| *** joanna has joined #schooltool | 00:20 | |
| *** joanna has quit IRC | 00:25 | |
| *** joanna has joined #schooltool | 00:26 | |
| joanna | hello | 00:26 |
|---|---|---|
| *** joanna has left #schooltool | 00:27 | |
| FarcePest | buh bye | 00:27 |
| *** gintas has quit IRC | 00:35 | |
| *** FarcePest has quit IRC | 01:44 | |
| *** Velmont has quit IRC | 12:10 | |
| *** bskahan has joined #schooltool | 20:05 | |
| *** predictive has joined #schooltool | 20:22 | |
| predictive | lo | 20:22 |
| th1a | Hello? | 20:22 |
| predictive | I just finished a first cut at implementing a couple of american edu standards | 20:23 |
| predictive | specifically AICC and SCORM | 20:23 |
| predictive | any interest in integrating those? | 20:23 |
| th1a | Implementing them in Python? | 20:24 |
| predictive | I only have a vague idea of your codebase but a friend said I should pop in and ask | 20:24 |
| bskahan | predictive: do you have a web reference for them, I'm not familiar with the acronyms (th1a probably is) | 20:24 |
| predictive | sure, one sec | 20:24 |
| th1a | I don't know AICC. | 20:24 |
| predictive | th1a: they are already in python | 20:24 |
| predictive | http://www.aicc.org/ | 20:25 |
| predictive | AICC is mostly used to let one LMS talk to another LMS to launch courses | 20:25 |
| predictive | and do tracking, etc | 20:25 |
| th1a | OK. | 20:25 |
| predictive | SCORM is a course interchange format | 20:25 |
| th1a | Hm... | 20:25 |
| predictive | that has an API along with it | 20:25 |
| predictive | yes, they are dry, boring, way too long standards | 20:25 |
| th1a | LMS isn't going to be a part of our core functionality. | 20:25 |
| predictive | which means I'm probably one of the very few people ever who have slogged through it in python | 20:26 |
| predictive | ah, I did not know that | 20:26 |
| th1a | But we'd certainly like people to add LMS to SchoolTool. | 20:26 |
| predictive | I have 3/4 of one written | 20:26 |
| bskahan | aicc.org is something about aviation | 20:26 |
| th1a | Is it web based? | 20:26 |
| predictive | among other things, yep | 20:26 |
| th1a | Lots of training needed in aviation. | 20:27 |
| predictive | bskahan: the aviation community created the standard a long time ago | 20:27 |
| th1a | Tracking specialized skills.. | 20:27 |
| predictive | but it's become a general format | 20:27 |
| th1a | Need to get that right. | 20:27 |
| bskahan | ah | 20:27 |
| th1a | Or else... CRASH! | 20:27 |
| predictive | haha | 20:27 |
| predictive | yeah | 20:27 |
| th1a | SCORM is sort of the same thing for the military. | 20:27 |
| bskahan | ah | 20:27 |
| th1a | Or else... BOOM! | 20:28 |
| predictive | th1a: the LMS I'm working on uses twisted | 20:28 |
| predictive | so I have a couple of frontends | 20:28 |
| predictive | like it talks xmlrpc which nobody else seems very interested in atm | 20:28 |
| bskahan | artillery operator: gee, I wish I'd paid attention in Trig | 20:28 |
| th1a | We talk REST. | 20:28 |
| predictive | I've only briefly read about that | 20:28 |
| predictive | it's the um, keep state in a url idea? | 20:28 |
| predictive | is that right? | 20:28 |
| bskahan | I, for one, welcome our new WSGI overloards | 20:29 |
| th1a | Not really. | 20:29 |
| th1a | More like just sending XML documents over HTTP. | 20:29 |
| predictive | well, doesn't seem like a good fit then' | 20:29 |
| predictive | er | 20:29 |
| predictive | haha xmlrpc is sending xml docs over http | 20:29 |
| th1a | I guess one difference is that XMLRPC is 'RPC,' while REST is documents. | 20:30 |
| th1a | I can't remember at the moment if XMLRPC uses regular HTTP verbs. | 20:31 |
| predictive | yep | 20:31 |
| th1a | Somehow I don't feel up to the task of giving a proper description of REST... | 20:31 |
| predictive | that's okay I'm googling | 20:31 |
| th1a | http://www.xml.com/pub/a/2004/08/11/rest.html | 20:32 |
| predictive | how does it deal with non latin charsets? | 20:34 |
| bskahan | xml.com has a new regular column on rest | 20:34 |
| bskahan | I don't think that particular article is from the comlumn though | 20:35 |
| predictive | given the security problems with IDN, though those are really problems of mixing charsets | 20:35 |
| th1a | IDN? | 20:35 |
| predictive | international domain names | 20:36 |
| predictive | you know, domain names in unicode charsets like say, japanese | 20:36 |
| th1a | Right. | 20:36 |
| th1a | REST uses the architecture of the web, so the places the web arch. is weak, REST web services are weak. | 20:36 |
| predictive | yeah, I'm just asking, not um picking | 20:37 |
| predictive | every technology has weak spots | 20:37 |
| th1a | Sam Ruby has a sobering presentation on the subject. | 20:37 |
| th1a | Basically, we're approaching interop more as scriptability at this point than supporting the big enterprise specifications. | 20:38 |
| predictive | so schooltool is really aimed at the administrative portions of education? | 20:38 |
| th1a | Because our primary motivation is philanthopic, schools that currently have no administrative software. | 20:38 |
| predictive | righto | 20:38 |
| th1a | Or just crappy, expensive. | 20:38 |
| predictive | well, sorry I couldn't be more helpful | 20:38 |
| predictive | yeah | 20:39 |
| predictive | haha I do most of my work in the USA edu market, and there's a lot of that | 20:39 |
| th1a | Well, in the long run our projects should work together nicely, actually. | 20:39 |
| predictive | I hope so | 20:39 |
| th1a | Is your project open source? | 20:40 |
| predictive | I'm waiting to finish it before I go to the trouble of releasing code | 20:40 |
| predictive | lest I be struck by the curse of sourceforge | 20:40 |
| predictive | 10,000 abandoned projects | 20:40 |
| predictive | it will be, BSD licensed | 20:40 |
| th1a | Cool. | 20:41 |
| predictive | I see | 20:41 |
| predictive | you know AICC works exactly this way except it uses ini type structure instead of xml | 20:41 |
| th1a | Up to this point, we've been focusing on calendaring, as a way to get our foot in the door of schools. | 20:41 |
| predictive | that's a pretty good problem | 20:41 |
| predictive | scheduling is hard | 20:41 |
| predictive | and no cal spec has won yet | 20:42 |
| predictive | hard to know what to target | 20:42 |
| bskahan | icalendar would win if outlook supported it | 20:42 |
| th1a | Yeah, and calendaring isn't quite as mission critical as most of the rest of the stuff we'll be doing: grades, attendance, etc. | 20:42 |
| th1a | Hopefully CalDAV will come through. | 20:42 |
| th1a | That seems most promising. | 20:42 |
| * predictive googles | 20:42 | |
| predictive | be nice if there was some RSS functionality to that | 20:43 |
| predictive | or something like it | 20:43 |
| th1a | Anyway, we're just getting warmed up to start dealing with curriculum and assessment. | 20:43 |
| bskahan | to what? | 20:43 |
| predictive | that caldav | 20:43 |
| predictive | unless it is and is just several pages down | 20:44 |
| predictive | s/it/there/ | 20:44 |
| predictive | what are you guys going to use for a interchange format generally | 20:44 |
| predictive | or are you | 20:44 |
| th1a | Once we get our curriculum model set up, we can look at how it can be tied into your stuff. | 20:44 |
| predictive | there's a standard here called SIF that addresses letting different pieces of software like schedulers, curriculum, etc talk to each other | 20:45 |
| th1a | I know SIF well. | 20:45 |
| th1a | We may do that someday... | 20:45 |
| predictive | are you going that way, like writing a ZIS? | 20:45 |
| predictive | yeah adam and I (my friend) have been playing with a zis prototype | 20:45 |
| th1a | We would be an agent, I guess, or we'd have to write an agent. | 20:46 |
| predictive | basic idea is easy, it's the hugenormous data model that sucks | 20:46 |
| th1a | A Python zis? | 20:46 |
| predictive | yea | 20:46 |
| th1a | I'd like to see that. | 20:46 |
| th1a | Twisted? | 20:46 |
| predictive | yep | 20:46 |
| th1a | Ooh. | 20:46 |
| predictive | it's like made for twisted | 20:46 |
| predictive | event based spec | 20:46 |
| th1a | Can I see it? | 20:47 |
| predictive | I'll ask adam, he wrote it for his school so we'd have to get permission from them | 20:47 |
| predictive | but I'll find out | 20:47 |
| th1a | Does it actually work? | 20:47 |
| predictive | it's a teeny prototype | 20:47 |
| predictive | nothing to get excited over | 20:47 |
| predictive | figuring out the interesting bits | 20:47 |
| predictive | for instance | 20:48 |
| th1a | Well, I'm curious. | 20:48 |
| predictive | scalability over a large deployment | 20:48 |
| th1a | I went to a couple of SIF meetings a couple of years ago. | 20:48 |
| predictive | twisted's pb looks like a potential solution | 20:48 |
| predictive | yea I've looked at like parts of the spec that are interesting | 20:48 |
| predictive | seems overdone like most edu specs I've seen | 20:49 |
| th1a | Enough to start understanding it, but then we decided it wasn't sufficiently relevant to the project I was working on. | 20:49 |
| predictive | well if you're not in north america it's pretty useless | 20:49 |
| th1a | Having worked on part of it, it seems impossible to not overdo that kind of spec. | 20:49 |
| predictive | anyway I asked because it would be interesting to make a similar iinternationally oriented spec | 20:49 |
| predictive | oh you did? | 20:49 |
| predictive | neat | 20:50 |
| predictive | novell is integrating zis functionality into dirxml too | 20:50 |
| th1a | Well, I sat at the table for several hours, nodded my head thoughtfully, and made a couple of suggestions. | 20:50 |
| predictive | someday | 20:50 |
| predictive | haha | 20:50 |
| th1a | Sat there thinking "this should all be RDF." | 20:50 |
| predictive | but RDF is all pull, whereas if I read the bits I read correctly | 20:51 |
| predictive | it's more of a async notification deal where the zis is responsible for the managing of them | 20:51 |
| predictive | I dunno RDF well though, I could be totally wrong | 20:52 |
| th1a | If you're thinking about the data and not how you move it around, you'd want it to be RDF. | 20:52 |
| th1a | I need to take my wife to the beach. | 20:52 |
| predictive | good plan | 20:52 |
| predictive | well, great talking to you guys | 20:52 |
| th1a | predictive: stay in touch. let me know what's up with the zis. | 20:53 |
| predictive | I'll pop in again when I'm close to having a useful deal to dl | 20:53 |
| th1a | Cool. Have a good one. | 20:53 |
| predictive | later | 20:53 |
| *** predictive has quit IRC | 20:53 | |
| bskahan | have fun at the beach | 20:53 |
| th1a | I'll try. | 20:54 |
| th1a | I think I should buy a skimboard. | 20:54 |
| * bskahan is a big fan of skimboards | 20:54 | |
| bskahan | for about 30 minutes | 20:54 |
| th1a | 30 minutes once or 30 minutes periodically. | 20:55 |
| bskahan | 30 minutes periodically | 20:55 |
| bskahan | vs. surfing which is entertaining for hours at a time | 20:56 |
| bskahan | damn, was trying to remember where this was | 20:56 |
| bskahan | http://en.wikipedia.org/wiki/Representational_State_Transfer#REST_versus_RPC | 20:56 |
| bskahan | that's the simplest rest vs. xmlrpc I've found | 20:57 |
| *** bskahan has quit IRC | 22:13 | |
| *** bskahan has joined #schooltool | 23:40 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!