:42
menesisth1a: I need the three images for branding on launchpad19:42
*** menesis has quit IRC19:47
th1aActually, I should probably ask Vinny for fresh versions because mine are all flat at the bottom for some reason.19:51
* replaceafill looks his logos19:52
*** ignas has quit IRC20:13
*** menesis has joined #schooltool20:51
*** menesis has quit IRC21:16
*** menesis has joined #schooltool23:03
*** replaceafill has quit IRC23:47

Generated by 2.15.1 by Marius Gedminas - find it at!