| *** tpb has joined #melange | 00:00 | |
| *** dmitrig01|afk has quit IRC | 00:03 | |
| *** dmitrig01|afk has joined #melange | 00:03 | |
| *** ChanServ sets mode: +v dmitrig01|afk | 00:03 | |
| *** dmitrig01|afk has quit IRC | 00:04 | |
| *** dmitrig01|afk has joined #melange | 00:07 | |
| *** ChanServ sets mode: +v dmitrig01|afk | 00:07 | |
| *** SRabbelier has joined #melange | 00:14 | |
| *** ChanServ sets mode: +o SRabbelier | 00:14 | |
| *** dmitrig01|afk is now known as dmitrig01 | 00:39 | |
| *** dmitrig01 has quit IRC | 00:50 | |
| *** dmitrig01|afk has joined #melange | 00:50 | |
| *** ChanServ sets mode: +v dmitrig01|afk | 00:50 | |
| *** dmitrig01_ has joined #melange | 00:51 | |
| *** ChanServ sets mode: +v dmitrig01_ | 00:51 | |
| *** dmitrig01_ has quit IRC | 00:52 | |
| *** dmitrig01|afk has quit IRC | 01:09 | |
| *** dmitrig01|afk has joined #melange | 01:15 | |
| *** ChanServ sets mode: +v dmitrig01|afk | 01:15 | |
| *** dmitrig01|afk has quit IRC | 01:17 | |
| *** solydzajs has joined #melange | 01:48 | |
| *** ChanServ sets mode: +o solydzajs | 01:48 | |
| *** florinciu has joined #melange | 03:03 | |
| *** madrazr has joined #melange | 04:32 | |
| *** tansell--laptop has quit IRC | 04:41 | |
| *** florinciu has quit IRC | 04:50 | |
| *** madrazr has left #melange | 05:00 | |
| jamtoday | SRabbelier: Made progress on the GUI, but still having problems on the back-end. I tried using two seperate views as you suggested but it turns out that this would probably be even more problematic | 05:18 |
|---|---|---|
| jamtoday | Or at least that's what I concluded...Now I'm back to using a generic ModelForm class, but it's not recognizing dynamic properties | 05:20 |
| jamtoday | I have a basic spike solution working, but there are a couple big potential problems. db.Expando() is great for dynamic properties, but I'd essentially be making a unique property for each individual survey question on each survey, and that doesn't seem to be a very scalable approach. | 05:24 |
| *** ankitg has quit IRC | 05:26 | |
| *** jbailey has quit IRC | 05:26 | |
| *** sandy|lurk has quit IRC | 05:26 | |
| *** jbailey has joined #melange | 05:29 | |
| *** ankitg has joined #melange | 05:29 | |
| *** sandy|lurk has joined #melange | 05:29 | |
| *** irc.freenode.net sets mode: +o jbailey | 05:29 | |
| *** forever has joined #melange | 05:33 | |
| *** jamtoday has quit IRC | 06:03 | |
| *** jamtoday has joined #melange | 06:47 | |
| *** Merio has joined #melange | 07:01 | |
| *** mithro has quit IRC | 07:04 | |
| *** Merio1 has joined #melange | 07:14 | |
| *** Merio2 has joined #melange | 07:14 | |
| *** Merio has quit IRC | 07:15 | |
| *** Merio1 has quit IRC | 07:15 | |
| *** Merio2 has quit IRC | 07:15 | |
| *** Merio has joined #melange | 07:15 | |
| *** ankitg has quit IRC | 07:23 | |
| *** MatthewWilkes has joined #melange | 07:47 | |
| *** MatthewWilkes has joined #melange | 07:48 | |
| *** solydzajs has quit IRC | 08:16 | |
| *** dr__house has joined #melange | 08:17 | |
| *** veelck_ has quit IRC | 08:52 | |
| *** veelck has joined #melange | 08:52 | |
| *** ankitg has joined #melange | 08:59 | |
| *** forever has quit IRC | 09:04 | |
| *** Merio has quit IRC | 09:07 | |
| *** Merio has joined #melange | 09:31 | |
| *** kblin has quit IRC | 09:32 | |
| *** kblin has joined #melange | 09:32 | |
| *** solydzajs has joined #melange | 10:27 | |
| *** ChanServ sets mode: +o solydzajs | 10:27 | |
| *** dr__house has quit IRC | 10:43 | |
| *** dr__house has joined #melange | 10:47 | |
| *** florinciu has joined #melange | 10:48 | |
| *** veelck_ has joined #melange | 11:07 | |
| *** veelck has quit IRC | 11:08 | |
| *** madrazr has joined #melange | 11:14 | |
| *** madrazr1 has joined #melange | 11:18 | |
| *** madrazr has quit IRC | 11:21 | |
| *** madrazr1 is now known as madrazr | 11:22 | |
| *** dr__house has quit IRC | 11:27 | |
| *** dbentley-laptop has joined #melange | 11:51 | |
| *** dbentley-laptop has left #melange | 11:51 | |
| *** Merio1 has joined #melange | 11:59 | |
| *** Merio has quit IRC | 12:15 | |
| *** Merio has joined #melange | 12:18 | |
| *** Merio1 has quit IRC | 12:19 | |
| *** dr__house has joined #melange | 12:30 | |
| *** dr__house has quit IRC | 13:17 | |
| *** dr__house has joined #melange | 13:17 | |
| SRabbelier | jamtoday: ping | 13:18 |
| *** MatthewWilkes has quit IRC | 13:41 | |
| *** jamtoday has quit IRC | 13:46 | |
| *** dhaun has joined #melange | 13:49 | |
| *** dr__house has quit IRC | 13:50 | |
| *** James--Crook has joined #melange | 14:01 | |
| *** James--Crook has left #melange | 14:05 | |
| *** dr__house has joined #melange | 14:15 | |
| *** ankitg is now known as ankitg|sleeping | 14:30 | |
| *** dhans has joined #melange | 14:36 | |
| Merio | Hi Daniel :) | 14:40 |
| *** sandy|lurk has left #melange | 14:41 | |
| SRabbelier | dhans, Merio: did you two research statistics from previous years yet? | 14:51 |
| Merio | SRabbelier: last year I've found some statistics in google OS blog | 14:52 |
| SRabbelier | dhans: you? | 14:53 |
| *** Lennie has joined #melange | 14:55 | |
| *** ChanServ sets mode: +o Lennie | 14:55 | |
| dr__house | the no power friend greets Lennie | 14:56 |
| Lennie | hi Gregory | 14:56 |
| dr__house | Lennie: heya | 14:57 |
| madrazr | Lennie: Gregory? | 14:57 |
| dhans | Merio: hello :) | 14:57 |
| dr__house | Lennie: pretty odd, nobody usually refers to me from my first name :P | 14:57 |
| dhans | SRabbelier: present:) | 14:57 |
| dr__house | madrazr: Gregory is Dr. House's first name | 14:57 |
| madrazr | dr__house: I must anyways say that we must shut up :P there is a meeting at 00:30 our time | 14:57 |
| dr__house | :P | 14:57 |
| madrazr | dr__house: he he Ok | 14:57 |
| dr__house | madrazr: ouch! didn't know that, sorry | 14:57 |
| madrazr | dr__house: NP actually :P | 14:58 |
| madrazr | but I just told you :) | 14:58 |
| *** James--Crook has joined #melange | 15:00 | |
| SRabbelier | Mhhhh | 15:00 |
| SRabbelier | Let's wait 5 more mins for Pawel | 15:00 |
| solydzajs | I'm back | 15:00 |
| * James--Crook waves to pawel | 15:01 | |
| solydzajs | rest room quickly and brb | 15:01 |
| SRabbelier | solydzajs: wb then :) | 15:01 |
| * SRabbelier will brb as well | 15:02 | |
| James--Crook | Hi Merio! What URL? (last year's stats)? | 15:03 |
| madrazr | I would like to take this time to thank James--Crook | 15:03 |
| madrazr | James--Crook: thanks so much! probably I cannot stop telling that :P | 15:03 |
| madrazr | for all the GHOP related wiki updates! | 15:04 |
| Merio | Hi James--Crook! | 15:04 |
| James--Crook | 'tis the wiki way :-) | 15:04 |
| madrazr | James--Crook: :) | 15:04 |
| James--Crook | Hi Merio. Thanks for the e-mail. | 15:04 |
| James--Crook | Hi Daniel! | 15:04 |
| solydzajs | ok i'm back | 15:04 |
| solydzajs | dhans: ping | 15:04 |
| solydzajs | Merio: ping | 15:04 |
| dhans | James-Crook: Hello:) | 15:05 |
| dhans | solydzajs: pong | 15:05 |
| James--Crook | and we're off...! | 15:05 |
| solydzajs | ok lets wait for Sverre and we are good to go | 15:05 |
| Merio | solydzajs: pong | 15:05 |
| * SRabbelier is back | 15:06 | |
| solydzajs | ok | 15:07 |
| James--Crook | So, what are the decisions that need to be made? | 15:07 |
| SRabbelier | James--Crook: good question! | 15:07 |
| SRabbelier | solydzajs: do we have an agenda? | 15:07 |
| solydzajs | how we are going to gather the stats data, where to store then in what format, how long to cache (we are going to use Jobs for that purpose for sure) | 15:08 |
| solydzajs | probably saving data in some model as JSON object | 15:08 |
| solydzajs | for easy processing later | 15:08 |
| Lennie | we are talking implementation? | 15:09 |
| Lennie | Do the two gentlemen even know what they will be doing :)? | 15:09 |
| SRabbelier | Lennie: +many | 15:09 |
| solydzajs | we need to decide what kind of stats are required | 15:09 |
| SRabbelier | solydzajs: you didn't get around to sending the email to Merio and Daniel, right? | 15:09 |
| solydzajs | like: | 15:09 |
| James--Crook | Yes. And can we add some specific questions, from Daniel, about what the improtant stats are, and a general one about display. Should easily keep us busy for the hour :-) | 15:09 |
| solydzajs | students per country | 15:09 |
| solydzajs | students per school | 15:09 |
| solydzajs | mentors per country | 15:10 |
| solydzajs | amount of mentors/students/org admins | 15:10 |
| SRabbelier | solydzajs: mhhh, can we save the 'what stats do we want to gater' part for later? | 15:10 |
| solydzajs | sure | 15:10 |
| solydzajs | this should go on the wiki | 15:10 |
| SRabbelier | solydzajs: yes, but I think we have other things to discuss first :) | 15:10 |
| Lennie | my plan would be, figure out who is doing what. Then let them write a wiki on what they are going to gather | 15:10 |
| James--Crook | Right, and this one has some discussion of stats too in the comments: http://google-opensource.blogspot.com/2008/04/two-top-10s-for-google-summer-of-code.html | 15:10 |
| solydzajs | sure bring them on then | 15:10 |
| tpb | <http://ln-s.net/36Nx> (at google-opensource.blogspot.com) | 15:10 |
| Lennie | refine on that | 15:10 |
| James--Crook | Lennie: Sounds good. Go for it. | 15:11 |
| solydzajs | http://opentouch.info/tmp/ghop/ghop-stats.html | 15:11 |
| tpb | Title: GHOP Statistics (at opentouch.info) | 15:11 |
| SRabbelier | (ftr, I invited James--Crook to share his view on things, similar to how we had Matthew comment during our earlier meeting on GHOP) | 15:11 |
| SRabbelier | solydzajs: I think it's important that we first get the who-does-what worked out; we've promised to do this for a few weeks now, so let's get that out of the way | 15:13 |
| Merio | SRabbelier: agree ... looking forward to it :) | 15:13 |
| solydzajs | Merio and dhans need to start working on way of gathering the stats data first | 15:14 |
| solydzajs | so we need a wiki page | 15:14 |
| * Merio is probably a little bit lagged... | 15:14 | |
| SRabbelier | solydzajs: mhhh, ok, but we need to be more specific | 15:14 |
| SRabbelier | solydzajs: we should scope that down a bit more I think to say... a few days of work | 15:14 |
| solydzajs | if you want to be more specific then we go into design of how this should work | 15:14 |
| SRabbelier | solydzajs: no, I don't mean implementation details | 15:15 |
| SRabbelier | solydzajs: I mean more like, functional requirements | 15:15 |
| solydzajs | SRabbelier: design is not implementation | 15:15 |
| SRabbelier | solydzajs: what exactly does this part of the system need to do | 15:15 |
| SRabbelier | solydzajs: (not the how should it do it) | 15:15 |
| James--Crook | So, perhaps lats start with stats that don't need the 1000 record solution, if there are any such? Let's say we want to do a repeat of, http://google-opensource.blogspot.com/2008/04/two-top-10s-for-google-summer-of-code.html | 15:16 |
| tpb | <http://ln-s.net/36Nx> (at google-opensource.blogspot.com) | 15:16 |
| solydzajs | James--Crook: 1000 record solution is not a problem here | 15:17 |
| James--Crook | ...go on? | 15:17 |
| SRabbelier | James--Crook: we plan to use Jobs, to batch-process the data | 15:17 |
| *** gchaix has left #melange | 15:17 | |
| SRabbelier | James--Crook: (say, in chunks of 50 entities at a time) | 15:18 |
| dhans | James:afaik we'wll use Jobs | 15:18 |
| solydzajs | ok lets start with a little description | 15:18 |
| SRabbelier | I was thinking about something like | 15:18 |
| solydzajs | we need stats module to display all sort of stats tables, charts and visualizations | 15:18 |
| solydzajs | we need Jobs to gather the data and save that data into some Stats Model | 15:19 |
| SRabbelier | A basic framework for gathering statistics, setting up a system that executes regularly, process a chunk of data, writes it out, lather rinse repeat | 15:19 |
| solydzajs | this is one part of the system | 15:19 |
| *** Merio has quit IRC | 15:19 | |
| *** Merio has joined #melange | 15:19 | |
| solydzajs | those Jobs will basically fetch all the necessary for processing data | 15:19 |
| * SRabbelier resumes | 15:19 | |
| SRabbelier | On the frontend we want a system that can accept a set of statistics, and display it to the user in a relevant way | 15:20 |
| James--Crook | dhans do you have a good picture of what that would entail? Good enough to write something up for it on a wiki page? | 15:20 |
| solydzajs | yes in form of table, chart (using Chart API, so static image) or Visualization API | 15:20 |
| James--Crook | (by that I mean the jobs data collection) | 15:20 |
| SRabbelier | This "relevant way" should be easily configurable (e.g., making it easy to change the type of the visualization), so preferrably not something hard-coded into a page | 15:20 |
| dhans | James--Crook: I read something about the Jobs (it's preety new in GAE), but I think I could do it | 15:21 |
| solydzajs | the stats will be collected on Dashboard pages | 15:21 |
| solydzajs | Org admin/Mentor/Program Admin/Developer dashboards | 15:21 |
| SRabbelier | solydzajs: ping | 15:22 |
| solydzajs | SRabbelier: pong ? | 15:22 |
| SRabbelier | solydzajs: I want to take a higher-level overview first, I feel you're focussing on the details too early already | 15:22 |
| SRabbelier | solydzajs: there's always time to work things out, taking a birds view is something you can't do later on | 15:23 |
| SRabbelier | solydzajs: (well, you can, but usually it's too late to change things by then) | 15:23 |
| solydzajs | SRabbelier: go ahead | 15:23 |
| James--Crook | dhans: so need to decide what the job does and where it puts the results. Could be memcache, could be GAE entity/property. I am not sure. How would we decide? | 15:23 |
| Merio | SRabbelier: do you mean talking about frontend features? | 15:23 |
| SRabbelier | solydzajs: I didn't mean I wanted to take the word, the opposite, I'd prefer to hear your oppinion on this | 15:23 |
| SRabbelier | Merio: no, I mean talking about design first | 15:24 |
| solydzajs | James--Crook: it needs to be in some GAE model and cached too | 15:24 |
| dhans | James-Crook: I thought about GAE entities at first, but solydzajs tells something about some JSON objects | 15:24 |
| James--Crook | so both, no decision needed :-) Cool with that. | 15:24 |
| SRabbelier | dhans, Merio: Pawel and I think it's important that you two work together on part of the system for the first few days (say, up to a week) | 15:24 |
| solydzajs | SRabbelier: no no if you know what you want to say go ahead, I will talk when you finish | 15:24 |
| Merio | SRabbelier: so, I don't know much about the Python end, but we can cache the data on the JS end sending JSON objects, and then switch the visualization as desired with the same data | 15:25 |
| James--Crook | Merio: Nice. | 15:25 |
| SRabbelier | dhans, Merio: so that, after that part of the system is finished, you two won't get into one another's way as much | 15:25 |
| SRabbelier | Merio: that sounds like a good solution | 15:26 |
| SRabbelier | what I want to get clear is a description of this "first week" system | 15:26 |
| SRabbelier | as said, something basic that you two can work from | 15:26 |
| solydzajs | but we also need to use GAE memcache I think | 15:26 |
| solydzajs | what we first need is python module that will use Jobs for fetching all the necesarry data | 15:27 |
| solydzajs | frontend we can work on later | 15:27 |
| SRabbelier | dhans, Merio: from my descripion earlier, do you two understand what I mean with what that system should do? | 15:27 |
| James--Crook | solydzajs: so wiki page should sketch out what we want to store in the memcache... | 15:27 |
| solydzajs | James--Crook: yes | 15:27 |
| Merio | SRabbelier: for the first week you mean something that handles all the roundtrip or just collect some data? | 15:28 |
| SRabbelier | Merio: the former | 15:28 |
| SRabbelier | Merio: something that you both can use and improve on in your code | 15:28 |
| SRabbelier | Merio: without duplicating any work if at all possible :) | 15:28 |
| James--Crook | and as part of testing it we want some front end, and it might be easy to do say a pie chart without too much jscript so it is doable as a spike solution?? | 15:28 |
| SRabbelier | dhans: ping? | 15:29 |
| solydzajs | I think right now the most important it to collect the data save it in some model as JSON | 15:29 |
| dhans | pong | 15:29 |
| solydzajs | and display that data in raw format | 15:29 |
| James--Crook | solydzajs: OK and we can view raw. | 15:29 |
| SRabbelier | dhans: same question to you | 15:29 |
| James--Crook | Yes. | 15:29 |
| solydzajs | later we will worry about charts etc | 15:29 |
| solydzajs | I think we should focus on this backend first | 15:29 |
| Merio | James--Crook: Well... actually I think it's plain easy to get one chart from the JSON object, dont' know about the real complexity on getting that JSON on the Python end | 15:30 |
| solydzajs | since this will be the "core" of the stats module | 15:30 |
| dhans | SRabbelier: I think I understand | 15:30 |
| James--Crook | Merio: Yes, I know. Agree. Focus is still on collection. It's harder, at this stage. | 15:30 |
| SRabbelier | dhans: can you give a short description in your own words? | 15:30 |
| James--Crook | Merio: Later javascript gets harder as we add more sophistication to it. | 15:30 |
| dhans | can I copy/paste what Merio wrote? :P | 15:31 |
| * SRabbelier kicks dhans :P | 15:31 | |
| Merio | James--Crook: I love hard JavaScript ^__^ :P | 15:31 |
| James--Crook | Merio: Just for record we could also have gone down a route with no jobs, and javascript doing multiple requests. (Might still in the end, but it's not the first route into it) | 15:31 |
| dhans | I undestand that what we need to do at first is to collect the data using Jobs and put them in some kind of model | 15:31 |
| dhans | as we just decided it will be some JSON objects | 15:32 |
| SRabbelier | dhans: go on :) | 15:32 |
| James--Crook | dhans: keep talking... | 15:32 |
| solydzajs | James--Crook: it will take too long | 15:32 |
| Merio | solydzajs: also with the async calls as done in the duplicate script? | 15:32 |
| solydzajs | Merio: yes, when you request the stats view you should get it right away that's why we want to use Jobs, memcache etc so it's fast | 15:33 |
| James--Crook | solydzajs: anyway what is decided is we go with the jobs solution first. | 15:33 |
| solydzajs | James--Crook: that will be the only solution we will use | 15:34 |
| solydzajs | James--Crook: I really don't see any other efficient options | 15:34 |
| SRabbelier | dhans: I do would like it if you two can work together on some client side stuff too though | 15:35 |
| SRabbelier | dhans, Merio: perhaps you will want to do a work-together week later on, when you work on the client side the first time? | 15:35 |
| dhans | it's possible | 15:35 |
| Merio | SRabbelier: working on different things (say, dashboard functionalities and charts) or focusing on same thing both of us? | 15:35 |
| SRabbelier | Merio: working on getting the basics down, so you two don't end up doing the same thing twice ;) | 15:36 |
| James--Crook | solydzajs: discussion on wiki page? Jobs will certainly give a faster answer where stats have already been collected and there has to be some cache as you say because we do not want this to happen on every query! | 15:36 |
| SRabbelier | James--Crook: what to happen? | 15:36 |
| Merio | SRabbelier: I think (for the frontend) we can work together on some JS module or object that can do some common work on the data. Furthermore... well the dashboard skeleton should be the same for every type of user | 15:38 |
| solydzajs | SRabbelier: fetching of data from stats model | 15:38 |
| James--Crook | SRabbelier: Multiple slow queries to Melange... | 15:38 |
| SRabbelier | James--Crook: ah, but the queries should be fast | 15:38 |
| SRabbelier | James--Crook: the jobs collect data in the background | 15:38 |
| James--Crook | exactly. | 15:38 |
| SRabbelier | James--Crook: (as in, all the time, even when there's no requests) | 15:38 |
| James--Crook | what solydzajs said. | 15:38 |
| SRabbelier | ah, well that shouldn't be slow | 15:38 |
| James--Crook | hmmm... no I think not... | 15:38 |
| SRabbelier | fetching data from one entity is really fast | 15:38 |
| James--Crook | most of the data is static. | 15:39 |
| James--Crook | solydzajs: opinion... do the jobs keep running collecting data? | 15:39 |
| James--Crook | even when nothing is changing? | 15:39 |
| SRabbelier | James--Crook: we can easily set the interval | 15:39 |
| James--Crook | (easier...) | 15:39 |
| SRabbelier | James--Crook: (they're already runnign atm) | 15:39 |
| Merio | Actually some jobs can be stopped... e.g. when students get accepted... => number of accepted students are fixed that time | 15:39 |
| SRabbelier | James--Crook: (we have 10 pokes each minute) | 15:39 |
| SRabbelier | James--Crook: (it's barely causing any load atm) | 15:40 |
| Merio | So we can have something that starts and stops jobs dinamically depending on which part of the timeline we are into | 15:40 |
| James--Crook | Fine. Ok. | 15:40 |
| James--Crook | (to SRabbelier) | 15:40 |
| SRabbelier | Merio: yes, indeed | 15:40 |
| SRabbelier | dhaun: what's your view on this? | 15:40 |
| SRabbelier | solydzajs: and you? | 15:40 |
| James--Crook | aha... so multiple jobs for different stats then? | 15:40 |
| solydzajs | yes I think there should be some Stats Settings view | 15:41 |
| James--Crook | seems to me simpler to have just one job type that collects all the stats. | 15:41 |
| solydzajs | where you can define what stats you want to collect in what interval | 15:41 |
| SRabbelier | James--Crook: we have different priority groups atm | 15:42 |
| SRabbelier | James--Crook: so that makes it possibel to define various types of stats gatherers, that have different priorities | 15:42 |
| James--Crook | So what is this json object going to llok like, (stop me if this is asking too much detail) | 15:42 |
| Merio | James--Crook.., but then you will come gathering stats that are fixed like, as said formerly, accepted students by country. Waste of work to me | 15:42 |
| dhans | don't we have this restriction that one application may have up to 20 scheduled tasks? | 15:42 |
| SRabbelier | Merio: no, not like that | 15:43 |
| Lennie | SRabbelier: I rather see the prio groups being used for high level purposes ^_^ | 15:43 |
| SRabbelier | Merio: we can create such jobs once | 15:43 |
| Merio | James--Crook: I think it would be a different structure depending on what we'll found useful to iterating through | 15:43 |
| SRabbelier | Merio: and then once they're done, they won't run again | 15:43 |
| Lennie | dhans, we work arround that look into our job system :) | 15:43 |
| dhans | Lennie: ok:) | 15:43 |
| SRabbelier | Merio: but we can have a general "statistics gatherer" job run regularaly, to gather any statistics that are outdated | 15:43 |
| Merio | SRabbelier: sorry, what does mean? | 15:43 |
| SRabbelier | Merio: does what I just said explain? | 15:44 |
| James--Crook | The most extreme option is that we could build up a json object with all the data (minus proposal details) in our sweeps, and put it in memcache. | 15:44 |
| James--Crook | Then we deliver extracts from that. | 15:44 |
| Lennie | dhans, basically we have 10 cron jobs calling the same URL every 1 minute, this URL calls code that fetches a "job" to be done and then executes it. Therefore 1 url can do all kinds off jobs | 15:44 |
| James--Crook | Not sure if that is a good plan or not. | 15:44 |
| Merio | SRabbelier: ah great.. but then we can have that job gathering data depending on the timeline | 15:44 |
| SRabbelier | Merio: correct | 15:44 |
| dhans | lennie: good to know that | 15:44 |
| SRabbelier | James--Crook: I would rather cache the statistics themselves | 15:45 |
| SRabbelier | James--Crook: and keep track of when we last gathered statistics | 15:45 |
| SRabbelier | James--Crook: and then update our records based on what changed | 15:45 |
| James--Crook | So for spike solution, lets decide on some stats. | 15:45 |
| SRabbelier | James--Crook: I am thinking something really simple | 15:45 |
| SRabbelier | James--Crook: like, what we can do in the interactive console in 5 minutes | 15:46 |
| James--Crook | example? | 15:46 |
| SRabbelier | James--Crook: like, how many students there are per country | 15:47 |
| James--Crook | OK. So that's our first target. | 15:48 |
| James--Crook | Do we push a button to set it collecting? | 15:49 |
| Merio | SRabbelier: so you mean focusing on a particular statistic instead of working together on something more general? | 15:49 |
| James--Crook | I see two pages. The dashboard page (set it running) and the results page (raw json) | 15:49 |
| SRabbelier | Merio: nono | 15:50 |
| SRabbelier | Merio: I mean that you should work on the general framework first | 15:50 |
| SRabbelier | Merio: but that to work on that, it might be easy to use an easy stat like that to test it | 15:50 |
| Merio | SRabbelier: ah ok, agree completely | 15:50 |
| SRabbelier | James--Crook: yes, I think we should have some sort of admin interaface there | 15:50 |
| James--Crook | OK. So if we have that one, we can then add applicants-by-country. | 15:50 |
| James--Crook | Then we can have two push buttons, or tick boxes or some such. | 15:51 |
| James--Crook | ...and build on that. | 15:51 |
| SRabbelier | James--Crook: mhhh, I'd be fine with having a simple page with two buttons/urls on it for now :) | 15:52 |
| James--Crook | dhans: Is there enough information so far as to the design of the collection task? | 15:52 |
| James--Crook | Merio: Is there enough information so far as to the design of the pages (and their likely future development)? | 15:52 |
| dhans | James--Crook: I think I get what we want to achieve | 15:52 |
| James--Crook | dhans: Will you write that up on wiki? | 15:53 |
| James--Crook | (after meeting) | 15:53 |
| SRabbelier | solydzajs: you're real quiet, is something bothering you? | 15:53 |
| solydzajs | SRabbelier: no no just reading what you guys write :-) takes time :-) | 15:54 |
| dhans | James: I'll try, but I may end up posting it tomorrow | 15:54 |
| SRabbelier | solydzajs: ok, no worries then, just don't want anyone to feel excluded and all | 15:54 |
| James--Crook | np. | 15:54 |
| Merio | James--Crook: actually I have some other ideas for the design of the pages, so perhaps I can write a wiki page on it to show the ideas, or just write something in a PDF and send to the list... or wel.. just discuss it now :) | 15:54 |
| James--Crook | great! | 15:54 |
| solydzajs | SRabbelier: I think it will be best to post my ideas how this should work on the wiki, we will not achieve anything during this meeting I think | 15:54 |
| SRabbelier | solydzajs: to the contrary | 15:55 |
| solydzajs | Merio: dhans : I suggest you create the wiki page and start thinking about Jobs , data collecting part of the Stats project | 15:55 |
| Merio | solydzajs: agree, we should discuss on something more defined... but however this meeting it's good for brainstorming to me ATM :) | 15:55 |
| SRabbelier | solydzajs: I think we achieved a great deal in getting clear to dhans and Merio what we expect from them the next few weeks | 15:55 |
| solydzajs | SRabbelier: ok cool :-) next few weeks is community bonding period so I expect wiki page from them only cause they need to work on testing too | 15:57 |
| dhans | what's not clear for me: what did we decide? are we going to store the data in json object and then memcache it or not? :) | 15:57 |
| James--Crook | dhans: I understood we memcache it as a json object. | 15:58 |
| James--Crook | solydzajs: check? | 15:58 |
| SRabbelier | dhans: we'll store a JSON string in the datastore containing the statistics | 15:58 |
| Merio | and then served to JS when needed | 15:58 |
| SRabbelier | dhans: I suggested that we also cache that same string | 15:58 |
| dhans | James--Crook: ok, now it makes more sense:) | 15:59 |
| solydzajs | yep :-) we store the JSON string in some stats model and also cache it using memcache | 15:59 |
| Merio | So for the moment just wiki pages and working on testing... perhaps something more detailed to work together on with Daniel will come up from that wiki pages too :P | 16:00 |
| solydzajs | ok | 16:00 |
| James--Crook | sounds good. | 16:00 |
| solydzajs | Merio: dhans : so you are guys good ? you know what you need to put in the wiki page ? | 16:00 |
| Merio | solydzajs: do you mean a stats model that holds the whole JSON string, so? | 16:01 |
| dhans | I'll describe the data gathering process | 16:01 |
| Merio | solydzajs: I was thinking also for some model with timed collected JSON... for the animated graphs if we want some | 16:01 |
| solydzajs | Merio: what kind of animated graphs you are talking about ? | 16:02 |
| James--Crook | Merio: Wiki page for the win! | 16:02 |
| Merio | solydzajs: actually I've talked about them in my proposal. Say proposals submitted by country per day in a world view map... :) | 16:03 |
| James--Crook | Merio: Would be very cool. Section on wikipage for 'animation'? | 16:03 |
| solydzajs | Merio: oh ok got it, yep put your suggestions on the wiki | 16:04 |
| Merio | solydzajs, James--Crook: ok! | 16:04 |
| dhans | Merio: what did you mean by 'timed collected' JSON? | 16:04 |
| Merio | dhans: I mean (actually not expressing myself too well... :P) to say e.g. [JSON object, day] | 16:05 |
| dhans | Merio: clear now :) | 16:05 |
| solydzajs | ok anything else to discuss right now ? | 16:05 |
| Merio | dhans: ok, sorry for the bad former description :P | 16:05 |
| dhans | Merio: np :) | 16:06 |
| Merio | solydzajs: it's ok to work together with dhans on the wiki page, to exchange ideas etc...? | 16:06 |
| Merio | solydzajs: I think it would be great for brainstorming also between us to have a common vision on it | 16:06 |
| SRabbelier | solydzajs: I think we can brainstorm a bit about what kind of stats we want to gather, but that can happen after you close the meeting :) | 16:06 |
| * SRabbelier will be idle for a bit | 16:07 | |
| SRabbelier | feel free to continue talking :) | 16:07 |
| solydzajs | Merio: yes it's ok | 16:07 |
| Merio | solydzajs: great. dhans: so, let's keep us posted by mail :) | 16:08 |
| dhans | Merio: sure :) | 16:08 |
| James--Crook | Thanks everyone. I felt that was very productive. I'll give a hand on the wiki as soon as there is any content there. | 16:09 |
| Lennie | James--Crook, awesome :) | 16:09 |
| dhans | James-Crook: thank you:) | 16:09 |
| *** James--Crook has left #melange | 16:10 | |
| solydzajs | SRabbelier: i'm closing it now :-) | 16:11 |
| solydzajs | bb in 1h | 16:12 |
| Merio | SRabbelier: do we want to talk about the stats.. or ... well.. back to housewarming? :PpP ;) | 16:16 |
| SRabbelier | Merio: go back to your housewarming :) | 16:16 |
| SRabbelier | Merio: we can talk later too | 16:16 |
| Lennie | dhans | 16:16 |
| Lennie | iirc your proposal defined some interesting stats | 16:17 |
| Merio | SRabbelier: well... if it's better now than let's do it now, just to know if go away from the PC or not :P | 16:17 |
| Lennie | maybe worth putting them up on the wiki? | 16:17 |
| SRabbelier | Merio: go ;) | 16:21 |
| SRabbelier | Merio: I'm busy anyway :) | 16:21 |
| Merio | SRabbelier: ok ^__^ ... c u all ^__^ | 16:21 |
| SRabbelier | Merio: have fun :) | 16:21 |
| Merio | SRabbelier: thx :) | 16:22 |
| *** madrazr1 has joined #melange | 16:33 | |
| *** madrazr has quit IRC | 16:34 | |
| *** madrazr1 is now known as madrazr | 16:34 | |
| *** dr__house has quit IRC | 16:35 | |
| *** dr__house has joined #melange | 16:36 | |
| *** dhans has quit IRC | 17:09 | |
| *** dhaun has quit IRC | 17:11 | |
| *** dr__house has quit IRC | 17:16 | |
| *** jameslevy has joined #melange | 17:20 | |
| *** madrazr has left #melange | 17:54 | |
| Lennie | nn | 17:58 |
| *** Lennie has quit IRC | 18:01 | |
| *** solydzajs has quit IRC | 18:34 | |
| *** SRabbelier has quit IRC | 18:37 | |
| *** florinciu has quit IRC | 18:59 | |
| *** Merio has quit IRC | 19:41 | |
| *** dmitrig01|afk has joined #melange | 20:14 | |
| *** ChanServ sets mode: +v dmitrig01|afk | 20:14 | |
| *** r0bby has quit IRC | 20:19 | |
| *** r0bby has joined #melange | 20:20 | |
| *** dmitrig01|afk is now known as dmitrig01 | 20:40 | |
| *** dmitrig01 has quit IRC | 20:48 | |
| *** ankitg|sleeping is now known as ankitg | 20:54 | |
| *** r0bby has quit IRC | 22:13 | |
| *** r0bby has joined #melange | 22:13 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!