Wiki Spaces


Get Help from Others

Q&A: Ask OpenMRS »
Discussion: OpenMRS Talk »
Real-Time: IRC Chat

IRC Logs

Child pages
  • 2008-07-17 - OpenMRS
Skip to end of metadata
Go to start of metadata
02:10:16 *** james_regen has joined #openmrs
02:10:16 *** ChanServ sets mode: +v james_regen
02:34:58 *** sprasanna has joined #openmrs
02:38:37 *** sioraiocht has joined #openmrs
03:45:44 *** Nzeyi has joined #openmrs
03:54:59 *** Nzeyi has quit IRC
03:56:04 *** TorLye has quit IRC
04:07:55 *** sprasann1 has joined #openmrs
04:08:35 *** TorLye has joined #openmrs
04:08:38 *** docpaul has joined #openmrs
04:08:38 *** ChanServ sets mode: +o docpaul
04:11:39 *** sherazfn has joined #openmrs
04:13:57 <sherazfn> Sheraz F Noormohammad. Hi. Just to get myself introduced briefly: I did my medical school in India after which I was a practicing intnernist there before joining IU and Regenstrief as a Health Informatics graduate student and a fellow.
04:14:51 *** docpaul_ has joined #openmrs
04:14:53 <sherazfn> My current interested in OpemMRS is that I would like to implement and evaluate the ability to add clinical reminders to the paper summary.
04:16:58 *** atomicturtle has quit IRC
04:18:45 <nribeka> wow ... wellcome sherazfn
04:18:52 <nribeka> i think everyone is in the devs call right now
04:23:58 *** jmiranda has joined #openmrs
04:23:58 *** ChanServ sets mode: +o jmiranda
04:24:03 *** jmiranda_ has joined #openmrs
04:25:05 <jmiranda> docpaul: tell burke the tool is called bal-sa-mic
04:25:36 <docpaul_> oh is it? like the vinegar? :)
04:25:39 <jmiranda> exactly
04:25:59 <jmiranda> it's spelled "balsamiq"
04:26:33 *** sprasanna has quit IRC
04:26:35 *** Nzeyi has joined #openmrs
04:27:59 <TorLye>
04:28:29 *** burke_away has joined #openmrs
04:28:29 *** ChanServ sets mode: +o burke_away
04:28:33 *** bmckown has joined #openmrs
04:28:33 *** ChanServ sets mode: +o bmckown
04:37:10 *** docpaul has quit IRC
04:37:58 *** jmiranda_ has left #openmrs
04:37:59 *** docpaul_ is now known as docpaul
04:38:00 *** ChanServ sets mode: +o docpaul
04:54:41 *** sherazfn has quit IRC
04:56:33 *** shadowdoc has joined #openmrs
04:57:01 *** shadowdoc is now known as shadowdoc_away
04:57:45 *** sherazfn has joined #openmrs
05:01:31 <Nzeyi> Hi folks
05:03:53 <nribeka> hi Nzeyi
05:04:06 <nribeka> everyone is on the devs call right now
05:04:07 <docpaul> hi Antoine. :)
05:04:14 <Nzeyi> I was unable to join the conf call
05:04:18 <Nzeyi> any help ?
05:04:21 <docpaul> Maros talked about your work earlier.
05:04:33 <nribeka> shri is explaining his project right now
05:04:35 <docpaul> Justin might be able to add you with skype
05:04:54 <Nzeyi> Ok hope it will work
05:04:55 <jmiranda> jmiranda62 is my skype id
05:05:21 <Nzeyi> let me add it to my list
05:07:34 <nribeka> jmiranda, i'm adding you to my list too
05:07:39 <jmiranda> just got it
05:08:05 <Phantal> jmiranda, welcome back :)
05:08:25 <jmiranda> nribeka, do you want to be added to the call too?
05:09:06 <nribeka> i'm in the dev call
05:09:14 <Phantal> jmiranda, so, on my question yesterday, what I was trying to determine is whether there exists either ETL scripts to generate an appropriate schema, or a schema already constructed
05:09:17 <jmiranda> hey Phantal
05:09:22 <nribeka> using my cell
05:09:30 <jmiranda> Phantal, yes
05:09:39 <jmiranda> Phantal, i was going to get back to you today
05:09:44 <Phantal> oh, cool
05:10:02 <jmiranda> the etl scripts are in the code on the google code site
05:10:16 <Phantal> Reason I ask is, Tush has a learning curve ahead of him for learning Kettle, and I'm swamped with work otherwise I'd try to help more with the ETL work he needs to do, so I figured I could get started with the existing stuff and get a mondrian schema put together for you
05:10:18 <jmiranda> the same zip files i referenced in the email to tush last week
05:10:28 <docpaul> guys
05:10:29 <Phantal> alright, i'll check those out
05:10:38 <docpaul> who is working on ETL stuff?
05:10:42 <jmiranda> docpaul, the typing might be me ... but i'm muted
05:10:44 <docpaul> sorry, wasn't following
05:11:02 <Phantal> Is the call you guys are doing something I should be apart of?
05:11:05 <jmiranda> so i have no idea how that's working
05:11:12 <Phantal> I'm in the middle of a training session
05:11:15 <jmiranda> Phantal, it's a Google summer of code thing
05:11:25 <Phantal> alright
05:11:25 <jmiranda> not a data warehouse thing
05:11:32 <docpaul> the lead developer of pentaho (the chief geek) is 1/3 of the way through a real "pentaho approved" prototype
05:11:39 <Phantal> docpaul, Tushar G<something> is doing ETL for the OMRS dwh project
05:11:41 <jmiranda> very cool
05:11:42 <docpaul> we just got the design document last night
05:11:48 <jmiranda> could you give us an update on that?
05:11:52 <docpaul> and he's done an outstanding job
05:11:54 <jmiranda> via email
05:11:57 <docpaul> sure
05:12:11 <docpaul> phantal, tell me more
05:12:23 <jmiranda> yeah, we have a project in which a student in the UK is doing some work for his thesis
05:12:25 <docpaul> who is tushar g?
05:12:35 *** njero has quit IRC
05:12:36 <jmiranda> he's the student i just spoke about
05:12:44 <Phantal> docpaul, Justin knows more abotu him than I
05:12:51 <docpaul> ok
05:13:12 <Phantal> docpaul, I was going to do the Mondrian schema when he got the underlying oltp schema put together
05:13:26 <docpaul> of course it's great that he's working on it, but hopefully he can gain some economies by working alongside james and the pentaho folks
05:13:43 <jmiranda> docpaul, yeah, i was hoping that would happen at some point
05:13:44 <docpaul> phantal: what's your email address?
05:13:46 <Phantal> docpaul, Who is doing what you were just talking about? Julian? Matt?
05:14:01 <Phantal> docpaul, you can reach me either at, or
05:14:04 <docpaul> James Dixon, chief geek at Pentaho
05:14:16 <Phantal> Oh, I don't know james at all
05:14:26 <docpaul> well, let me send you both an email... sec
05:14:30 <jmiranda> Phantal, dmoran is also probably a part of it
05:14:39 <docpaul> doug definitely is
05:14:42 <Phantal> doug i do know, he hangs out in #Pentaho sometimes
05:14:43 <jmiranda> at least at a high level
05:14:54 <docpaul> james is the lead developer at pentaho
05:14:56 <jmiranda> Phantal, yeah exactly
05:15:21 <Phantal> bah
05:15:50 <docpaul> sent... check this out. :)
05:16:05 <Phantal> I'm in this training and the trainers ran into some issues. They got it resolved, but the extra time spent is going to force me to take the 10am train instead of 9:30
05:16:17 <docpaul> this is a generic solution for all openmrs installations
05:17:04 <Phantal> speaking of dwh stuff, I came up with a new technique for dealing with some things that should come in handy
05:17:52 <docpaul> phantal: did you get my email?
05:18:06 <docpaul> justin?
05:18:14 <Phantal> It's a little in-depth, so I won't bore you with it right now, but anyway ...
05:18:41 *** njero has joined #openmrs
05:18:41 *** ChanServ sets mode: +v njero
05:19:48 <jmiranda> docpaul, yeah i just got it
05:20:02 <Phantal> docpaul, thank you. Just got it, it's open on my work machine now, but I'm VPNd, in, tunneling remote desktop through two servers, so I'm not going to try to read it until I ge tinto the office
05:20:04 <docpaul> cool, eager to hear your thoughts
05:20:10 <jmiranda> this is cool
05:20:24 <docpaul> i hope you dont mind me driving this a little bit
05:20:40 <jmiranda> docpaul, btw tushar approached me at the beginning of the summer
05:20:45 <Phantal> docpaul, not at all. ETL is absolutely the most time consuming part of the construction phase of a warehouse
05:20:59 <Phantal> Writing the mondrian schema that sits on top of it is a walk in the park by way of comparison
05:21:20 <jmiranda> and wanted to get him (and brian, haza) involved in the pentaho-run project
05:22:12 <Phantal> jmiranda, i would highly recommend you keep your eye on postgres, and a product called 'everest'
05:22:19 <docpaul> there were some specific constraints related to openmrs that i wanted to make sure we accounted for
05:22:29 <jmiranda> but i didn't want there to be too many cooks in the kitchen
05:22:41 <Phantal> everest is a custom postgres database engine that's column-orientd, but built by Yahoo
05:22:41 <jmiranda> stirring the kettle
05:22:43 <docpaul> 1) it's up to the specific implementation as to what concepts are measured and stored
05:22:51 <Phantal> Yahoo hasn't decided whether to release it back to the postgres community or not
05:22:53 <jmiranda> i've heard of everest actually
05:23:15 <Phantal> My experience thus far with warehousing is, if you have a lot of data, you need a beefy machine to handle the load
05:23:25 <docpaul> because we've normalized clinical data away from the database design
05:23:35 <Phantal> but if you can get a column-oriented database to back your warehouse, it's able to handle the load better
05:23:53 <docpaul> isn't that some of the magic behind ROLAP?
05:24:08 <docpaul> optimized column behavior?
05:24:17 <Phantal> well, lemme give a simple example to illustrate what ROLAP does
05:24:21 <Phantal> here's a very simple MDX query
05:24:49 <Phantal> select {[Measures].[Sales Amount]} on columns, {[Products].[All Products].children} on rows from [sales cube]
05:24:56 <Phantal> That would be translated into a couple of SQL queries
05:25:02 <docpaul> ah
05:25:22 <Phantal> select sum(sales_amount), productid from sales_facts group by productid;
05:25:48 *** sgrannis has joined #openmrs
05:26:00 *** ChanServ sets mode: +v sgrannis
05:26:07 <Phantal> select productid, product_name from product_dimension, sales_facts <appropriate stuff to join the tables)
05:26:14 <sgrannis> nribkea: noyomna - you around?
05:26:25 <Phantal> Then the resultset is put together by comparing the totals from the first query with the product names & ids in the 2nd
05:26:33 <Phantal> and there may be an extra query or two to add in other important details
05:27:06 <Phantal> So, all ROLAP really does is translate MDX to SQL, get resultsets, then cache the resultsets for future use to avoid hitting the OLTP db again with the same (or similar) queries
05:28:01 <Phantal> As far as speed/optimizations go, using something like Vertica, Brighthouse, Everest, or MonetDB would improve things drastically over something like Mysql, postgres, oracle, etc
05:28:18 <docpaul> so rolap is database independent?
05:28:26 <Phantal> I'd actually suggest we use monetdb if it weren't buggy, and limited to in-memory storage space
05:28:33 <docpaul> i thought it was related to a specific db codebase
05:28:43 <Phantal> docpaul, to the extent that the underlying db supports the SQL mondrian generates
05:28:51 *** njero_ has joined #openmrs
05:28:55 <docpaul> phantal: have you joined our developer's list?
05:28:59 *** njero has quit IRC
05:29:10 <docpaul> we need your brain and insight as we evolve the DW concept
05:29:18 <Phantal> nope, not yet
05:29:34 <jmiranda> docpaul, are the pentaho guys lurking on the dev list?
05:29:47 <Phantal> I've been so busy with things going on at work, home, and school starting again in 6 weeks, I've hardly had time to spend any time on this project
05:29:58 <docpaul> they will be soon. we're just trying to get a strawman realized, and then we'll share it with the team
05:30:05 <jmiranda> i created a google data warehouse group that is being used as a "dump" of all our materials
05:30:31 <docpaul> we've given them a 28 million obs dataset, a sample report, and some design specifications
05:30:41 <docpaul> this document was the first pass at the general solution
05:30:42 <jmiranda> i saw that
05:31:07 <jmiranda> i'm very excited to see a first draft of the schema
05:31:26 <docpaul> yeah, i am too... b/c they've proved that my dream scenario is possible
05:31:54 <docpaul> we can get to a place where a report builder simply chooses the data points of interest, and then a schema is built on the fly
05:32:24 <docpaul> then the "slicing and dicing" tools can help build reports
05:32:39 *** sherazfn has quit IRC
05:32:40 <docpaul> they're going to prototype the PEPFAR form just as a proof of concept
05:32:57 <docpaul> er, PEPFAR report
05:33:30 <docpaul> the dream of course is that they can use the exisiting report tools to get started, but as the implementation grows, they can layer on this DWH option for scale
05:35:04 <jmiranda> yeah, that's what i've been saying for years ... we (read: james arbaugh) needs the DW option
05:35:06 <jmiranda> glad to see this moving forward
05:35:10 <docpaul> phantal: understand how life is... perhaps now that Pentaho themselves are doing a lot of the heavy lifting, that there will be opportunities for you to contribute incrementally vs. conceptualizing the entire enchilada
05:35:22 <sprasann1> is that ok if I leave the call now? I gotta meet my doctor now.
05:35:30 <docpaul> sri: of course... thank you!
05:36:08 <sprasann1> docpaul: thanks a lot paul... seeing you in the channel after a very longtime :)
05:36:22 *** sprasann1 is now known as sprasanna_away
05:36:23 <docpaul> have a good doc visit! :)
05:36:28 <jmiranda> docpaul, one thing i would like to see more of, could we start to get more developers into the conversation
05:36:45 <jmiranda> i'm guessing now is not the right time since we'd like to let pentaho actually get something done
05:36:45 *** sherazfn has joined #openmrs
05:36:51 <docpaul> of course... i just wanted to give them the space to get focused and prove that the prototype works
05:37:01 <jmiranda> before adding noise to the spec/design of the schema
05:37:21 *** sherazfn has quit IRC
05:37:23 <docpaul> they want to do all this PR and stuff, and I think we need to crawl before we walk.... i'm holding them to getting something real instantiated
05:37:56 <jmiranda> yeah, completely agree
05:42:08 <jmiranda> Phantal, did you check out the docs that tush uploaded yesterday?
05:47:46 *** njero_ is now known as njero
05:49:05 <nribeka> yayyy ... hope my english is ok
05:49:16 <jmiranda> :)
05:49:36 <nribeka> :)
05:49:47 <TorLye> wow, everyone survived
05:49:55 *** rfoecking has joined #openmrs
05:50:25 *** bmckown has quit IRC
05:51:39 *** Nzeyi has quit IRC
05:54:45 *** TorLye has left #openmrs
05:57:22 *** jmiranda has quit IRC
05:58:28 *** TorLye has joined #openmrs
06:03:30 *** TorLye_ has joined #openmrs
06:07:03 *** bmckown has joined #openmrs
06:07:03 *** ChanServ sets mode: +o bmckown
06:10:02 *** sprasanna has joined #openmrs
06:13:02 *** atomicturtle has joined #openmrs
06:19:54 *** TorLye has quit IRC
06:20:32 *** sprasanna_away has quit IRC
06:21:51 *** docpaul has quit IRC
06:26:34 *** burke_away has quit IRC
06:34:55 *** sprasanna has left #openmrs
06:44:37 <r0bby> meh
06:44:48 *** nribeka is now known as nribeka_away
06:54:59 *** njero has quit IRC
07:07:45 <OpenMRSBot> Recent updates in the world of openmrs: OpenMRS Tickets: Ticket #915 (task closed): PersonService.getPeople(String, boolean) returns duplicate Person objects <> || OpenMRS Tickets: Ticket #915 (task created): PersonService.getPeople(String, boolean) returns duplicate Person objects <>
07:16:00 *** [mharrison] has quit IRC
07:19:56 <r0bby> TorLye_: sorry i missed your presentation :(
07:21:33 *** njero has joined #openmrs
07:21:33 *** ChanServ sets mode: +v njero
07:23:07 <TorLye_> r0bby: aw, too bad... everyone had cake! and balloons!
07:23:52 *** TorLye_ is now known as TorLye
07:27:59 <r0bby> the cake is a lie
07:28:03 <r0bby> it always is \
07:28:15 <r0bby> i blame shit sleep habits
07:39:10 <TorLye> you did not assume the party escort submission position :)
07:39:30 <r0bby> that is also a lie
07:43:40 <TorLye> really? you, <subject name here> must be the pride of <subject home town here>
07:52:35 <TorLye> sorry, when someone is quoting Portal, I can't help myself :)
08:32:17 *** james_regen has quit IRC
08:55:44 *** shadowdoc has joined #openmrs
08:55:54 *** james_regen has joined #openmrs
08:55:54 *** ChanServ sets mode: +v james_regen
08:57:44 * r0bby dances
09:00:45 *** jbrown has joined #openmrs
09:03:16 * r0bby throws his shoe at jbrown to welcome to him
09:03:17 <r0bby> :)
09:03:19 <r0bby> Welcome :>
09:03:31 <r0bby> Welcome to the channel, i'm the hyperactive something
09:03:32 <jbrown> thank you
09:03:35 <r0bby> Also a soc student
09:03:36 <r0bby> :>
09:12:34 *** burke_away has joined #openmrs
09:12:34 *** ChanServ sets mode: +o burke_away
09:13:54 *** shadowdoc_away has quit IRC
09:13:54 *** shadowdoc is now known as shadowdoc_away
09:26:35 *** shadowdoc_away is now known as shadowdoc_
09:27:14 *** shadowdoc_ is now known as shadowdoc_away
09:29:12 *** shadowdoc_away is now known as shadowdoc
09:37:28 *** ekrub has joined #openmrs
09:37:28 *** ChanServ sets mode: +o ekrub
09:43:34 <r0bby> oh mai
09:43:43 <r0bby> you can run but you can't hide
09:43:53 * r0bby snickers
09:44:27 <r0bby> sink full of dishes -> template processing -> controller validation -> view generation
09:44:41 <r0bby> documentation time
09:45:04 <r0bby> ekrub: did you see my email to dev@
09:45:50 <r0bby> I decided to not use the template engine -- it introduces complexity that's not worth it; instead i'll just do it the way i'm doing it
09:46:15 <r0bby> it seems half-baked, but while you were AWOL i had to come up w/ something :P
09:47:02 <r0bby> so far, i make 4 AJAX calls in succession only upon successful completion of the previous
09:47:21 <r0bby> we can move away from that when DWR is upgraded and I can do it in one DWR method :)
09:47:27 <r0bby> from a servlet :>
09:47:38 * r0bby guilt-trips
09:47:50 * r0bby brb
09:54:07 *** burke_away has quit IRC
09:55:18 <r0bby> ekrub: whenever you have a chance, i'll open up the port and let you see what its like so you don't have to go the trouble of deploying
09:56:06 <r0bby> also weird problem where i get some JS syntax errors when i first deploy the module; but when i do the deploy-web task the errors go away w/o any changes to the JSP
09:56:10 <ekrub> r0bby: cule. sorry. in a meeting at the moment. looking forward to checking it out
09:56:13 <r0bby> not sure what that is
09:56:36 <r0bby> i'll let you be
09:56:48 <r0bby> apologies
09:59:40 *** bmckown has quit IRC
10:00:05 *** bmckown has joined #openmrs
10:00:05 *** ChanServ sets mode: +o bmckown
10:06:38 *** jbrown has left #openmrs
10:13:34 *** james_regen has quit IRC
10:17:25 * r0bby throws his koosh man at bmckown
10:17:34 <r0bby> I still have that
10:18:33 <r0bby> man i miss my koosh :<
10:19:19 <bmckown> ?
10:20:12 <r0bby> you ever hear of koosh balls :P
10:20:13 <bmckown> r0bby has bad aim.
10:20:23 <r0bby> who'd I hit?
10:20:28 <bmckown> no i haven't.
10:20:35 <r0bby> aw come on
10:20:40 <r0bby> you're older than I am!
10:20:42 <bmckown> hit your monitor
10:21:16 <r0bby>
10:21:22 <r0bby> these things were neat
10:21:39 <r0bby> I had a koosh sling rosie o'donnell had on her show :>
10:21:47 <r0bby> that thing was neat to play w/ :)
10:22:29 * bmckown interrupts more important items of the day to search what koosh balls are
10:22:41 <bmckown> ah yes. I just did not know their name
10:22:43 <r0bby> i just gave you a wikipedia article on it!
10:22:58 <bmckown> i wasn't looking because i was on google
10:23:34 <bmckown> yes saw the wikipedia article.
10:23:40 <bmckown> hmm
10:23:48 * bmckown goes back to work
10:24:42 * bmckown looks one more time at wikipedia article...
10:25:07 <bmckown> "developed in 1986" eh... I was too old for them in 1986.
10:25:16 *** shadowdoc has quit IRC
10:25:30 * bmckown really goes back to work
10:30:41 *** shadowdoc has joined #openmrs
10:31:57 * r0bby wasn't
10:32:07 <r0bby> I was i like 6 :>
10:32:14 <r0bby> er no 4
10:34:23 *** rfoecking has quit IRC
10:40:22 <bmckown> wow
10:42:54 * r0bby is 26
10:48:20 <bmckown> 38
10:48:34 <r0bby> old man
10:48:42 <bmckown> *real* old
10:48:51 <r0bby> I bet burke is older
10:48:58 * r0bby runs
10:49:18 <bmckown> burke is 24
10:49:20 <r0bby> probably not the best thing to do when the mentors can change their evals :-)
10:49:25 <r0bby> lies
11:24:05 *** shadowdoc has quit IRC
11:26:18 <njero> hey all, Leslie isn't around is she?
11:34:21 <bmckown> njero, She has an e-mailed to the google mentors list as late as yesterday
11:34:37 <njero> Aha...
11:34:47 <njero> thanks bmckown :)
11:34:52 <bmckown> sure :-)
11:51:59 *** njero has quit IRC
11:52:16 *** docpaul has joined #openmrs
11:52:16 *** ChanServ sets mode: +o docpaul
11:52:56 *** docpaul has quit IRC
12:12:37 *** TorLye has quit IRC
12:13:13 *** ekrub has quit IRC
12:41:49 *** nribeka_away has quit IRC
12:46:44 *** bmckown has quit IRC
13:04:41 *** [mharrison] has joined #openmrs
13:56:27 <OpenMRSBot> Recent updates in the world of openmrs: OpenMRS Tickets: Ticket #919 (defect created): BIRT throws "Failed to prepare the following query ..." during report generation <> || OpenMRS Tickets: Ticket #918 (defect created): BIRT throws "Failed to prepare the following query ..." during report generation <> || OpenMRS Tickets: Ticket #917 (defect created): BIRT throws "Cross-table query is not supported" error during report generation <> || OpenMRS Tickets: Ticket #916 (defect created): Report generation not handling date columns correctly (possible parsing error) <> || OpenMRS Modules: BirtReportModule 1.8.1 uploaded to OpenMRS Module Repository <>
14:31:35 <sioraiocht> anyone here speak german?
14:32:21 <r0bby> Not I
15:23:16 *** atomicturtle has quit IRC
18:19:03 *** atomicturtle has joined #openmrs
18:34:42 *** sioraiocht has quit IRC
19:09:49 *** jacobb has quit IRC
19:09:50 *** Echidna has quit IRC
19:11:28 *** Echidna has joined #openmrs
19:11:55 *** jacobb has joined #openmrs
19:12:56 *** jacobb has quit IRC
19:13:02 *** jacobb has joined #openmrs
20:44:32 *** njero has joined #openmrs
20:44:32 *** ChanServ sets mode: +v njero
20:53:42 *** njero has quit IRC
20:57:39 *** njero has joined #openmrs
20:57:48 *** ChanServ sets mode: +v njero
22:32:13 *** njero_ has joined #openmrs
22:32:19 *** njero has quit IRC
22:45:37 *** njero_ is now known as njero
22:45:39 *** ChanServ sets mode: +v njero
23:33:46 *** njero has quit IRC
23:47:59 *** njero has joined #openmrs
23:47:59 *** ChanServ sets mode: +v njero
  • No labels