| th1a | I don't understand why. | 00:00 |
|---|---|---|
| jinty | the backend lookup mechanism in duplicity is via protocol | 00:00 |
| jinty | so you can do: duplicity / scp://host/xxx | 00:01 |
| jinty | or: duplicity / ssh://host/xxx | 00:01 |
| th1a_linux | Oh. | 00:01 |
| jinty | but as close to the url scheme of s3 would be better, perhaps w can do something like s3+http:// | 00:02 |
| jinty | a la subversion | 00:02 |
| th1a_linux | That seems reasonable. | 00:03 |
| jinty | ok, th1a if you have a new account/bucket, I can do some more testing, then try a complete backup | 00:16 |
| th1a_linux | We may have to resume this at a later date. | 00:17 |
| jinty | k then, just ping me | 00:19 |
| *** srichter has quit IRC | 00:29 | |
| *** didymo has joined #schooltool | 01:20 | |
| *** Gwynn has quit IRC | 01:47 | |
| *** jinty has quit IRC | 02:01 | |
| *** Gwynn has joined #schooltool | 04:01 | |
| *** pcardune has quit IRC | 06:51 | |
| *** pcardune has joined #schooltool | 07:52 | |
| *** ignas has joined #schooltool | 10:29 | |
| *** srichter has joined #schooltool | 10:45 | |
| *** tiredbones has quit IRC | 11:20 | |
| *** alga has joined #SchoolTool | 11:22 | |
| *** tiredbones has joined #schooltool | 11:24 | |
| *** Aiste has quit IRC | 12:38 | |
| *** jinty has joined #schooltool | 12:45 | |
| *** mgedmin has joined #schooltool | 12:49 | |
| *** Aiste has joined #schooltool | 13:01 | |
| povbot | /svn/commits: * ignas committed revision 5854: | 13:11 |
| povbot | /svn/commits: Refactored SectionInstructor view. | 13:11 |
| povbot | /svn/commits: * ignas committed revision 5855: | 13:19 |
| povbot | /svn/commits: Renamed template as it will be used for more than selecting instructors. | 13:19 |
| *** faassen has joined #schooltool | 13:26 | |
| *** didymo has quit IRC | 13:34 | |
| povbot | /svn/commits: * ignas committed revision 5856: | 15:09 |
| povbot | /svn/commits: Refactored Relationship Edit views even further. Made Section Learner views reuse the same base class and template. | 15:09 |
| povbot | /svn/commits: * ignas committed revision 5857: | 15:35 |
| povbot | /svn/commits: ZCML cleanup. | 15:35 |
| *** thisfred has joined #schooltool | 15:38 | |
| povbot | /svn/commits: * ignas committed revision 5858: | 15:49 |
| povbot | /svn/commits: Added resources BoundRelationshipProperty to the Section. Added resource booking view for sections. | 15:49 |
| *** vmx_ is now known as vmx_afk | 15:50 | |
| povbot | /svn/commits: * ignas committed revision 5859: | 15:52 |
| povbot | /svn/commits: Added list of booked resources to the section view. Added link to the resource booking view to the section view. | 15:52 |
| *** ignas has quit IRC | 16:50 | |
| *** alga has quit IRC | 16:55 | |
| *** mgedmin has quit IRC | 16:55 | |
| *** Aiste has quit IRC | 16:57 | |
| *** Reza_M has joined #schooltool | 17:00 | |
| *** alga has joined #SchoolTool | 17:06 | |
| *** mgedmin has joined #schooltool | 17:06 | |
| *** Aiste has joined #schooltool | 17:08 | |
| *** ignas has joined #schooltool | 17:13 | |
| *** vmx_afk is now known as vmx | 17:13 | |
| * th1a shuffles some papers around. | 17:28 | |
| * faassen jumps up. | 17:29 | |
| th1a | Good morning everyone. | 17:30 |
| alga | er, hi | 17:30 |
| th1a | Evening for some. | 17:30 |
| th1a | Most... | 17:30 |
| faassen | afternoon. | 17:31 |
| *** ignas_ has joined #schooltool | 17:31 | |
| *** ignas has quit IRC | 17:31 | |
| mgedmin | hi | 17:31 |
| th1a | srichter: Are you going to be with us, or are you occupied sprinting? | 17:31 |
| *** ignas__ has joined #schooltool | 17:31 | |
| th1a | Agenda: | 17:32 |
| th1a | * should we change the start time for the meeting? | 17:32 |
| th1a | * work plans from POV & Infrae | 17:32 |
| th1a | * should we reshuffle POV's contract a bit? | 17:33 |
| th1a | * the next ST 2006 release. | 17:33 |
| th1a | So... is this time ok? | 17:33 |
| alga | move to 1330Z! | 17:34 |
| alga | I've got a lecture at 1500Z | 17:34 |
| th1a | Is 1330 UTC ok for everyone? | 17:34 |
| faassen | that works for me, though as soon asyou get to daylight savings.. | 17:35 |
| faassen | it all goes an hour earlier.. | 17:35 |
| faassen | or something. | 17:35 |
| faassen | oh, wait. | 17:35 |
| faassen | no, it won't. | 17:35 |
| faassen | it's fine for me. | 17:35 |
| th1a | We're on daylight savings now. | 17:35 |
| th1a | It is 10:30 here. | 17:35 |
| faassen | oh, right, it's just we're on UTC for the meetings that's bugging us. | 17:35 |
| th1a | Yes. | 17:35 |
| faassen | anyway, fine for me. | 17:35 |
| mgedmin | any hour on tuesday works for me | 17:36 |
| th1a | OK, 1330 it is. | 17:36 |
| srichter | th1a: here | 17:36 |
| th1a | Hi srichter. Where are you? Switzerland? | 17:36 |
| *** mgedmin changes topic to "SchoolTool development | IRC logs at http://source.schooltool.org/irclogs/ | Dev meetings Tue, 13:30 UTC (16:30 EEST)| CanDo dev meetings Tue, 4pm EST" | 17:36 | |
| th1a | mgedmin: Thanks. | 17:37 |
| th1a | OK. POV, what are you guys planning on working on this week? | 17:37 |
| *** ignas has joined #schooltool | 17:38 | |
| alga | I'm working on deleting STTs and terms | 17:38 |
| alga | Gintas is working on resource management in timetables | 17:38 |
| srichter | th1a: yeah | 17:38 |
| alga | Later, we'll start on timezones in timetables | 17:38 |
| srichter | (not sprinting yet, though) | 17:38 |
| alga | we seem to be in a bit of a crunch | 17:38 |
| th1a | To hit your deadline? | 17:39 |
| alga | yep | 17:39 |
| alga | but I've still got hope | 17:39 |
| th1a | Are ignas and mgedmin working on ST this week? | 17:40 |
| alga | unfortunately, mgedmin is held off for most of this week | 17:41 |
| alga | but the rest of us are on schooltool. | 17:41 |
| th1a | OK. | 17:41 |
| th1a | ignas: Did we ever finish an estimate of how long it would take to fix the acl for taking attendance? | 17:42 |
| alga | ignas is flipping on and off for some reason | 17:42 |
| alga | but today we finished estimating it | 17:42 |
| alga | there are two alternatives: | 17:43 |
| alga | erm, | 17:43 |
| alga | making admins and clerks access attendance is easy | 17:43 |
| *** _ignas has joined #schooltool | 17:43 | |
| alga | making teachers who teach a given sections poses problems | 17:43 |
| *** th1a has quit IRC | 17:43 | |
| alga | there are two ways to handle that: | 17:44 |
| *** th1a has joined #schooltool | 17:44 | |
| th1a | Whoops. | 17:44 |
| alga | 1) hacky-simple: check in the view, show a polite message if a teacher cannot access a sections attendance | 17:44 |
| th1a | Colloquy is a great IRC client except that it crashes periodically. | 17:45 |
| alga | 2) the-right-thing: introduce real relationships-based security | 17:45 |
| alga | our estimates are 6 and 16 nuts for these solutions, respectively | 17:45 |
| alga | we would recommend going for the Right Thing (TM) | 17:46 |
| alga | in return, we suggest to drop corrections (18 nuts) | 17:46 |
| alga | it is the most isolated story | 17:46 |
| th1a | I was just thinking the same thing. | 17:47 |
| alga | wouldn't make it either easier or harder to do it later | 17:47 |
| th1a | Right. | 17:47 |
| th1a | Can you explain relationship-based security a bit more. | 17:47 |
| *** ignas_ has quit IRC | 17:48 | |
| alga | we want to be able to define permissions on objects like this: | 17:48 |
| th1a | I know we talked about it in Vllnius, but I'd like some feedback from faassen and mgedmin. | 17:48 |
| alga | <object class="foo.bar.baz" permission="schooltool.edit" relationship="foo.bar.URITeacher" /> | 17:48 |
| alga | it would give us a flexible and generic mechanism to implement different security models | 17:49 |
| alga | based on relationships between objects | 17:49 |
| faassen | are you thinking about implementing a new security policy? | 17:49 |
| faassen | i.e replacing Zope's default one? | 17:49 |
| alga | faassen: perhaps | 17:49 |
| alga | most probably we would extend the zope one | 17:49 |
| faassen | okay, I urge you to look into it. | 17:49 |
| faassen | okay, I'd recommend against that. | 17:49 |
| faassen | with the documentlibrary, I've used the zope one. | 17:50 |
| faassen | and it worked, but I figured in the end having gone with a custom security policy might've made things clearer. | 17:50 |
| faassen | the containment nature of the default policy (objects have to have a __parent__) is also very frustrating. | 17:50 |
| alga | the zope security policy is very tightly optimized | 17:50 |
| mgedmin | I was thinking about custom security checkers and the default zope security policy | 17:51 |
| alga | I'm somewhat afraid of having to do through all that ourselves | 17:51 |
| th1a | That might take a few more nuts. | 17:51 |
| faassen | well, I wonder whether we couldn't come up with a new security policy. I don't think we need to worry too much about optimization. | 17:51 |
| mgedmin | although I haven't investigated the feasibility of that in detail | 17:51 |
| faassen | I mean, the security story has a strong tie-in with user interface. | 17:51 |
| faassen | currently it sucks. | 17:51 |
| faassen | too many options, and too hard to reason about. | 17:51 |
| alga | yes | 17:52 |
| faassen | perhaps just dumping the options completely and replacing it with one security policy that's hard-coded but at least one we can reason about would make sense. | 17:52 |
| alga | but coming up with a new security model is too mindbogling | 17:52 |
| faassen | you could base it on location. | 17:52 |
| mgedmin | we could steal the one SteveA presented last EuroPython | 17:52 |
| mgedmin | adapters | 17:52 |
| faassen | there's also one by zope corp, here.. | 17:52 |
| th1a | faassen: Generally, yes, giving people much fewer options is the way to go. | 17:53 |
| *** ignas__ has quit IRC | 17:53 | |
| faassen | can't find that policy right now. | 17:53 |
| faassen | anyway, this needs some design and discussion, I'd suggest. | 17:53 |
| faassen | like, what kind of security use cases we have in the first place, overall. | 17:54 |
| faassen | who needs to be able to access what? | 17:54 |
| th1a | Now would be a good time to discuss it. | 17:54 |
| alga | indeed | 17:54 |
| faassen | perhaps there's something in there we can express in some easy declarative way. | 17:54 |
| faassen | there seem to be notions of ownership.. | 17:54 |
| alga | these are the scary questions we're all talking about for at least a year | 17:54 |
| faassen | and notions of 'everybody in a group can access this' | 17:54 |
| th1a | Well, the specific case we're talking about now relates to attendance. | 17:54 |
| faassen | and then you have those relationship issues you came up with. | 17:54 |
| th1a | Specifically, only the instructors of this section can take attendance for it. | 17:55 |
| th1a | Not all instructors. | 17:55 |
| th1a | Or all teachers. | 17:55 |
| th1a | Only the instructors of this section. | 17:55 |
| th1a | And that's defined by a relationship. | 17:55 |
| th1a | This pattern will come up elsewhere. | 17:55 |
| *** ignas_ has joined #schooltool | 17:56 | |
| th1a | Parents will need to be given access to only their children's records (I mean human parents and children, not objects). | 17:56 |
| faassen | you could also model it as ownership, I think. | 17:56 |
| faassen | but yeah, relationship would work as it, too. | 17:56 |
| th1a | I think ownership would be dicey for expressing this. | 17:57 |
| faassen | Zope corp's alternative security policy is here: | 17:57 |
| *** ignas has quit IRC | 17:57 | |
| faassen | http://svn.zope.org/zc.sharing/trunk/src/zc/sharing/ | 17:57 |
| faassen | see policy.txt | 17:57 |
| th1a | You can have multiple instructors, and they aren't the people who will create the object. | 17:57 |
| *** ignas__ has joined #schooltool | 17:57 | |
| faassen | it would make sense to investigate that. | 17:57 |
| faassen | okay, yeah, that's dicey. | 17:58 |
| mgedmin | I'm a bit afraid checking relationships on every attribute access might result in a system that cannot be described by the phrase "fast enough" | 17:58 |
| faassen | mgedmin: have you investigated zc.relationship? :) | 17:58 |
| faassen | mgedmin: it seems to have performance optimizations. | 17:58 |
| mgedmin | no | 17:58 |
| faassen | anyway, perhaps someone can write this down. | 17:59 |
| faassen | first. | 17:59 |
| faassen | and then we can discuss it. | 17:59 |
| faassen | like, what are really the requirements for this system? | 17:59 |
| faassen | how might it look in the user interface? | 17:59 |
| faassen | what would a typical user want to be able to do with it? | 17:59 |
| faassen | that kind of stuff. | 17:59 |
| th1a | Well, I've talked about this problem a lot with POV. | 18:00 |
| faassen | well, I haven't and it'd be useful for us to become involved for 2 reasons. | 18:00 |
| faassen | 1) fresh perspective | 18:00 |
| faassen | 2) user interface perspective. | 18:00 |
| ignas__ | faassen, you see - we haven't talked about this from the user perspective | 18:00 |
| ignas__ | as we do not really have any real users yet | 18:00 |
| faassen | well, currently there *is* a user interface. | 18:01 |
| faassen | that I don't want to show to real users. | 18:01 |
| faassen | so something's gotta happen. | 18:01 |
| th1a | Well, are we trying to talk about acl in general, or this attendance problem? | 18:01 |
| ignas__ | acl in general | 18:01 |
| faassen | I'd recommend at least starting a conversation about access in general. | 18:01 |
| faassen | having a particular use case is useful. | 18:01 |
| ignas__ | faassen, we know the state is awful, but imho we should base modifiactions on user input not on semi wild guesses ... | 18:02 |
| th1a | Well, I think we all agree that giving the user/site manager far fewer options is essential. | 18:02 |
| alga | sorry, I gotta run to the uni... | 18:02 |
| *** alga has quit IRC | 18:02 | |
| th1a | I don't think we're just making wild guesses. | 18:03 |
| th1a | At least I don't think I am. | 18:03 |
| *** _ignas has quit IRC | 18:03 | |
| ignas__ | we are not making any guesses i am afraid | 18:03 |
| faassen | ignas__: you cannot expect end users to give sensible input about complicated security issues. you need to think about what end users really want. | 18:03 |
| ignas__ | as i haven't seen any documentation describing exact roles of clerks/school admins/ trachers | 18:03 |
| faassen | I mean, the current system is not a good base to get user input about. | 18:03 |
| ignas__ | as well as their commonly performed tasks | 18:04 |
| th1a | Essentially, I don't think there is any way to have one ACL screen that can be used with every object. | 18:04 |
| th1a | For an application as varied as SchoolTool. | 18:04 |
| faassen | I hope we can have no acl screen altogether. | 18:04 |
| ignas__ | +1 | 18:04 |
| th1a | faassen: Exactly. | 18:04 |
| faassen | and just hardcode the story for now. | 18:04 |
| th1a | Yes. | 18:05 |
| ignas__ | we must have the "story" first | 18:05 |
| faassen | well, we seem to have some stories. | 18:05 |
| faassen | perhaps someone ought to write them down? | 18:05 |
| faassen | like, "student needs access to these objects".. "teacher teaching a section needs access to these objects" | 18:05 |
| mgedmin | +1 | 18:06 |
| mgedmin | for writing down the security spec | 18:06 |
| faassen | so who is going to write it? | 18:06 |
| th1a | Could you be more specific about that? | 18:06 |
| faassen | th1a: can you? | 18:06 |
| th1a | Since it falls on me. | 18:06 |
| faassen | th1a: well, a student would need read access to ..um, his calendar events.. | 18:07 |
| ignas__ | th1a, we need default policies for schooltool - who can see what, who can do what | 18:07 |
| faassen | th1a: and write access to his calendar events. and read access to the sections he's following or perhaps all of them. | 18:07 |
| faassen | th1a: and someone teaching a section needs write access to that section, plus ..some form of access to all the students in the section, say, their demographics data. | 18:07 |
| th1a | I mean, if we can come up with a list of questions now, I can answer them. | 18:07 |
| faassen | th1a: can you give an example question? | 18:08 |
| *** srichter has quit IRC | 18:08 | |
| th1a | I think coming up with the questions and answers at the same time is a bit overwhelming to me. | 18:08 |
| faassen | th1a: i don't know what questions you're asking for. :) | 18:08 |
| th1a | Like, "What objects does a student need to access?" | 18:08 |
| faassen | th1a: I mean, we can make a list of a lot of things in the system that someone could access. | 18:08 |
| faassen | th1a: demographics data of a student, attendance of students, section.. | 18:09 |
| faassen | th1a: we also have a list of the roles/groups in the system. | 18:09 |
| faassen | th1a: students, teachers, etc | 18:09 |
| ignas__ | faassen, you see - there are objects that do not have clear policy, schooltool should be adaptable for schools all over the world, and policies will be different ... | 18:09 |
| th1a | So make a grid. | 18:09 |
| faassen | th1a: we need stories where the system is seen from the perspective of the student..what information can they access and what can they alter? as well as a teacher of a section, etc. | 18:09 |
| th1a | ignas__: That's why we've got partner schools. | 18:09 |
| th1a | We've only got to deal with a limited number of cases right now. | 18:10 |
| faassen | ignas__: yeah, but in this case having at least one functional story would be useful. we can always plug in different policies. | 18:10 |
| ignas__ | th1a, that's why i am saying that we must talk to them first and then design security, not try reading their mind | 18:10 |
| faassen | ignas__: they won't be able to understand your question half the time. :) | 18:10 |
| faassen | ignas__: and a large section of the rest of the time they'll give answers that are at most incomplete. | 18:10 |
| th1a | It will be easy enough to turn certain permissions on or off for them manually. | 18:11 |
| faassen | ignas__: though I fully agree we need to talk to them, I think we have some ideas ourselves to improve the present situation drastically. | 18:11 |
| ignas__ | role security will and should simplify everything (as soon as we will have it eworking), but for that we need clearly defined roles wit hclearly defined/sparated functions | 18:11 |
| faassen | ignas__: role-based security would be nice. | 18:11 |
| ignas__ | s/role/role based/ | 18:12 |
| faassen | ignas__: I didn't know that was on the agenda already. it won't solve everthing though, though that's where the relationships come in too, I hope. | 18:12 |
| ignas__ | i would actually suggest starting form clearly defining roles | 18:12 |
| faassen | ignas__: that'd be good. student, teacher, etc. but are these roles local or global? | 18:12 |
| ignas__ | and then thinking from their perspective will be a lot easier | 18:12 |
| th1a | I'm not trying to NOT talk to the partners. I have, in fact, discussed this with them, and I think just talking to me is good enough for getting 80% of the way there. | 18:12 |
| ignas__ | th1a, cool :) | 18:13 |
| faassen | ignas__: and do you need global roles like 'student' and then local roles like 'student_on_this_section'? | 18:13 |
| faassen | anyway, okay, what if I write something down about all this? | 18:13 |
| faassen | and then we go from there? | 18:13 |
| faassen | I'll write something down and post it to the list. | 18:13 |
| th1a | faassen: That would be fine with me. | 18:14 |
| *** ignas_ has quit IRC | 18:14 | |
| th1a | Might loosen the log jam. | 18:14 |
| faassen | okay. I can compare with document library issues, where we had complicated issues with security. | 18:14 |
| ignas__ | faassen, i think thinking from the point of view of the "acting" role shoul clear up such issues, thus not "can students of this section access this attribute" but "can i see attribute foo of section i am a member of" | 18:14 |
| faassen | ignas__: is an acting role like a local role? | 18:14 |
| ignas__ | ? | 18:15 |
| th1a | ? | 18:15 |
| ignas__ | can you define the term "local role" | 18:15 |
| faassen | ignas__: a role you have on an object (and its subobjects), but not elsewhere.. | 18:15 |
| th1a | I don't think ignas is a Zope 2 hacker. | 18:15 |
| ignas__ | nope | 18:15 |
| faassen | ignas__: i.e. "On this section I am the teacher" | 18:15 |
| ignas__ | i am not | 18:15 |
| faassen | they exist in Zope 3 too, local roles. | 18:16 |
| faassen | that is, all roles are local, but right now.. in schooltool, you only have global 'roles', in the form of groups. | 18:16 |
| faassen | i.e. they're global to all of schooltool. | 18:16 |
| faassen | you're either a teacher or you're not. | 18:16 |
| faassen | but you can also model things by giving people local roles. | 18:16 |
| th1a | Hm... perhaps I didn't know Zope 3 had local roles. | 18:16 |
| faassen | you have local permissions, but that's harder to reason about. | 18:16 |
| faassen | th1a: all roles in Zope 3 are local. | 18:16 |
| faassen | th1a: that is, you can assign people roles locally. | 18:17 |
| faassen | th1a: I think you can also create new roles locally but that's not interesting. | 18:17 |
| faassen | that is, in the standard Zope 3 security policy. | 18:17 |
| faassen | we could have a security policy which doesn't use roles, anyway, we'll figure it out. | 18:17 |
| faassen | I propose we move on. | 18:17 |
| th1a | I think we could use local roles to handle the attendance use case, as long as the roles were kept in sync with the relationships. | 18:17 |
| th1a | OK. | 18:18 |
| faassen | th1a: (when you say in sync ..I think custom security policy, though a custom adapter might do too) | 18:18 |
| th1a | I was thinking "subscriber" but I don't know what I'm talking about. | 18:18 |
| ignas__ | th1a, we were bittent by subscribers way too many times :/ | 18:18 |
| faassen | oh, it's solvable with events, it's just probably more hairy. | 18:18 |
| th1a | Seemed like what srichter would say. | 18:18 |
| ignas__ | th1a, they are causing problems when evolving the system | 18:19 |
| th1a | th1a: OK, let's move on. This was useful, though. | 18:19 |
| faassen | anyway, we'll mull this over, I'll post something to the list later this week, we'll take it from there. | 18:19 |
| faassen | okay, good, I always worry I am perceived as throwing spanners in the works. | 18:19 |
| th1a | ignas: When would you guys need to start writing this stuff? | 18:19 |
| faassen | th1a: they can start thinking along this week already. :) | 18:20 |
| th1a | I mean, they'll have to start it soon if they're going to meet their deadline. | 18:20 |
| th1a | Which is in two weeks. | 18:20 |
| faassen | oh, that's soon. | 18:20 |
| ignas__ | i am not sure, probably in the begining of the next week, but as our schedule is extremally compressed it might get pushed into this week/weekend | 18:21 |
| mgedmin | reworking the entire schooltool's security infrastructure in two weeks is not practical | 18:21 |
| th1a | Yes, this is certainly true. | 18:21 |
| ignas__ | we are not going to rework the infrastructure | 18:21 |
| ignas__ | i can promise you that | 18:21 |
| ignas__ | i still want to have most of the functionality for the deadline | 18:21 |
| th1a | Perhaps cheap option 1 is best for now. | 18:21 |
| *** Reza_M has left #schooltool | 18:22 | |
| ignas__ | th1a, if we are going to spend time fixing it asap (next contract) then maybe | 18:22 |
| th1a | We would be. The problem is that an attendance system with no ACL is an incomplete toy, no matter how featureful it is. | 18:23 |
| th1a | So we don't want to finish a second round of attendance work with a toy. | 18:23 |
| th1a | Even if it is a slightly hacky non-toy. | 18:24 |
| th1a | Does that make sense? | 18:24 |
| ignas__ | th1a, sorry i have lost it now ... | 18:24 |
| ignas__ | so we want - ugly hacky solution for april 18 | 18:24 |
| ignas__ | or we want the generic relation based security policy | 18:24 |
| th1a | I think ugly hacky. | 18:25 |
| ignas__ | or we want to rework the security of schooltool but we will not do thet because of time constraints | 18:25 |
| ignas__ | and what are we going to do in the next contract then ? | 18:25 |
| th1a | ACL! | 18:25 |
| ignas__ | ugly hacky means - hack something up so that: | 18:25 |
| ignas__ | Teachers can only acces attendance views for sections they are | 18:26 |
| ignas__ | teaching to. | 18:26 |
| ignas__ | th1a, ACL is a bit meaningless, sorry | 18:26 |
| ignas__ | we have ACL and they are not really working that well :/ | 18:26 |
| ignas__ | what do you mean by ACL ... | 18:26 |
| th1a | I guess AC not ACL. | 18:26 |
| th1a | I mean a new access control story for SchoolTool. | 18:27 |
| ignas__ | AC as in - desing proper AC, or just add relationship based access control. | 18:27 |
| th1a | Yes. | 18:27 |
| faassen | AC as in we're going to figure it out. :) | 18:27 |
| th1a | Yes. | 18:27 |
| faassen | I mean, this really needs a bit of thought and writing things down. | 18:28 |
| faassen | it's easy to get lost in all kinds of edge cases with it. | 18:28 |
| faassen | so we'll see whether we can get some good ideas on it. | 18:28 |
| th1a | The problem is right now it is like we're designing an automated teller system, except we haven't written the part where you enter your PIN number, | 18:28 |
| faassen | I think the relationship model could help there too, indeed. anyway, we'll see. | 18:28 |
| th1a | so if you show it to a bank at this point, it is hard for them to take it seriously. | 18:28 |
| faassen | the access control story has repercussions throughout schooltool on many levels. | 18:29 |
| th1a | See what I mean? | 18:29 |
| th1a | And this is pretty much the case because I've been afraid of dealing with access control. | 18:29 |
| th1a | So it isn't your fault. | 18:29 |
| mgedmin | we were all afraid, I think | 18:30 |
| faassen | it's a hairy area. | 18:30 |
| faassen | so that's all understandable, but it needs to be tackled if we want to expose it to users. | 18:30 |
| faassen | we can then of course tweak in response to feedback as ignas is suggesting. | 18:30 |
| faassen | that's essential. | 18:30 |
| th1a | OK. Well, perhaps faassen can get us started in resolving it. | 18:30 |
| faassen | but we need to have a better base to start from. | 18:30 |
| faassen | what's the next topic? | 18:30 |
| ignas__ | the big bang ? | 18:31 |
| * mgedmin checks the watch | 18:31 | |
| th1a | Particularly if we can come up with a systematic way of getting requirements from me & users. | 18:31 |
| th1a | Yes, we can wrap up. | 18:31 |
| faassen | okay, well, as to Infrae's status.. | 18:31 |
| th1a | Oh... I forgot Infrae! | 18:31 |
| th1a | Sorry. | 18:31 |
| faassen | last week I didn't have a lot of time, this week I started again and I hope to get at least another developer in for feedback. | 18:31 |
| faassen | I've made some progress with a prototype still hiding out in infrae.com's svn as it's more a sketch than real code. | 18:32 |
| faassen | of the demographics story. | 18:32 |
| th1a | Let me know when I can take a look. | 18:32 |
| faassen | in the sketch i have table-based display, formlib based forms, catalog driven search, batching, and a todo file. :) | 18:32 |
| faassen | but it all looks horrible and it's flakey and doesn't have tests. | 18:32 |
| faassen | I've also been spending time trying to update the schooltool buildout infrastructure.. | 18:33 |
| faassen | so I can start pulling in some dependencies like the various zc.* packages. | 18:33 |
| faassen | I've worked a bit with jinty as he did some interesting work with eggs. | 18:33 |
| faassen | and I've been doing some experiments eggifying zc.resourcelibrary today. | 18:33 |
| faassen | they do take away cruft from the makefile into Python code we didn't write ourselves (and thus need not maintain), so I think that's the right direction | 18:34 |
| faassen | this work needs to be done for my work to be able to land in schooltool proper, as we need a number of dependencies for that. | 18:34 |
| th1a | Good. I'll be happy when we've got those problems straightened out. | 18:35 |
| ignas__ | th1a, btw, you can see the resource booking for sections in the newest checkout (no evolution scripts yet, thus should only be tested on a clean Data.fs) | 18:35 |
| th1a | ignas: OK. Thanks. | 18:35 |
| faassen | okay, that's infrae's status for now. | 18:35 |
| faassen | I expect more time this week and next week. | 18:36 |
| th1a | Thanks faassen. | 18:36 |
| th1a | OK, that wraps up this meeting. | 18:36 |
| faassen | so I hope I can finally make progress into Tom-visible stuff. :) | 18:36 |
| * th1a bangs the virtual gavel. | 18:36 | |
| *** mgedmin has quit IRC | 18:45 | |
| *** Aiste has quit IRC | 18:46 | |
| *** ignas__ has quit IRC | 18:53 | |
| *** Aiste has joined #schooltool | 18:55 | |
| *** mgedmin has joined #schooltool | 18:56 | |
| tiredbones | I just read this morning's irc meeting dicussion. I haven't used use-cases to define data relationships before, but have used it to define UIs'. Could someone point me to docs that explains this type of analysis? | 20:11 |
| tiredbones | When you folks use the term security, is this another term for data integrity/quality? | 20:16 |
| *** mgedmin has quit IRC | 20:24 | |
| *** mgedmin has joined #schooltool | 20:24 | |
| povbot | /svn/commits: * gintas committed revision 5860: | 20:25 |
| povbot | /svn/commits: Ripped out some other parts using section locations that I missed last time. Added a missing test case. | 20:25 |
| povbot | /svn/commits: * gintas committed revision 5861: | 20:25 |
| povbot | /svn/commits: Added an evolution script that gets rid of the old ISection.location and updates the new relationship attribute. | 20:25 |
| *** jinty has quit IRC | 20:45 | |
| *** faassen has quit IRC | 20:48 | |
| *** alga has joined #SchoolTool | 20:48 | |
| *** jinty has joined #schooltool | 20:56 | |
| *** Aiste has quit IRC | 21:03 | |
| *** mgedmin has quit IRC | 21:08 | |
| *** alga has quit IRC | 21:14 | |
| *** Aiste has joined #schooltool | 21:35 | |
| *** thisfred has left #schooltool | 21:41 | |
| *** pcardune has quit IRC | 21:45 | |
| *** srichter has joined #schooltool | 22:16 | |
| *** jinty has quit IRC | 22:33 | |
| *** Aiste is now known as Aiste|away | 22:35 | |
| *** pcardune has joined #schooltool | 22:38 | |
| *** jelkner has joined #schooltool | 22:59 | |
| jelkner | pcardune: why mr. cardune, how r u? | 23:00 |
| pcardune | hi jelkner | 23:00 |
| pcardune | i'm good | 23:00 |
| pcardune | how are you? | 23:00 |
| jelkner | great, do u know if welsh will be joining us? | 23:00 |
| pcardune | i believe so, He might call me though. | 23:01 |
| jelkner | ahh | 23:01 |
| pcardune | but I'll tell him to get onto irc | 23:01 |
| jelkner | if he does call, get on irc | 23:01 |
| jelkner | ;-) | 23:01 |
| pcardune | jelkner: welsh is getting onto irc | 23:12 |
| *** dwelsh has joined #schooltool | 23:12 | |
| dwelsh | hello | 23:13 |
| jelkner | hi dave! | 23:13 |
| pcardune | hi dwelsh | 23:13 |
| dwelsh | sorry about being late | 23:13 |
| jelkner | no problem | 23:14 |
| jelkner | what is on the agenda | 23:14 |
| jelkner | i have a few things | 23:14 |
| jelkner | 1. pcardune's airline tickets | 23:14 |
| pcardune | I'd like to here about the NSF stuff if there's time | 23:14 |
| jelkner | 2. copyright history | 23:14 |
| jelkner | 3. nfs stuff | 23:14 |
| jelkner | 4. test server | 23:15 |
| jelkner | that's all i have | 23:15 |
| pcardune | sounds good to me | 23:15 |
| dwelsh | 5. check recent improvements | 23:15 |
| jelkner | dwelsh ? | 23:15 |
| jelkner | problems with (4) may make that difficult | 23:15 |
| dwelsh | 6. next mtg. to walk thru CanDo06 progress | 23:15 |
| pcardune | jelkner: i have a test server running on maddog | 23:16 |
| jelkner | good | 23:16 |
| pcardune | jelkner: until bonnieb starts working again | 23:16 |
| dwelsh | yup, but let's get the old girl back on her feet | 23:16 |
| jelkner | i'll look into it tomorrow | 23:16 |
| dwelsh | good; you take the lead jeff | 23:17 |
| jelkner | should we start? | 23:17 |
| dwelsh | yup | 23:17 |
| pcardune | yes | 23:17 |
| jelkner | ok, we started with 4 already, so let's finish up with that | 23:17 |
| jelkner | the machine we are using for bonnieb seems to be flaky | 23:17 |
| dwelsh | need new hardware? | 23:17 |
| jelkner | i'll check it tomorrow, but we've had problems with it for awhile | 23:17 |
| jelkner | yes | 23:17 |
| jelkner | do you have any available? | 23:18 |
| dwelsh | could use the server I gave you | 23:18 |
| jelkner | that's what we are using | 23:18 |
| jelkner | it shuts down and won't come back on | 23:18 |
| dwelsh | what???? | 23:18 |
| jelkner | unless you unplug it, wait, and plug it back in | 23:18 |
| dwelsh | the new server with the prop wash??? | 23:18 |
| jelkner | oh, no | 23:18 |
| dwelsh | I mean the new one | 23:19 |
| jelkner | don't you need that for dhcp? | 23:19 |
| dwelsh | dhcp is holding up for now | 23:19 |
| dwelsh | I'll ask for it back if we need it | 23:19 |
| jelkner | well, then, yes | 23:19 |
| jelkner | i'll use that | 23:19 |
| dwelsh | good | 23:19 |
| jelkner | it would be much better | 23:19 |
| jelkner | ok, done | 23:19 |
| dwelsh | good | 23:19 |
| jelkner | next issue: paul's plane ticket | 23:20 |
| jelkner | he is ready to book his flight here | 23:20 |
| jelkner | can we cover that? | 23:20 |
| dwelsh | does he know which flight??? | 23:20 |
| jelkner | he knows when | 23:20 |
| jelkner | we were waiting for this discussion to make reservations | 23:20 |
| dwelsh | he'll give me a carrier and reservation number | 23:21 |
| dwelsh | and I'll arrange for payment from here! | 23:21 |
| dwelsh | DON | 23:21 |
| jelkner | excellent! | 23:21 |
| dwelsh | DONE | 23:21 |
| jelkner | pcadune: you got that? | 23:21 |
| pcardune | yep | 23:21 |
| pcardune | Seattle -> DC -> Seattle | 23:21 |
| jelkner | since it is more than 30 days, you may not be able to get round trip | 23:22 |
| jelkner | anyway, we'll figure that out later | 23:22 |
| jelkner | ok, 2 | 23:22 |
| jelkner | did you see the emails tom, steve and i sent to each other? | 23:22 |
| *** vmx_ has joined #schooltool | 23:23 | |
| pcardune | jelkner: when? | 23:23 |
| dwelsh | yes; went over them with Mike Morton today | 23:23 |
| dwelsh | our changes are not huge | 23:23 |
| dwelsh | will make changes and send to everyone | 23:23 |
| jelkner | pcardune: you're not on the cc list for that | 23:23 |
| jelkner | i'll send them to you if you would like to see them | 23:23 |
| dwelsh | seems to me we should be printing code for CanDo06 | 23:24 |
| jelkner | basically, we are trying to get our talking points in order regarding copyright for cando | 23:24 |
| dwelsh | and noting what contributions came from the CanDo team | 23:24 |
| jelkner | yes | 23:24 |
| pcardune | jelkner: either way... just as long as I know what to put at the file headers | 23:24 |
| dwelsh | That's what Tom's lawayer basically said | 23:24 |
| jelkner | yes | 23:24 |
| jelkner | th1a: are you here? | 23:24 |
| *** vmx has quit IRC | 23:25 | |
| jelkner | pcardune: you should ask tom about that | 23:25 |
| jelkner | it is a basic GPL license header | 23:25 |
| dwelsh | yes, that's not really an issue or problem | 23:25 |
| dwelsh | the issue has to do with copyright holder to privately license | 23:26 |
| dwelsh | we're the copyright holder of what "we" (CanDo team) wrote | 23:26 |
| dwelsh | we should just document that | 23:26 |
| jelkner | the copyright holder will be APS, yes? | 23:26 |
| dwelsh | Let's just say CanDo team for now | 23:27 |
| jelkner | i don't know if that works | 23:27 |
| jelkner | CanDo team is not a legal entity | 23:27 |
| dwelsh | Right; in that case, APS would be the entity | 23:27 |
| jelkner | that makes more sense | 23:27 |
| dwelsh | They're the ones most of us work for or are students at | 23:27 |
| dwelsh | or were hired by | 23:28 |
| jelkner | so someone who can represent APS should contact the shuttleworth foundation | 23:28 |
| jelkner | to arrange for the contributors agreement | 23:29 |
| dwelsh | correct5 | 23:29 |
| jelkner | that will give us dual copyright on cando | 23:29 |
| dwelsh | we have permission to use the APS legal counsel | 23:29 |
| jelkner | great | 23:29 |
| jelkner | so you are on top of that? | 23:29 |
| jelkner | and i'm done, yes? | 23:29 |
| dwelsh | yes, I'll follow up | 23:29 |
| jelkner | is my email all you need for now? | 23:29 |
| dwelsh | yes; just look at the corrections | 23:30 |
| jelkner | will do | 23:30 |
| jelkner | ok, next topic | 23:30 |
| jelkner | NFS | 23:30 |
| jelkner | take it away, dave | 23:30 |
| dwelsh | would rather hear your version :) | 23:30 |
| jelkner | i thought it went great | 23:30 |
| dwelsh | do tell | 23:30 |
| jelkner | i told wdickers that they were as good as magnets ;-) | 23:31 |
| jelkner | a bunch of important people started heading our way to see who these students were | 23:31 |
| jelkner | and it appeared you got to have a lengthy converstion with the women who had been running the meeting when we arrived | 23:31 |
| jelkner | so i figured getting the students there was a big help for our cause | 23:32 |
| jelkner | they had fun | 23:32 |
| jelkner | that's all i have | 23:32 |
| pcardune | that sounds really good | 23:32 |
| dwelsh | Just to complete the story | 23:33 |
| dwelsh | the lady was as program officer for the Advanced Tech Ed grant program | 23:33 |
| dwelsh | these are very big grants, up to $900K over 3 years | 23:33 |
| dwelsh | she encouraged use to submit CanDo as a project | 23:33 |
| dwelsh | which would mean the exec. summary would go in at the end of this month | 23:34 |
| dwelsh | with a full proposal to go in around mid-October | 23:34 |
| dwelsh | APS has dedicated a grant writer (Ashley) to the project | 23:34 |
| dwelsh | So we couldn't ask for more | 23:34 |
| jelkner | wow! | 23:34 |
| pcardune | incredible | 23:34 |
| dwelsh | Overall, a very successful mtg. with students, teachers, admin, grant writer | 23:34 |
| dwelsh | and even community college reps. who were singing our praises | 23:34 |
| *** srichter has quit IRC | 23:35 | |
| dwelsh | one of the community college guys was on our panel to review the tv program | 23:35 |
| dwelsh | so they knew us and had great things to say | 23:35 |
| dwelsh | anyway, a good showing | 23:35 |
| jelkner | what's not to love? | 23:35 |
| dwelsh | Ashley and I will get working on prospectus tomorrow | 23:35 |
| jelkner | dwelsh: let me know when i could meet with the 2 of you | 23:36 |
| jelkner | i don't want to rush to fast, but i have some other ideas for this summer | 23:36 |
| jelkner | http://www.cs.umd.edu/~jspacco/marmoset/ | 23:37 |
| jelkner | take a quick look at that link | 23:37 |
| pcardune | \me looking | 23:38 |
| * pcardune looking | 23:38 | |
| dwelsh | jelkner: Ashley and I will pull you in to look at drafts | 23:38 |
| jelkner | i spoke at some length with the umd prof who is leading this project | 23:38 |
| jelkner | i've been dreaming of something like it in python for several years already | 23:38 |
| jelkner | our zope3 developers may be able to work on something like this | 23:39 |
| jelkner | and it could eventually become part of the cs curriculum in cando | 23:39 |
| jelkner | especially if we got long term grant funding | 23:40 |
| pcardune | that looks interesting | 23:40 |
| pcardune | very interesting | 23:40 |
| jelkner | indeed | 23:40 |
| *** srichter has joined #schooltool | 23:40 | |
| jelkner | ok, next topic | 23:41 |
| jelkner | we are almost done | 23:41 |
| jelkner | check recent improvements | 23:41 |
| jelkner | i think that is your's mr. carduner | 23:41 |
| pcardune | http://maddog.yhspatriot.net:7080 | 23:41 |
| dwelsh | actually, paul and I will check them Friday 4-6pm | 23:42 |
| dwelsh | same time as last week | 23:42 |
| jelkner | ok | 23:42 |
| pcardune | jelkner: will you be there also? | 23:42 |
| dwelsh | the long walkthru really helped last week | 23:42 |
| jelkner | yes | 23:42 |
| dwelsh | think we could use a couple just to prime the pump | 23:43 |
| dwelsh | we're talking philosophy and features | 23:43 |
| dwelsh | very helpful | 23:43 |
| jelkner | here is what i would like to do about testing | 23:43 |
| jelkner | pcardune: could you develop a list of "zope3 competencies" for the cando course? | 23:43 |
| pcardune | certainly | 23:44 |
| dwelsh | jeff; need to go; talk to you soon | 23:44 |
| jelkner | bye | 23:44 |
| *** dwelsh has left #schooltool | 23:44 | |
| pcardune | will these be on how to use CanDo? | 23:44 |
| pcardune | or on how to us Zope? | 23:44 |
| jelkner | the latter | 23:44 |
| pcardune | certainly | 23:44 |
| jelkner | this is what we are morphing your class into | 23:44 |
| jelkner | we will enrole wdickers, linda, dwoo, rjelliso, and william haddon | 23:45 |
| jelkner | and me | 23:45 |
| jelkner | and flint | 23:45 |
| pcardune | it makes perfect sense | 23:45 |
| jelkner | and tiredbones | 23:45 |
| jelkner | we will learn about cando by using it to track our own competencies | 23:45 |
| jelkner | over time, will add real curriculum that will help folks master zope3 | 23:45 |
| pcardune | where do you want this to run? | 23:45 |
| jelkner | on bonnieb | 23:46 |
| pcardune | ok | 23:46 |
| pcardune | I'll start coming up with the competency lists now | 23:46 |
| jelkner | can you get a rough draft together by friday? | 23:46 |
| pcardune | well.... of how much | 23:46 |
| jelkner | doesn't matter | 23:46 |
| pcardune | there are maybe 1000 zope competencies... | 23:46 |
| pcardune | if not more | 23:46 |
| jelkner | yes, but just the ones you want us to master now | 23:47 |
| pcardune | ok | 23:47 |
| jelkner | you can add to the list as we go | 23:47 |
| jelkner | and we can setup the class this friday | 23:47 |
| jelkner | 4 to 6 | 23:47 |
| pcardune | ok | 23:47 |
| pcardune | sounds good | 23:47 |
| jelkner | great | 23:47 |
| jelkner | sounds like we have everything | 23:47 |
| pcardune | yep | 23:47 |
| jelkner | i'll talk to you on friday | 23:47 |
| jelkner | did you write to whaddon? | 23:47 |
| pcardune | was i supposed to write to him? | 23:48 |
| jelkner | yes | 23:48 |
| jelkner | to give him something to do | 23:48 |
| jelkner | i sent you an email | 23:48 |
| jelkner | it bounced | 23:48 |
| jelkner | (don't know why) | 23:48 |
| pcardune | I don't remember talking about this | 23:48 |
| jelkner | so i forwarded it | 23:48 |
| pcardune | ooh | 23:49 |
| pcardune | let me check | 23:49 |
| jelkner | hold on... | 23:49 |
| pcardune | yeah, i see it now | 23:49 |
| jelkner | good | 23:50 |
| jelkner | i looked at his progress | 23:50 |
| jelkner | he is really moving along | 23:50 |
| jelkner | he needs to be brought into the loop now | 23:50 |
| pcardune | ok | 23:50 |
| pcardune | the timeclock loop? | 23:50 |
| jelkner | hmm | 23:51 |
| jelkner | i think wdickers and woo have that covered | 23:51 |
| jelkner | anything else you could give him to do? | 23:51 |
| jelkner | he is reading and studying on his own | 23:51 |
| pcardune | you mean come up with another zope project? | 23:51 |
| jelkner | hold on | 23:52 |
| jelkner | are you near a phone | 23:52 |
| jelkner | my fingers are getting tired ;-) | 23:52 |
| pcardune | yeah | 23:52 |
| jelkner | regular number? | 23:52 |
| *** jelkner has quit IRC | 23:53 | |
| th1a | Someone ping me? | 23:57 |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!