Vesuvius April 30, 2013 IRC about GSoC

Listed here because official logging didn't capture this session.

 menu

Status #sahana-meetingX 


(no topic set)


[12:00] == glennp_ [9c287a0c@gateway/web/freenode/ip.156.40.122.12] has joined #sahana-meeting

[12:02] == mayank_jain [~mayank@14.139.82.6] has joined #sahana-meeting

[12:08] <glennp_> Hi, all

[12:08] <Sneha> hi!

[12:08] <mayank_jain> Hi glennp_

[12:09] <glennp_> Looks like a light turnout today, compared to last week's mega-session.

[12:11] <glennp_> That probably means folks have fewer questions & are beavering away on their blueprints & Melange proposals.

[12:11] <glennp_> Or studying for their school exams

[12:12] <mayank_jain> yes.. it seems so ..

[12:14] <glennp_> Any questions or concerns?

[12:14] <Sneha> yes.  i have questions regarding the demo instances.

[12:14] == IsuruMadhu [af9dc362@gateway/web/freenode/ip.175.157.195.98] has joined #sahana-meeting

[12:15] <theVISHESHone> Hey sorry.

[12:15] <rami__> glennp_ let's hope so :D

[12:15] <theVISHESHone> I was away.

[12:15] <theVISHESHone> ok let her finish

[12:16] <Sneha> rami__:  you had asked about the way i wanted to implement sessions in purging the database. can i elaborate it here?

[12:17] <glennp_> Sneha, rami, go for it.

[12:18] <rami__> Sure go ahead

[12:20] == Akila [~akila@112.135.68.208] has joined #sahana-meeting

[12:20] <theVISHESHone> ...

[12:21] <Sneha> when a demo user logs in, his session id will also be stored along with a timestamp. now suppose the trial period is for 60 mins, then during this time he can make any changes he wants to with his instance, he can create a new report search for people etc. or even import his database.

[12:21] <Sneha> then compare the time with every tab or page the user travels

[12:21] <Sneha> or compare with a timer

[12:22] <Sneha> when the time gets over, a query to delete all the newly added data with that session id will fire

[12:23] <rami__> Aha

[12:23] <Sneha> and the database will be restored to its original state

[12:23] == dilantha [7087911e@gateway/web/freenode/ip.112.135.145.30] has joined #sahana-meeting

[12:24] <rami__> Hmmm

[12:24] <rami__> We were thinking more along the lines of

[12:24] <rami__> Maintaining an instace

[12:24] == kaushikaruk [67f732be@gateway/web/freenode/ip.103.247.50.190] has joined #sahana-meeting

[12:24] <rami__> Where users could create accounts

[12:24] <rami__> And report persons etc

[12:25] <kaushikaruk> hii

[12:25] <rami__> Then after a while

[12:25] <rami__> The data would reset automatically based on an arbitrary time value set by the demo instance admin

[12:26] <Sneha> ya so the time will start when a new user will log in right?

[12:26] <rami__> Plus we want to pull in new changes from the latest release of Vesuvius

[12:26] <rami__> Yup that's correct

[12:27] <glennp_> Sneha: To add to what rami said - Sounds too much trouble to me, to track every database change and undo it.  I guess it could be done with MySQL logging or fancy transactioning.  But better to just throw ALL data away and reload known data.

[12:27] <theVISHESHone> can't you just drop the database and import it again?

[12:27] <theVISHESHone> ok its probably the same thing, anyways continue.

[12:28] <glennp_> Yeah, drop the database is one way to delete all data.

[12:28] <glennp_> Easiest, may not be the fastest tho.

[12:29] <Sneha> but if we drop the database altogether then even the ones we don't want to remove will be related. i mean we have to delete the values for that particular user only. so somehow we need to track the user.. this is why i thought of sessions

[12:29] <theVISHESHone> OK, I don't really have the complete picture, but won't it be better if you could just keep the data, and maybe block the user from loading any of the pages?

[12:30] <theVISHESHone> and showing an expiration message instead.

[12:30] <rami__> We're more concerned with the particular instance than the user

[12:31] <rami__> But I guess it's okay to purge data entered by a particular user

[12:31] <rami__> However it would be good to have some data in the db as well

[12:32] <rami__> We have to decide whether the purge is going to work per user session or on a temporal basis

[12:32] <Sneha> ohk..

[12:32] <glennp_> Hmm, is our model 1 user = 1 instance of db, or multiple users per db instance?  I was kinda thinking the first.

[12:33] <theVISHESHone> not my project instance, but I just wasn't clear on one thing, is the demo a demo version of the whole product itself? or a particular user of one of the instances?

[12:33] <kaushikaruk> hi Ramindu

[12:33] <theVISHESHone> I think she is referring to the 2nd one!

[12:33] <Sneha> i thought it was the first one. so i kind of thought along the lines of the session.

[12:34] <theVISHESHone> ok i added some more confusion :P sorry

[12:34] <rami__> Hi kaushikaruk

[12:35] <rami__> 1st one sounds ok

[12:35] <rami__> In which case it will be purge data per user session

[12:36] <kaushikaruk> hi drafted a blueprint and a proposal and i will share them with u, mind u hv a look and gv a feed back

[12:37] <theVISHESHone> ok on a totally different topic, rami__, on the bug tracker, is the entry " Search does not return any results " confirmed a bug or not?

[12:37] <kaushikaruk> as we discussed I looked it to the code and identified the main tasks for a single module merge

[12:39] <theVISHESHone> because mayank_jain did not get the same error, and I don't think there will be any such error in the main nlp instance either.

[12:40] <glennp_> rami: both these questions are for you.

[12:41] <rami__> Vishesh, I guess it's not confirmed yet

[12:41] <Sneha> rami__, glennp_ i have doubts regarding the "pulling changes from vesuvius"

[12:41] <theVISHESHone> and rami__, can you please check for the "Resource Pages Translate view WYSIWYG editor is displayed too low " bug? we ALL submitted patches :P

[12:41] <theVISHESHone> that was the only one left.

[12:41] <rami__> But my instance is still kota showing seat results

[12:41] <rami__> Yeah I was going to check today but LP was down

[12:41] <rami__> In the morning

[12:42] <theVISHESHone> LP?

[12:42] <kaushikaruk> launch pad

[12:42] <theVISHESHone> oh lol ok.. :)

[12:42] <rami__> I will check all the patches tomorrow

[12:42] <rami__> Thanks for your effort

[12:42] <theVISHESHone> Thank you

[12:42] <Sneha> there are so many questions. i think we should form a queue sort of thing and ask questions one by one. :P!

[12:42] <kaushikaruk> I also updated my patch according to ur feedback

[12:43] <theVISHESHone> lol., ok We shall wait till Sneha finishes. :)

[12:43] <rami__> Yup kaushikaruk, I noticed

[12:43] <rami__> Will check soon

[12:43] <rami__> Yeah Sneha go ahead

[12:43] <Akila> - I was able to reproduce the "search results" bug

[12:43] <kaushikaruk> seems sneha not going to finish today.. :P

[12:43] <Sneha> he he! i can wait! i already asked one question.

[12:44] <rami__> No go ahead we'll handle projects one by one

[12:44] <glennp_> Sneha: we probably won't be pushing changes from NLM to Vesuvius during GSoC (except for exceptions :->) , so don't worry about that.

[12:45] <Sneha> okk..

[12:45] <rami__> glennp_, don't you think it might be a good idea to be able to pull in changes from new releases to the demo instance?

[12:45] <Sneha> and the new changes in the codebase? like bug fixes or added features? or even new disaster entries??

[12:46] <rami__> I know exactly aren't doing a lot of releases, but it might be good to have the framework in place

[12:46] <rami__> Nope you don't need to worry about unreleased changes

[12:47] <rami__> Like stuff on the repository

[12:47] <Sneha> ok.

[12:47] <rami__> But you don't need to consider releases, IMHO

[12:47] <rami__> *do need

[12:47] <Sneha> IMHO??

[12:48] <mayank_jain> Sneha: In my humble/honest opinion ;)

[12:48] <rami__> In My Humble Opinion :)

[12:48] <Sneha> ohh!!

[12:49] <Sneha> i know updating the new releases manually is a bad idea. but i couldn't think of any other way.

[12:49] <rami__> Anyway, I think everyone should get their proposals up on Melange soon so that everyone will read and comment on them

[12:49] <rami__> Hmmm...

[12:50] <rami__> We'll have to figure that out

[12:50] <rami__> Or maybe just make the framework so that we can build new demo instances from new releases as well

[12:50] <kaushikaruk> rami: r we going to stop developing in Kilauea after the merge in Vesuvius?

[12:51] <glennp_> I wish releases were easier to get approved here.  Sigh.

[12:51] <rami__> So that would be about making the framework tolerant to change

[12:51] <Sneha> ok.

[12:52] <kaushikaruk> if we have to consider every possible modules to merge with vesuvius

[12:53] <rami__> Sneha, are you done?

[12:53] <rami__> Any other questions?

[12:54] <Sneha> yes. i guess. so should i start drafting my final proposal based on the purging session idea?

[12:55] <glennp_> Works for me.  rami?

[12:56] <Sneha> and i guess we can always include new modules in our demo instance. that will be done manually but obviously it will be done in the same way as in the NLM one.

[12:57] <glennp_> ok

[12:57] <rami__> Yup sounds great

[12:58] <Sneha> ok.

[12:58] <Sneha> i am done now! :D

[12:58] <glennp_> kaushirkaruk, rami: maybe alex (in the chat last week from CUNY) would be interested in co-mentoring the merge, providing guidance on relative importance of modules?

[12:59] <rami__> Yes that sounds good

[12:59] <rami__> We should contact him

[12:59] <rami__> Get him involved as well

[12:59] <kaushikaruk> sounds good

[12:59] <glennp_> I'll drop him a line.

[12:59] <rami__> Anyway, kaushikaruk, how goes the progress?

[13:00] <rami__> Thanks, Glenn!

[13:01] <mayank_jain> can i ask questions now ? :)

[13:02] <kaushikaruk> I tried to merge a module to vesuvius and identify the common steps we need to consider

[13:03] <kaushikaruk> some of them are dependent from other modules so we might need to start with a simple independent module

[13:03] <rami__> Yuo

[13:03] <rami__> *yup

[13:04] <mayank_jain> glennp_: I tried looking for categories into which events can be divided into ... Nothing solid so far..

[13:04] <rami__> What would you suggest?

[13:04] <kaushikaruk> i tried with crs module, but it dependent with fms module

[13:05] <rami__> kaushikaruk, I guess documenting the steps for merging a new module would be the first action items for you

[13:05] <kaushikaruk> In each module there are some core important area to be merged

[13:05] <rami__> How about we go with that for now until we get some feedback from Alex?

[13:06] <mayank_jain> Some basic classification looked the best though..http://www.redcross.org/prepare/disaster... The ones mentioned here.. I don't think there would be any ambiguities in this type of classification

[13:06] <mayank_jain> What do you think about it?

[13:06] <rami__> Ok guys I'm going to have to step out now

[13:07] <theVISHESHone> :O ?

[13:07] <kaushikaruk> as i fell we should consider merging those area rather merging everything at the same time

[13:07] <rami__> I'm not at home, been chatting from my phone

[13:07] <rami__> You can direct any questions to me via email

[13:07] <rami__> Thanks and goodnight

[13:07] <kaushikaruk> of course I drafted a blueprint and I will share it with u

[13:07] <theVISHESHone> good night, bye

[13:08] <kaushikaruk> thanks Ramindu for ur effort

[13:08] <theVISHESHone> glennp_: have you decided already who is going to mentor which project?

[13:08] <kaushikaruk> good nyt and enjoy.. ;)

[13:08] <rami__> Of course, any time

--------------------------------------------------------------------------------

[13:09] == mayank_jain [~mayank@14.139.82.6] has quit [Quit: Leaving]

[13:09] == mayank_jain [~mayank@14.139.82.6] has joined #sahana-meeting

[13:09] <glennp_> mayank_jain: the red cross categories don't look bad

[13:09] <mayank_jain> I am still going through EDXL-SITREP ..

[13:10] <mayank_jain> But it has considered lot of points for classification

[13:10] <mayank_jain> Geography/Severity etc..

[13:10] == kaushikaruk [67f732be@gateway/web/freenode/ip.103.247.50.190] has quit [Quit: Page closed]

[13:10] <glennp_> I did look at SITREP a long time ago, but not with eye to the categories

[13:10] <mayank_jain> Red-Cross one is simpler

[13:11] <mayank_jain> So should i proceed with unless I get something better ?

[13:12] <glennp_> I think it unlikely that we'd want hundreds of categories to have to make templates for, so 10-20 categories should be good.  Yeah, proceed unless you get something better  :->

[13:13] <glennp_> theVISHESHone one:  the final decision of who is the "formal" mentor on each project isn't made until the projects are essentially awarded.  There's horse trading involved.

[13:13] == rami__ [~rami@61.245.163.25] has quit [Quit: Bye]

[13:13] <theVISHESHone> oh lol ok.. :)

[13:13] <mayank_jain> This is the blueprint of the project .. http://www.mediawiki.org/wiki/User:Mayankkrjain  .. Can you please have a look and suggest any changes

[13:14] <mayank_jain> PS: I have hosted it on mediawiki for time being

[13:14] <glennp_> mayank:  I'll do so this afternoon and/or tomorrow morning.

[13:15] <mayank_jain> Ohk.. Will submit the proposal then by tonight.. Will incorporate the changes later :)

[13:16] <glennp_> Probably safest that way.

[13:17] <glennp_> I'll send you my comments probably through melange.

[13:18] <mayank_jain> ok.. Thanks a lot :)

[13:19] <glennp_> I say probably because Melange always seems to have its moods.  I have your email address too.

[13:19] <glennp_> Any other questions, anyone?

[13:20] <mayank_jain> moods as in ?

[13:20] <theVISHESHone> Not questions actually, just if you could give some more clarification on what the installer does? I mean like, first should it check for the OS? then maybe it should check for which AMP stack is installed, etc?

[13:20] <theVISHESHone> and then perform the functions repectively?

[13:21] <theVISHESHone> Ok I need to go through the current implementation for that, just asking.

[13:21] <glennp_> moods: ...hostile, forgetful, braindead ... oh wait, that's me

[13:21] <theVISHESHone> hehe..

[13:21] <triune> I would just take the current installer and try to reverse engineer an answer to that question ;)

[13:22] <theVISHESHone> hmm. cool. i'll try thanks.

[13:22] <mayank_jain> :D

[13:23] <glennp_> thanks, triune.  Thanks, everyone else for joining us today.  Meet again next week, same time, same place.

[13:23] <theVISHESHone> great., thanks .. bye

[13:23] <mayank_jain> Bye glennp_ :)

[13:23] == mayank_jain [~mayank@14.139.82.6] has quit [Quit: Leaving]

[13:23] <glennp_> Bye, all

[13:23] <Akila> I guess we won't able to have a chat with Chamindra before the deadline

[13:24] <Akila> bye

[13:24] <glennp_> Akila: I'll mention that to him as a concern.

[13:24] <Sneha> bye! :)

[13:24] <glennp_> I know Tuesday's is not a easy day for him.

[13:25] <glennp_> In the meantime, work the discuss list and other ways that work for him.

[13:26] <glennp_> Bye, Sneha.

[13:27] <theVISHESHone> bye all

Akila
dilantha
glennp_
IsuruMadhu
lifeeth
rami
Sneha
theVISHESHonetriune