These are chat archives for codyhan94/cs135-data-visualization

3rd
Mar 2015
Cody Han
@codyhan94
Mar 03 2015 19:56
from chat with sam
@jvanbrug @minfawan @yaoshenghan @fictivelogic
in essence, the plan now is to get the data generation team to use this app while generating data, telling it where they're putting the data they generatd
I realize now that what we were trying to do is more or less completely impossible independent of the generation step
or, at least, makes no sense to do independently of it because if we can just roll this app for both GUI + data generation to use, we're in good shape
i.e. all of the data that we want to store is available while the data is being generated, so it makes sense that they should tell the database to store it as they generate it. and there's various nice ways Jan can call on the app to get his files
Jan Van Bruggen
@jvanbrug
Mar 03 2015 20:03
I don't understand.
Cody Han
@codyhan94
Mar 03 2015 20:06
sorry what I posted doesn't make too much sense out of the context of my chat with sam
Jan Van Bruggen
@jvanbrug
Mar 03 2015 20:06
So in the DataFile model you put a way for the stat generator to point to the csv location
Cody Han
@codyhan94
Mar 03 2015 20:07
I have an app on django/db-setup called data that, once complete, will expose an interface for stat generation to log the stats it generates and for GUI to grab the data that it asks for
Jan Van Bruggen
@jvanbrug
Mar 03 2015 20:07
yes.
what is the impossible thing you and sam were trying to do?
or should i not care?
fictivelogic
@fictivelogic
Mar 03 2015 20:08
I'm not sure what is impossible either..
Cody Han
@codyhan94
Mar 03 2015 20:08
I think I just had this idea that the data generation could be done completely independently of us cataloging it in the DB
or rather, should be done. and I don't think it should be that way at all because that would just incur a lot of extra work and a lot of hard-coded things
this way everything uses as naming conventions the fields I specify in data/models and is super easily extensible
to have more splitting dimensions all they have to do is add another field and remigrate the schemas
Jan Van Bruggen
@jvanbrug
Mar 03 2015 20:11
So the stat generator would be interacting with the database?
I assumed there would be some script in the data app that would check for new files and add them
Perhaps new files on a different server
I'm also not sure where the stat generator will run
@yaoshenghan @minfawan do you guys know where your library will run? should we separate it from the rest of the web server and put it on the simulation server?
fictivelogic
@fictivelogic
Mar 03 2015 20:13
It seems that having a script on the data app that searches for new files to add is probably easiest and most flexible.
Cody Han
@codyhan94
Mar 03 2015 20:18
that can be arranged. I was going to provide some functions for the stat generators to work with as well
am going to
Cody Han
@codyhan94
Mar 03 2015 20:30
crap! I didn’t mean to decline the PR
Jan Van Bruggen
@jvanbrug
Mar 03 2015 20:44
can you un-decline it?
fictivelogic
@fictivelogic
Mar 03 2015 21:25
CTRL+Z
Jan Van Bruggen
@jvanbrug
Mar 03 2015 21:26
haha
@codyhan94 @fictivelogic @obinnae Now that the new django project is in, let's plan to name all our branches like "gui/X" and "db/X"
unless we're working on a different or joint feature
fictivelogic
@fictivelogic
Mar 03 2015 21:41
rog @jvanbrug
Cody Han
@codyhan94
Mar 03 2015 22:54
ok