#dat

karissa: interesting, i'm in almost the opposite world

most people in bioinformatics live on the command line

at least, where i've been

heh yeah. its hard to tell honestly, I come from social science and general informatics land.

that's my original background

quantitative social science

yeah, that sounds about right :)

from my experience & rule of thumb, if it looks further away from english, analysts are less likely to touch it.

FYI we are trying out https://gitter.im/datproject/discussions as an alternative to IRC

also, windows is popular because of Word and Excel.

for the time being i will just stay logged into both

although that's less likely nowadays cause of mac, but its legacy carryover

and mr. gates' amazing monopoly superpowers!

ogd: cool

ogd: sweeet

that's my only real contribution is that i think it'd be nice to have something that's further away from bash because there's a reason why a lot of people don't even touch it (even a lot of programmers!)

karissa: although why break stride with the legions who already understand "|" and ">"

erikg: true.

and also, if we don't go for something glue-like, it'll be harder for the scripting environment to natively support those who know python/R

erikg: yeah, there should be gluey pieces

erikg: i would guess that people who understand | and > are already writing their own bash scripts

erikg: to do pipelining

erikg: i would say that we have two types of users: people who are comfortable w/ command line and those who are not.

we have to handle both of them gracefully without under or overwhelming

perhaps it makes sense to be able to run arbitrary bash, but by default have some nice grammar for those who aren't comfortable

karissa: I agree!

karissa: to flesh things out, we need the buyin of power users as much as we need more casual ones to use what is built and extend on it

mafintosh: https://github.com/maxogden/httpsocket

groundwater_: o/ (uses the approach you mentioned)

ogd: sweet

ogd, bmpvieira, erikg : thanks all for a really productive hangout today

ogd: oh nice!

so client fetches on notifications?

groundwater_: yea

cool!

groundwater_: check out the code, i need to consider error cases and cache/buffer limits more

groundwater_: but the basics work now

ogd: i imagine back-pressure is difficult to propagate

i.e. how does the server know when to stop buffering

groundwater_: yea i dont handle it at all

mafintosh: word!

ogd: mafintosh erikg great hangout