IRC log of #schooltool for Monday, 2011-03-14

*** aks has joined #schooltool05:18
*** svaksha has quit IRC07:42
*** Lumiere has quit IRC07:42
*** kbrosnan has quit IRC07:42
*** aks has quit IRC07:48
*** svaksha has joined #schooltool07:51
*** Lumiere has joined #schooltool07:51
*** kbrosnan has joined #schooltool07:51
*** aks has joined #schooltool07:53
*** aks has joined #schooltool07:53
*** aks has quit IRC08:02
*** aks has joined #schooltool08:03
*** svaksha_ has joined #schooltool09:33
*** svaksha has quit IRC09:35
*** svaksha_ has quit IRC09:35
*** svaksha has joined #schooltool09:36
*** alga has joined #schooltool09:58
*** yvl has joined #schooltool10:52
*** menesis has joined #schooltool10:59
*** yvl has quit IRC11:15
*** alga has quit IRC11:28
*** yvl has joined #schooltool11:46
*** aks has quit IRC13:20
*** menesis has quit IRC13:45
*** menesis has joined #schooltool14:51
*** replaceafill has joined #schooltool15:11
*** jelkner has joined #schooltool15:14
aelkneryvl, ayt?15:31
jelkneraelkner, hi from pycon, bro!15:32
aelknerhey jelkner, how's it going?15:32
jelknerincredible!15:32
yvlyes aelkner15:32
aelknerdid i miss a memo saying there's no meeting?15:32
jelkneri don't think so15:32
jelkneri know replaceafill had his alarm set15:33
jelkneri wonder if the change in time confused th1a?15:33
replaceafill:)15:33
aelknercould be15:33
yvlmeeting starts in an hour15:39
replaceafill:|15:39
aelknerwhere did you see that?15:39
yvlor did the time change again15:39
aelkneryes, the time changed again15:40
aelknerwhat time is it there?15:40
aelknerand is it going to change for you soon?15:40
yvlit's 15:40 EET, and 8:40 EST15:41
yvland 13:41 UTC15:41
aelknerit's not 8:40 EST, it's 9:4015:41
jelkneryvl and 9:40 "daylight savings time"15:41
jelkneraelkner, there is another letter for our current time15:41
jelknerit isn't EST15:41
yvlit's EDT, right?15:42
jelkneryvl, i think so15:42
yvl9:42 EDT, at least in NYC15:42
aelkneryvl, can we talk intervention catalog in the meantime?15:43
* yvl ducks and covers15:44
aelkner:)15:44
yvllet me just get my coffee :)15:45
yvl5-7 mins ;)15:45
aelknernp15:45
jelknerbtw. replaceafill (with support of the zope community here) has turned us all into pyramid fans15:45
replaceafillnah, mark ramm did it15:46
replaceafillhe was the one who convinced you15:46
jelknernah, it was replaceafill, mark ramm just helped ;-)15:46
jelknerits true that mark's talk was the "tipping point" though15:46
jelknerbut i was only even at that talk because of replaceafill15:47
jelknerso now fsufitchi and replaceafill are writing pybookbuilder in pyramid15:48
replaceafillyou'd still be following massimo if mark wouldnt have made that comment at the end ;)15:48
*** th1a has joined #schooltool15:52
jelknerreplaceafill, dots, dots... ;-)15:52
th1aIt isn't meeting time yet by utc, right?15:52
* th1a just woke up from his daylight savings time transition nap.15:53
* replaceafill wonders when is UTC going to change...15:53
jelknerth1a: will meetings be at 10 am now?15:53
th1aNo, we (I) just never remember to change the meeting time before DST changes.15:54
jelknerth1a: you mean your alarm15:54
th1aSo we traditionally have one odd meeting at the shift.15:54
th1aNo... next week we'll meet earlier by UTC, at 9:30 EDT.15:55
* yvl is back15:55
*** ChanServ sets mode: +o yvl15:57
*** yvl changes topic to "SchoolTool development | IRC logs at http://source.schooltool.org/irclogs/ | SchoolTool Dev meetings Mon, 13:30 UTC (15:30 EET, 9:30 EDT) | Use http://paste.lisp.org/new/schooltool for pasting"15:57
*** ChanServ sets mode: -o yvl15:57
yvlthere, it's fixed15:57
th1aThanks, yvl.15:58
yvlmy pleasure :)15:58
yvlaelkner, ping16:03
aelkneryes16:05
aelkneryvl, can we talk catalog?16:06
yvlsur16:06
yvle16:06
th1aFeel free to just start the meeting... ;-)16:06
* th1a is still feeling a little woozy.16:06
yvlaelkner, do you have some specific questions in mind?16:07
aelknerno, just confusion :)16:07
aelknerand i didn't wake up in a good sleep cycle, so my head is swimming16:08
yvloh :)16:09
yvlit seems that daylight change hit you hard16:09
yvlso... anything specific you're confused about?16:09
aelknerwell, i'm not sure your note exactly suggested a solution to the problem16:10
yvlit did not, actually16:10
yvlyou'll need to put "XXX: these indexes are not updated when IContact changes"16:11
yvlit just needs some work in core16:11
aelknerok, i have a first question16:12
yvlcool :)16:12
aelknerso forgetting evolution for the moment16:13
aelknerwhen i changed the persons_responsible/recipients attributes to be relationship properties16:13
aelknerthat caused the getPersonsResponsible adapter used by the catalog to return the contact list16:14
aelkneras a property rather than a list of ids16:14
aelkneris that wrong already?16:15
yvlno, it's fine16:16
yvlbut you should use a different converter for the index16:17
yvlgetPersonsResponsibleIntIds(object)16:19
yvlthat returns, say, a tuple of int ids for IInterventionPersonsResponsible(object)16:19
yvlon the other hand16:20
yvlumm, scrap that16:20
yvlI don't want to dvelve too deep today :)16:21
yvlI was going to comment on IInterventionMarker interface having no attributes or whatever, but when you adapt an object, you suddenly expect attributes like created16:22
aelkneri was already thinking of getting rid of IInterventionMarker16:23
yvlcool! :)16:23
aelknerand just adapting each object separately16:23
aelkneri see now that such a design is better, more explicit16:23
aelknerbut when could you delve deeper, tomorrow?16:26
yvloh, I can today :)16:26
yvlI just thought you are somewhat sleepy :)16:26
yvlso16:26
aelknerwell, unfortunately, i am, but i'll make due16:26
yvlsay, we get rid of the intervention marker16:27
* yvl is looking at the code16:28
* yvl still looking16:31
aelkneryvl, i had a thought16:32
aelknerhow soon could you get started tomorrow morning your time?16:32
aelkneri could program my sleep clock to still be awake then16:32
aelknerand have a chance to look at your contact branch16:33
yvlhmm, we can do that16:34
yvlhow about 2AM your time?16:35
aelknerthat sounds relaly good for me16:35
aelknerreally16:36
aelknerin the meantime, regarding the widget __init__16:36
aelkneri thought about having properties for things like student and choices16:37
aelknerbut that means calculating them each time they are needed16:37
aelknerwhich seems awfully inefficient16:37
yvlRight16:37
aelknerthat's why i chose to set them in __init__ and be done wth it16:37
yvlbut you could set them in other method16:38
yvls16:38
yvland call those methods from __init__ if you really need to16:38
aelknerdo you mean still do the math in the __init__ step, just break the code out into separate methods16:38
aelknerthat are called from __ini__?16:38
aelknerfor clarity sake?16:39
yvlyes16:39
yvlyes16:39
aelkneroh fine, consider it done :)16:39
yvland also - isAdding just creeps me out16:39
yvlif you had separate methods16:39
yvlit would also be possible to just write a child class16:39
yvlthat calculates two or three things differently16:39
yvlas in - gets the student in a different way, so on16:40
yvlone more thing16:40
yvlthere's from zope.cachedescriptors.property import Lazy16:40
yvlI'm not a big fan of it, but you can do16:40
yvl@Lazy16:40
yvldef calculate_a_lot(self):16:41
yvl   return [1,2,3]16:41
yvlthe calculation is done once16:41
yvlof course it makes it easy to screw up in tests16:41
yvlwhen you forget that this value is actually cached16:41
yvljust so you know that it can be done if you really need to some time16:42
aelkneri would shy away from making the widget even more complex then it is, but...16:42
yvlwell, it is actually made of two widgets, isn't it?16:43
yvlone - normal16:43
yvlas in - editing16:43
yvlanother - adding16:43
yvlthey differ very little, but they do differ16:43
aelkneryes16:43
yvlso splitting into two widgets should yield two simpler widgets16:43
yvlslightly simpler16:43
yvlI'm not asking to do that now, btw16:44
yvlbut separating calculation from __init__ to separate methods is a step in that direction16:44
yvl+hopefully better readability16:44
yvland it's easier to test of course16:44
aelknerhow would the two separate widgets be registered?16:45
th1aYou guys have five more minutes on this.16:45
yvldidn't think about it, frankly16:46
yvlthanks, th1a, we're rounding up I think16:46
replaceafillhey! yvl, nice moving format_table to trunk :)16:46
* replaceafill is looking at yvl's timetable branch...16:46
yvl:)16:46
yvlI think it was always in trunk16:47
aelknerwe can table the rest of the discussion for 2AM16:47
replaceafillyvl, i'm looking for those forms you showed in the sprint, are they here?16:47
yvlnope, no UI changes for tts16:47
yvlold interface stays this time16:47
replaceafill:(16:47
aelkneryvl,  in the meantime, could you email me how to register the widgets separately for add and edit views?16:47
yvlI'll look into it16:48
yvlit's one of those things that I assume that can be done somehow :)16:48
aelknerok, thanks16:48
aelknerth1a, shall i just report?16:49
aelknerwell, i'll take that as a yes :)16:50
aelknerso last week i did the changes for contact properties in intervention messages and goals16:51
th1aSorry, I had a phone call.16:52
aelkneryvl looked them over and has some minor suggestions for fixes, but mostly i think he is ok wth them16:52
yvlyes I am :)16:52
aelknerthe only outstanding issue is something we didn't discuss in setting the bonus target which is the catalog16:53
aelknerit is a sticky issue that i don't think yvl even has a clear plan for16:53
aelknersee, the advantage of relationships is the magical removal of links16:53
aelknerso that when a contact is removed16:53
aelknerit's presence in a message or goal will automatically disappear16:54
aelknerno more orphaned ids as before16:54
aelknerunfortunately, with the catalog, no such magic exists16:54
aelknerif a contact is present in the intervention catalog16:55
aelknerand it is removed by user action, it will still be sitting in the catalog16:55
aelkneryvl, am i right about that?16:56
yvlyes16:56
yvlI just never got to implementing proper relationship catalogs16:57
aelkneranyway, i'll do what i can to get that to work this week along with the test coverage i promissed16:57
aelknerto keep on schedule and all16:57
aelkneryvl and i will meet later tonight for me, tomorrow morning for him16:58
aelknerbut oterwise, that's my report16:58
th1aThanks, aelkner.16:59
th1areplaceafill?17:00
replaceafillok17:00
replaceafilli took friday off because of the conference, but i finished updating the average formulas for cambodia17:01
replaceafilli also added edit buttons for levels and classes17:02
replaceafillas part of:17:02
replaceafill* Improve navigation and action buttons now that most default17:02
replaceafillsuscribers are turned off17:02
replaceafillright now, i'm modifying the teachers add/edit forms to make assinging a teacher to a subject easier17:03
replaceafillyvl, that's why i was interested in seeing your forms :(17:03
yvloh17:03
replaceafilland for this week, i'm going to implement the feature requested by javier of disconnecting subjects from a level17:04
yvlhmm, is there anything in particular that interests you?17:04
yvl(they're not that good in my opinion)17:04
replaceafillyvl, i just wanted to see all those comments you had in your code about the weirdness of z3c.form :)17:05
replaceafillthey looked like warnings for me :D17:05
replaceafill"dont do this", "you have to do this because of this..." and such17:06
replaceafilland mostly, how you created those custom widgets, with dropdowns and stuff :)17:07
yvlthose are mostly my issues with list widget something and of course the ****** object widget17:07
replaceafillanyway, th1a, i also would like your comments on the uruguay sugar event, maybe after the meeting17:07
th1aWhat are your comments about it?17:08
th1aI mean, would it be a good venue for you to promote SchoolTool to the OLPC community?17:08
th1aDo you have a link handy for that?17:08
replaceafillth1a, i liked that people interested in schooltool like tony anderson is possibly attending17:08
th1aIt wouldn't be expensive, right?17:09
replaceafilland that there's people from the #treehouse group coming also17:09
replaceafillthey're the ones who provided the VM for the salvadorean pilot17:09
replaceafillth1a, right, that's a negative factor i guess17:09
th1aIt would or wouldn't?17:09
replaceafillit would17:10
th1aBecause?17:10
replaceafillbecause of the plane ticket17:10
th1aWhat, like $1000?17:10
replaceafillthe "big" airlines take me first to miami or houston and then to uruguay...17:10
th1aAh.17:11
replaceafilli was thinking of maybe going by bus to panama and take a flight there17:11
replaceafillit would be much much cheaper17:11
replaceafillbut i have to find out17:11
replaceafilli'll look today and report back17:11
replaceafillok?17:11
* replaceafill is of course interested in attending :)17:11
th1aOK.17:11
th1aI think it would be worth doing.17:11
th1aWill this be mostly a Spanish-speaking meeting?17:12
replaceafilli guess both17:12
replaceafillpeople from the US is coming17:12
replaceafilland from other south american countries17:12
th1aWell, if anyone is from the US, everyone has to speak English.17:12
th1aThat's the rule.17:12
aelkner:)17:13
replaceafillright17:13
* th1a is happy to be the hegemon.17:13
replaceafill:D17:13
th1aFor now...17:13
th1aOK, thanks replaceafill.17:13
replaceafillok, i guess that's it from me17:13
replaceafillthanks17:13
th1aNo, thank you!17:13
th1ayvl?17:13
yvlwell, timetable wizzard finally works17:14
th1aYay!17:14
yvltrue :)17:14
yvland basically that's it17:15
yvlI'll be working on other views this week17:15
yvlI also spent some time helping Alan17:15
th1aCool.17:16
yvland some time on thinking about things missing in core17:16
yvlinterventions module clearly points to some very reasonable API missing17:16
yvland by thinking I meant - trying to decide17:17
yvlif this should be just fixed in intervention17:17
yvlor it's something I'll need to develop soonish, and when's the best time to do that17:17
yvlrelationships will need a little love17:18
yvlcatalogs, also17:18
yvland something along the lines of weak reference17:18
yvlbasically wrap around int ids17:18
yvlok, that's my report17:19
yvloh, and NotYet exceptions annoyed me to no end17:19
th1aThanks yvl.17:19
th1amenesis?17:19
menesisnothing important to report17:20
menesishave been upgrading zope packages17:20
replaceafillmenesis, thanks for your reply to my skin resources question :)17:20
menesisthat are there since lucid but have deferred upgrading them to our versions.cfg till after alpha 317:21
menesishad problems with packaging branches, but now I know how to workaround that17:21
menesisstill todo < 10 packages17:22
*** jelkner has quit IRC17:23
menesishave looked at recent bugfix releases to include17:23
menesisabout the packages that are new in natty17:23
menesisincluding schooltool packages17:24
menesisI asked cjwatson to add the new packages to my packagesets17:24
menesistwice17:24
menesisbut had no reply since Mar 317:25
th1aLet me know if that's getting critical timeline-wise.17:25
th1aAnything else, menesis?17:27
menesisI would like to make a b1 release at the end of this week17:28
th1aSounds good to me.17:28
menesisthere are two bugs preventing installation that I wanted to upload two weeks ago17:29
menesisso I'm getting impatient17:29
th1aYes, I know.17:29
menesisI still have to merge translations before making a release17:30
th1aA little nagging becomes necessary.17:30
menesishad this planned for last week, but not finished17:30
menesisbtw, replaceafill's gradebook branch was merged at the start of march, finally :)17:31
th1aAh.  Good.17:31
replaceafillmenesis, yeah i noticed, thanks!17:31
menesissorry it took so long17:32
menesisbig branch to review!17:32
replaceafill:D17:32
menesisI do not see any merge requests, are there any branches that people want merged before release?17:32
replaceafilloops! the footer one i think17:33
replaceafilli'll make the request17:33
th1aYes, please.17:33
menesisI have not read all the scrollback of this meeting, please repeat so that I don't miss anything17:33
menesiswill read later17:33
th1aOK, thanks, menesis.17:34
th1aI guess I decided I'm going to try to find a designer for the homepage/book.17:34
th1aSo then I had to start writing up an explanation of the task,17:34
th1aand that led to the "SchoolTool brand values" issue.17:35
th1aBasically it is relevant because the tone of a "teachers working with teachers" project ("Get Involved!") is different than an "Open source tools for NGO's and governments" one.17:36
th1aAnd really we're more of the second now.17:36
th1aBut yes, I have to explain to the designer what feel we want.17:36
th1aI'll send a draft of the design brief today after I look over it again.17:37
th1aAlso, Pandaboard performance is much better with the SSD drive.  We need the SchoolTool benchmarking suite.17:38
yvl:D17:38
yvlI really really would like to do that :)17:38
th1aWell, hopefully this year.17:39
replaceafillmenesis, merge requested17:39
th1aAlso, replaceafill, I've exchanged a couple emails with a CS person in Vietnam who is interested in SchoolTool for a K-12 school SIS project there.17:40
replaceafillah yes!!!17:40
th1aI'm going to refer him to you on details of what you've been doing with Cambodia.17:40
replaceafilllots of interesting questions from him17:40
th1aYes.  The big question there also suggests we need benchmarks -- how many SchoolTools can you run on a central server.17:41
th1aOf course it depends on what you're doing with them.17:41
yvlth1a, just to check17:41
yvlumm17:41
yvlST as a service for multiple schools is not our focus at the moment, right?17:42
th1aI'm definitely in favor of multi-school reporting, etc. being a completely separate application which we aren't interested in writing.17:42
yvlwhere multiple > 10017:42
yvlor > 20 for that matter17:42
th1aI've been discouraged lately from thinking running multiple schools in one instance was a big advantage over multiple instances.17:43
th1ayvl:  Essentially no.17:43
*** jelkner has joined #schooltool17:43
th1aIt is not our focus.17:43
yvlok17:44
yvlI'm just thinking about performance, etc.17:44
yvlby the way17:46
yvlsome time ago you wrote about forking17:46
yvlI think that at some distant time in the future17:46
yvlthere is a possibility of having multiple configurations essentially without forking17:47
th1aCodebase forking or process forking?17:47
yvlcodebase17:47
th1aYes.17:47
th1aOK, let's wrap this up.17:48
th1aHave a great week gentlemen.17:48
yvlI kind of have this gut feeling that we could achieve little forking and "by techers for teachers" feel at some point17:48
th1aMeeting at 9:30 EDT next week.17:48
replaceafillthanks everybody17:48
* th1a drops the bag of gravel.17:49
yvlmaybe we could simply chat about that at some point... maybe next sprint17:49
yvlthanks guys!17:49
replaceafillbtw, i'm flying back home next wednesday, so i'll be probably offline17:49
replaceafill(unless AA gives free wifi in the plane again!)17:49
yvlfree wifi on airplanes17:49
yvlnow that is something17:50
replaceafill:D17:50
replaceafillcomplementary wifi by HP17:50
yvl:D17:50
replaceafill(im going to buy another printer when i get back) ;)17:50
th1alol17:50
yvl:DDDD17:50
th1aiPaq!17:50
replaceafill:D17:50
* aelkner takes a nap to be fresh for 2am meeting with yvl :)17:52
yvlgood night, aelkner  :)17:53
*** klausade has joined #schooltool18:12
*** jelkner has quit IRC18:51
*** replaceafill has quit IRC19:11
*** replaceafill has joined #schooltool19:41
*** svaksha has quit IRC19:43
*** menesis has quit IRC21:56
*** menesis has joined #schooltool22:42
*** menesis has quit IRC22:46
*** menesis has joined #schooltool23:01

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!