| *** ignas_ has joined #schooltool | 02:00 | |
| *** alga has quit IRC | 02:09 | |
| *** ignas_ has quit IRC | 02:32 | |
| *** aks has joined #schooltool | 04:52 | |
| *** aks has quit IRC | 05:09 | |
| *** aks has joined #schooltool | 05:09 | |
| *** yvl has joined #schooltool | 10:00 | |
| *** aelkner has quit IRC | 10:40 | |
| *** Khalsa has quit IRC | 10:40 | |
| *** Khalsa has joined #schooltool | 10:41 | |
| *** aelkner has joined #schooltool | 10:52 | |
| *** alga has joined #schooltool | 12:03 | |
| *** ignas_ has joined #schooltool | 13:25 | |
| *** ignas_ has quit IRC | 13:44 | |
| *** ignas_ has joined #schooltool | 14:34 | |
| *** menesis has joined #schooltool | 14:45 | |
| *** menesis has quit IRC | 14:46 | |
| *** aks has quit IRC | 15:58 | |
| *** hjbehling has joined #schooltool | 16:08 | |
| hjbehling | hi, is the CAS module still being actively developed? (or are there other attempts to make schooltool auth plugable into other systems ldap, etc?) | 16:10 |
|---|---|---|
| *** jelkner has joined #schooltool | 16:17 | |
| *** replaceafill has joined #schooltool | 16:19 | |
| yvl | hey hjbehling | 16:20 |
| hjbehling | yvl: hi! | 16:20 |
| yvl | at the moment CAS module is not dev focus | 16:20 |
| hjbehling | ok | 16:20 |
| hjbehling | are there plans for pluggable auth? | 16:20 |
| yvl | but it should work, as well as ldap | 16:20 |
| hjbehling | ldap auth is available? | 16:21 |
| yvl | there are plans to bring those up-to-date, but not immediate ones | 16:21 |
| yvl | ldap auth *should* work | 16:21 |
| yvl | though we haven't tested it for a while | 16:21 |
| hjbehling | are there directions anywhere? | 16:21 |
| hjbehling | is LDAP auth built-in or do I need to download a plugin | 16:22 |
| hjbehling | I'd actually prefer to do LDAP authbut I would be willing to do CAS in order to integrate schooltool into my SSO environment | 16:22 |
| yvl | it's not built in, and I'm afraid not documented as well | 16:23 |
| * yvl is trying to remember the status... | 16:23 | |
| yvl | https://lists.launchpad.net/schooltool-developers/msg00492.html | 16:23 |
| yvl | actually, we're having the annual ST sprint next week | 16:24 |
| yvl | I'll bring this up, if it's ok | 16:24 |
| *** aks has joined #schooltool | 16:26 | |
| hjbehling | that would be awesome | 16:27 |
| hjbehling | i'll take a look at this plugin and see if i can get it to work | 16:27 |
| yvl | ok, so here's this https://code.launchpad.net/~schooltool-owners/schooltool/schooltool.ldap | 16:27 |
| hjbehling | i tested the CAS plugin a while ago (with jasigCAS) and there was a small bug involving the site identification string that ST sent to CAS | 16:28 |
| yvl | http://bazaar.launchpad.net/~schooltool-owners/schooltool/schooltool.ldap/view/head:/README.txt | 16:28 |
| *** th1a has joined #schooltool | 16:28 | |
| hjbehling | I was going to try to modify the code but I didn't have time | 16:28 |
| hjbehling | but I'd really like to get ST running for our equipment checkout so I'm going to make time to focus on it again | 16:29 |
| * th1a notices his xchat was disconnected for some reason... | 16:29 | |
| yvl | so... you downloaded CAS and put a slug, right? | 16:29 |
| hjbehling | yes | 16:30 |
| yvl | cool | 16:30 |
| yvl | you could try doing a similar thing with LDAP | 16:30 |
| th1a | hjbehling: The extent to which CAS and LDAP are a priority is directly proportional to the number of people who ask about it. | 16:30 |
| hjbehling | cool, i'll try it | 16:30 |
| th1a | hjbehling: Do you use something for SSO now? | 16:31 |
| hjbehling | th1a: gotcha | 16:31 |
| yvl | options are listed in README and the mail to schooltool-developers | 16:31 |
| yvl | hope it helps | 16:34 |
| hjbehling | th1a: we have a working kerberos+ldap SSO for webmail and system logins | 16:34 |
| hjbehling | i've used apache2-mod-auth-kerb for a small number of sites | 16:34 |
| * th1a is trying to remember what kerberos would do for us... | 16:34 | |
| th1a | Where "remember" means checking wikipedia... | 16:34 |
| hjbehling | kerberos can be used as a backend to CAS, or LDAP | 16:34 |
| th1a | The thing about this stuff is the standards tend to be not quite standard enough. | 16:35 |
| yvl | well, LDAP is quite widely used, IMHO | 16:35 |
| th1a | So it is the kind of thing that it is hard to provide free support for. | 16:35 |
| yvl | but I might be very wrong :) | 16:35 |
| th1a | I'm not saying they aren't widely used. | 16:35 |
| th1a | Just that at least 10 years ago, certain LDAP's didn't work together in certain configurations. | 16:36 |
| hjbehling | right, it seems like the most popular is for apps to support Active Directory Auth, which usually means they support LDAP auth | 16:36 |
| th1a | It just makes it hard for us to support remotely. | 16:36 |
| hjbehling | and if you can check passwords in LDAP, you can have LDAP check passwords in Kerberos... which sounds really convoluted but imo worth it | 16:37 |
| th1a | So at this point our CAS and LDAP "work" but the overall situation discourages me from trying to completely polish and promote them, since I don't have a lot of faith that they'd ever work without often requiring site-specific tweaks to setups we don't know about or understand.. | 16:38 |
| th1a | However, we should help people who are serious about SchoolTool use them. | 16:38 |
| hjbehling | cool. well i'll try out the ldap plugin and see where it gets me | 16:39 |
| th1a | OK. Don't hesitate to ask when you get stuck. | 16:39 |
| th1a | This is the kind of thing our developers should offer as paid support. ;-) | 16:39 |
| hjbehling | from what i remember the CAS plugin was really close to working but it wasn't adding a forward slash to the FQDN of the refering site (or was adding the forward slash) and that was stopping me from getting it going | 16:39 |
| hjbehling | although if LDAP works then CAS is one less thing I have to worry about | 16:40 |
| th1a | That's what I'd say. | 16:40 |
| th1a | Actually, one reason I liked pushing CAS is it makes the support easier for us. | 16:40 |
| hjbehling | jasigCAS is a serious pain... and i wasn't able to get anywhere with rubyCAS | 16:40 |
| th1a | There are just a couple CAS servers we have to talk to instead of myriad LDAP setups. | 16:41 |
| th1a | On the other hand, if CAS doesn't work, it is a bad idea. | 16:41 |
| th1a | So... | 16:41 |
| hjbehling | do you use java or rubycas? | 16:41 |
| th1a | You know, the one site we really used CAS with actually ended up writing their own in PHP! | 16:41 |
| th1a | So that's not exactly a ringing endorsement. | 16:41 |
| hjbehling | ha | 16:42 |
| th1a | I guess the bottom line is if you aren't using it for anything else, don't worry about it. | 16:42 |
| th1a | It would probably be easier to add kerberos support to SchoolTool (I'm guessing). | 16:42 |
| hjbehling | yeah i'm just not comfortable working in java/tomcat | 16:42 |
| * yvl too :D | 16:42 | |
| hjbehling | too much messy configuration, poor debuging | 16:43 |
| th1a | I should probably not encourage CAS at this point. | 16:43 |
| hjbehling | i wish there was a cas server in python | 16:43 |
| hjbehling | then i might have a fighting chance | 16:43 |
| th1a | It doesn't seem like it has gotten critical mass as a standard. | 16:44 |
| th1a | When I started this kind of work I was really excited about being a good citizen and supporting open standards whenever possible. | 16:45 |
| th1a | Unfortunately, it is almost always a waste of time and money. | 16:45 |
| th1a | Except for a few key (and usually simple) things. | 16:45 |
| hjbehling | maybe that's a good thing.. :) it would be great though if there was something out there that would work like it does in theory.. but ldap seems to fit the bill pretty well | 16:46 |
| th1a | OK, hjbehling, this is actually our developer meeting time, so I probably should start that. | 16:46 |
| hjbehling | thanks! | 16:46 |
| th1a | No problem. | 16:46 |
| th1a | hi yvl, aelkner, replaceafill. | 16:46 |
| replaceafill | good morning/afternoon | 16:46 |
| aelkner | morning | 16:46 |
| yvl | good morning guys :) | 16:47 |
| th1a | yvl: Did you send the links about the ajax lists to aelkner & replaceafill? | 16:48 |
| yvl | umm, no | 16:50 |
| yvl | http://69.164.203.135:38000/persons/@@index.html | 16:50 |
| yvl | http://69.164.203.135:38000/schoolyears/2008-2009/groups/students | 16:50 |
| yvl | and edit members link there: http://69.164.203.135:38000/schoolyears/2008-2009/groups/students/members_persons.html | 16:51 |
| th1a | So here's my feeling right now. | 16:51 |
| yvl | it su***! :D | 16:51 |
| th1a | I don't think this works as an evolution of the current view. | 16:51 |
| * yvl does not object | 16:52 | |
| th1a | I would pause to note that changing this view in the first place was all my idea, so all this is on me. | 16:52 |
| th1a | Basically, we need to spend an afternoon at the white board in Arlington re-thinking these views from scratch. | 16:53 |
| yvl | +1 | 16:53 |
| th1a | In particular whether or not the add/delete/edit view needs to be the main view for a group. | 16:53 |
| th1a | Or, I guess it isn't. | 16:54 |
| th1a | ;-) | 16:54 |
| th1a | But what that relationship should be. | 16:54 |
| th1a | That is, we should start with the group view and define how the edit views work in relationship a little more clearly. | 16:56 |
| th1a | Maybe use an accordion on the main group view, etc. | 16:56 |
| th1a | Or some other js widgetry. | 16:57 |
| * yvl can't imagine what an accordion could do there, but - sure | 16:57 | |
| th1a | I'm just brainstorming. We can do this next week. | 16:58 |
| th1a | One question, yvl: it doesn't work to have two dynamic lists where you can delete above and add below, right? | 16:58 |
| yvl | it's a bit tricky | 16:59 |
| th1a | But if you couldn't delete from the current list, could you (fairly easily) have dynamic adding? | 16:59 |
| yvl | mostly from the user experience part | 16:59 |
| th1a | That is, if you had an ajax add view and ajax delete view? | 16:59 |
| th1a | But don't try to do them both at once, is that easier? | 16:59 |
| yvl | yes | 17:00 |
| yvl | frankly, I'd like to have some grouping behaviour | 17:00 |
| yvl | but let's chat live | 17:00 |
| th1a | Yes. | 17:00 |
| yvl | it'll be faster | 17:00 |
| yvl | in any case | 17:00 |
| th1a | Now is a good time to stop. ;-) | 17:00 |
| th1a | What do you want to do the rest of the week? | 17:00 |
| yvl | it would help a lot if you guys thought a bit how you imagine that view to work | 17:01 |
| yvl | now that's a good question, th1a | 17:01 |
| th1a | Yes, I'm thinking. | 17:01 |
| yvl | thanks! :) | 17:01 |
| replaceafill | could i request selenium work? | 17:01 |
| replaceafill | i mean, for yvl :) | 17:01 |
| * yvl does not have a fast PC at hand | 17:02 | |
| yvl | mine broke last month | 17:02 |
| replaceafill | yvl, do tests pass in a slow pc? | 17:02 |
| yvl | yep | 17:02 |
| yvl | but I can look at it anyway | 17:02 |
| yvl | btw, I spent most of today looking at Django VS Pyramid | 17:03 |
| yvl | just something I'll bring up at the sprint | 17:03 |
| yvl | no need to make a decision, just to think about it a bit | 17:03 |
| yvl | ok... so I'll look at Selenium | 17:04 |
| yvl | what else... | 17:05 |
| yvl | can I take a short look at CanDo? | 17:05 |
| th1a | Definitely. | 17:05 |
| th1a | We need to triage what is worth salvaging and what should just be re-written. | 17:06 |
| th1a | That'll be a major topic of conversation. | 17:06 |
| yvl | my thoughts exactly | 17:06 |
| yvl | ok, if I run out of things to do, I can always get to fixing bugs | 17:06 |
| th1a | The biggest design question is how to store each evaluation of a competency. | 17:07 |
| th1a | Right now it is just one score per person/competency, right? | 17:07 |
| * yvl thinks so | 17:07 | |
| th1a | We need much better tracking of each evaluation, even though that will mostly be hidden from the user unless they want to drill down. | 17:09 |
| aelkner | yes, there is only one score per comp at the moment | 17:09 |
| yvl | like history? | 17:09 |
| th1a | So that could just be a list but maybe something else. | 17:09 |
| aelkner | a time-soreted list would be needed | 17:09 |
| th1a | At least. Maybe a set of separate indexed objects. | 17:10 |
| th1a | We have to talk through the more complicated use cases. | 17:10 |
| th1a | And what performance needs to be optimized. | 17:10 |
| yvl | good points | 17:11 |
| th1a | Bear in mind that competencies that are evaluated by multiple teachers are important. | 17:11 |
| th1a | And that teachers will want to see in their gradebook *their* evaluations. | 17:11 |
| yvl | thanks, this is useful info | 17:11 |
| yvl | very useful, actually | 17:12 |
| yvl | it's looking to be a busy sprint :) | 17:13 |
| th1a | I'd say the simplest implementation is a list of scores per standard for each person. The most complicated is a separate object for each evaluation. | 17:15 |
| th1a | I want aelkner to be able to write tests after the sprint. | 17:17 |
| th1a | So if there are specific issues/questions aelkner, now would be a good time to air them. | 17:17 |
| aelkner | we all need to be ale to write tests, not just me | 17:17 |
| aelkner | so far, we haven't really written any for flourish | 17:18 |
| aelkner | but i don't any specific questions, just waiting to follow someone's lead on flourish tests | 17:19 |
| aelkner | i remember that datepicker is holding us up though | 17:20 |
| yvl | one more reason for me to look at Selenium :) | 17:20 |
| yvl | and lang. settings | 17:20 |
| yvl | and something else | 17:20 |
| aelkner | yes | 17:20 |
| aelkner | replaceafill was right to ask yvl to look into that | 17:20 |
| yvl | apologies for delay on my side | 17:21 |
| th1a | replaceafill filed bugs, right? | 17:21 |
| yvl | yes | 17:21 |
| replaceafill | https://bugs.launchpad.net/schooltool/+bug/893186 | 17:21 |
| th1a | replaceafill & yvl: Stay in sync on this please. | 17:21 |
| yvl | will do | 17:21 |
| * replaceafill is setting up an oneiric vm | 17:21 | |
| replaceafill | want to try a slower "pc" :) | 17:22 |
| yvl | :D | 17:22 |
| yvl | oh, and if you pointed me to how reproduce 1. of the bug | 17:23 |
| yvl | it would be great | 17:23 |
| yvl | the modal buttons part | 17:23 |
| replaceafill | yvl, i remember i stop the person test in that part | 17:23 |
| replaceafill | let me check | 17:23 |
| replaceafill | person/browser/stests/password_edit.txt | 17:24 |
| replaceafill | line 213 | 17:24 |
| replaceafill | although the test is out of date now i think | 17:24 |
| replaceafill | since the password edit form changed later | 17:24 |
| yvl | ok, I'll try that one | 17:25 |
| replaceafill | oh | 17:26 |
| yvl | let's catch up tomorrow @ default meeting time on IRC | 17:26 |
| replaceafill | and about 3. | 17:26 |
| replaceafill | i think the right env variable should be LANGUAGE in ubuntu | 17:26 |
| yvl | I think I'll just disable translations | 17:26 |
| replaceafill | but i didnt try it yet | 17:26 |
| replaceafill | yes, that would be better :) | 17:26 |
| yvl | probably ;) | 17:27 |
| replaceafill | 2. is hackable | 17:27 |
| replaceafill | setting the dates first | 17:27 |
| replaceafill | and then the title | 17:27 |
| replaceafill | (at least in the add year form) | 17:27 |
| yvl | I think we just need some waitFor style helpers there | 17:28 |
| yvl | wait-until-the-dialog-is-not-visible | 17:28 |
| replaceafill | +1 | 17:28 |
| replaceafill | let me know if you want me to test anything you change | 17:28 |
| yvl | sure | 17:29 |
| replaceafill | th1a, could we have a couple of hours during the first day of the sprint for questions to yvl? :) | 17:29 |
| replaceafill | like asking his opinion on stuff | 17:29 |
| yvl | :D | 17:29 |
| replaceafill | i've put two bugs on pause because of that | 17:30 |
| th1a | Yes. | 17:30 |
| replaceafill | (the demo fields ones) | 17:30 |
| replaceafill | kk, thanks | 17:30 |
| th1a | Saturday is kind of business. | 17:30 |
| th1a | We'll do the CanDo stuff during the week. | 17:30 |
| th1a | So Sunday can be more general ST stuff. | 17:30 |
| th1a | I'm working on the agenda and will send out the hotel info, etc. | 17:31 |
| yvl | great | 17:32 |
| th1a | We have a studio and a suite. | 17:32 |
| yvl | was about to ask that :) | 17:32 |
| th1a | So someone has to sleep in a sofabed, but nobody has to share a room. | 17:32 |
| th1a | bedroom | 17:32 |
| th1a | replaceafill got the sofa last year. | 17:32 |
| replaceafill | 2010 :P | 17:33 |
| th1a | I keep thinking that was last year. | 17:33 |
| * yvl can take the sofabed if need be :) | 17:33 | |
| th1a | It is yours yvl. | 17:33 |
| th1a | Some things go by age. ;-) | 17:34 |
| aelkner | :) | 17:34 |
| th1a | OK. | 17:35 |
| th1a | Thanks yvl. | 17:35 |
| th1a | replaceafill? | 17:35 |
| replaceafill | ok | 17:35 |
| replaceafill | last week i worked on the css bug | 17:35 |
| replaceafill | styled the /mygrades view | 17:35 |
| replaceafill | and also found a new error in the description demo fields | 17:36 |
| replaceafill | it was a missing class declaration | 17:36 |
| replaceafill | that prevented the form from accessing the error's properties | 17:37 |
| replaceafill | i didnt report a bug, just fixed it and put a comment in the old bug | 17:38 |
| replaceafill | (although the cause was different) | 17:38 |
| replaceafill | then i moved to make the gradebook look like a spreadsheet | 17:38 |
| replaceafill | removing the extra padding for cells | 17:38 |
| replaceafill | i changed our border rules in css | 17:39 |
| replaceafill | from: | 17:39 |
| replaceafill | border: 1px solid #cccccc; | 17:39 |
| replaceafill | to border-width, border-style and border-color rules | 17:39 |
| replaceafill | so border-color could be on colors.css | 17:40 |
| replaceafill | which makes more sense to me | 17:40 |
| replaceafill | small stuff | 17:40 |
| replaceafill | also took a look at this one: https://bugs.launchpad.net/schooltool.gradebook/+bug/886257 | 17:41 |
| replaceafill | i was thinking about just marking the average menu | 17:41 |
| replaceafill | and setting an special rule for it, to move it to the left | 17:41 |
| replaceafill | instead of looking for a general solution that adjust the menus automatically | 17:41 |
| th1a | That sounds fine. | 17:42 |
| replaceafill | (that was a topic for conversation during the sprint for me btw) | 17:42 |
| replaceafill | :) | 17:42 |
| replaceafill | cool, i'll do that then | 17:42 |
| replaceafill | i think that's it | 17:42 |
| th1a | One question: if you started now how long do you think it would take to fix this date format nightmare? | 17:43 |
| replaceafill | i can have something ready to show on the wed meeting | 17:44 |
| replaceafill | i can stop css fixes | 17:44 |
| replaceafill | and move to that back | 17:44 |
| th1a | I'd rather get the dates fixed. | 17:44 |
| replaceafill | (i kind of inverted the order this past week) | 17:44 |
| replaceafill | ok | 17:44 |
| replaceafill | will work on that then | 17:44 |
| th1a | OK. Thanks replaceafill. | 17:44 |
| th1a | aelkner? | 17:45 |
| aelkner | i got the transcript request modal to work | 17:45 |
| aelkner | there is a need to fix ST.dialogs.submit | 17:46 |
| aelkner | right now the $.ajax() call has a success handler, but no error handler | 17:46 |
| yvl | there are more places like that | 17:47 |
| aelkner | that means, if the request template has a silly typo, hitting Download just results in a blank modal | 17:47 |
| yvl | and there's also the session timeout issue | 17:47 |
| aelkner | right, so we need to talk about that kind of stuff at the sprint | 17:47 |
| yvl | +1 | 17:47 |
| aelkner | luckily, i was able to figure out what the typo was in firebug, just time consuming | 17:48 |
| yvl | it's a < week's worth of a fix | 17:48 |
| aelkner | that's what i meant before about modals being difficult, not that they are not doable, just slow | 17:48 |
| aelkner | but anyway, we'll have that sped up before too long | 17:49 |
| aelkner | anyway, the pdf comes up now, but it doesn't have any data in it | 17:49 |
| aelkner | so i moved onto the layout view | 17:49 |
| aelkner | i thought that adding Transcript Layout to Customize would be appropriate | 17:50 |
| aelkner | right below Report Card Layout | 17:50 |
| aelkner | th1a, as per our conversation last wed, i thought that eventually a transcript | 17:50 |
| aelkner | type would need to be a dropfown in the layout view | 17:51 |
| aelkner | but for now we can assume that it is the first one we come up with | 17:51 |
| aelkner | eventually, the choice of transcript type would potentially change the | 17:51 |
| aelkner | set of choices that follow it | 17:51 |
| aelkner | assuming that we have just the first simple form of transcript we | 17:51 |
| aelkner | only have one drop-down (so far) that the user needs to choose from | 17:52 |
| aelkner | that is, the report activity source for the term final grade data | 17:52 |
| th1a | Well, there is with or without gpa in terms of type. | 17:52 |
| aelkner | well, before introducing type, i thought i'd get the simplest one to work for starters | 17:53 |
| aelkner | and that would be, just showing the one and only one score for each section attended | 17:53 |
| aelkner | based on the choice of which deployed report sheet activity is to be the source | 17:54 |
| aelkner | so the output would be a table with two columns, course and grade | 17:55 |
| th1a | I'm just saying I don't think it gets more complicated than that except you have to decide if there will be a gpa or not. | 17:55 |
| aelkner | with the term being a breaker for visual aid | 17:55 |
| th1a | Just remember that only the last term in a linked section should count for the score. | 17:55 |
| aelkner | ah, so don't show scores for earlier terms for linked sections? | 17:56 |
| th1a | No. | 17:57 |
| th1a | At least that's what we'll try to get away with. | 17:57 |
| aelkner | ok, that helps because there was going to be the issue of term score versus final score | 17:57 |
| th1a | The last term is the final score. | 17:57 |
| aelkner | cool | 17:57 |
| th1a | So anyway, get something simple going for me to look at to make sure you don't spend the week marching in the wrong direction. | 17:58 |
| aelkner | so for now the Transcript Layout view will have just the one drop-down to select deployed report sheet activity | 17:58 |
| th1a | Yes. | 17:58 |
| aelkner | and we can discuss type when we have the simplest one working | 17:59 |
| th1a | Yes. | 17:59 |
| aelkner | so i'm clear on what to do, and i'll press on with it this week | 17:59 |
| th1a | OK, so we can close on a nice round minute... | 17:59 |
| yvl | :) | 18:00 |
| th1a | Have a good couple of days and see you guys Wed. | 18:00 |
| * th1a drops the bag of gravel. | 18:00 | |
| yvl | looking forward to it! | 18:00 |
| aelkner | thanks guys | 18:00 |
| yvl | see you guys soon! | 18:00 |
| replaceafill | thanks everybody | 18:00 |
| aelkner | cya wednesday | 18:00 |
| *** aks has quit IRC | 18:37 | |
| *** Khalsa has quit IRC | 19:33 | |
| replaceafill | yvl, confirmed, if i set my VM to 40% cpu, the categories stest passes :) | 20:34 |
| replaceafill | (no dialog in that one though) | 20:35 |
| replaceafill | if i set the VM to 100% cpu, the test fails sometimes | 20:35 |
| th1a | replaceafill: I just sent a reply to Cambodia. | 20:51 |
| *** ignas_ has quit IRC | 21:32 | |
| replaceafill | th1a, nice reply, thanks | 21:40 |
| *** jelkner has quit IRC | 23:30 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!