<     May 2017     >
Su Mo Tu We Th Fr Sa  
    1  2  3  4  5  6  
 7  8  9 10 11 12 13  
14 15 16 17 18 19 20  
21 22 23 24 25 26 27  
_2_8 29 30 31
00:26 prophile joined
00:55 trolling joined
01:38 prophile joined
02:04 Pyrrh joined
02:07 <Pyrrh> jfhbrook: I finally did the things to get back on irc
02:14 <jfhbrook> nice!
02:14 <jfhbrook> welcome back
02:17 <Pyrrh> why thank you :3 I can't remember the other channel I was chillin' in before though
05:39 _whitelogger joined
05:54 _whitelogger joined
06:00 _whitelogger joined
06:21 _whitelogger joined
06:28 Pyrrh joined
07:57 _whitelogger joined
09:35 cloudhead joined
09:51 fujisan joined
10:25 mylesborins joined
12:56 prophile joined
21:04 <ec> Pyrrh!
21:04 <ec> Pyrrh: you don't need any other channel.
21:04 <ec> this is the ~Superior Channel.~
21:05 <Pyrrh> ec while that is true I like a well-rounded internet diet and I can't help it
21:05 <Pyrrh> YOU CAN'T HOLD ME BACK
21:05 <Pyrrh> i can't be tamed
21:10 <pikajude> I like a poorly rounded internet diet myself
21:10 <pikajude> like, real shit
21:11 <jfhbrook> In real life I eat exclusively super burritos
21:11 <jfhbrook> I like to think my internet habits mirror that
21:18 <Pyrrh> mmmmmmm burritos
21:18 <Pyrrh> I eat a LOT of vegetables
21:19 <Pyrrh> way more than I want to
21:24 <pikajude> don't you mean monads, jfhbrook
21:26 <jfhbrook> definitely not
21:26 <jfhbrook> I've spent the last week porting scala code to python and have at least 2 more weeks of that in my future
21:26 <jfhbrook> so monads functors and anything along those lines, are all on my shit list right now
21:27 <pikajude> oh
21:27 <pikajude> scala has monads+
21:27 <pikajude> ?
21:27 <jfhbrook> scala has everything
21:27 <pikajude> that's funny
21:27 <pikajude> i remember that about scala-
21:27 <pikajude> jesus christ
21:27 <pikajude> this keyboard is killing me
21:27 <jfhbrook> but on friday for instance I was dealing a lot with JsValue thingers
21:28 <jfhbrook> I think they're monads
21:28 <jfhbrook> they can either represent a json parsed thing or an error of some kinda
21:29 <pikajude> sounds ideal for monads
21:29 <pikajude> or applicatives
21:29 <jfhbrook> almost definitely a monad
21:30 <jfhbrook> either way, I think scala is awful and bad
21:31 <pikajude> might have been an applicative
21:31 <jfhbrook> not only is it way too complicated, but it's like the perfect storm of jvm assholes and fp assholes all in a convenient one-stop shop
21:31 <pikajude> yeah i hate scala
21:31 <pikajude> it's a mess
21:31 <pikajude> it's the voynich manuscript of languages
21:31 <jfhbrook> I'd forgotten about that
21:31 <jfhbrook> had to look it up
21:32 <jfhbrook> yeah so like
21:32 <jfhbrook> ugh I can't tell if this project is awesome or stupid
21:32 <jfhbrook> so kinja has a data pipeline that they use for internal comment and pageview counts, and somehow it's also driving recommendations
21:32 <jfhbrook> haven't pieced the whole thing together yet
21:33 <jfhbrook> but it's a scala/play app hosted in their DC writing to a kinesis queue, with a custom pipeline for writing to s3 and redshift that could be completely replaced with kinesis firehose but they don't want to
21:33 <jfhbrook> and then a bunch of spark jobs, of course also written in scala, on EMR
21:34 <jfhbrook> there's one person that fully understands it, and one other that's made minor tweaks, and that's claire and my boss respectively
21:34 <jfhbrook> claire always seems irritated at my questions, like I'm thick or something; doesn't seem to trust us (can't replace the thing with firehose, can't rewrite the play app)
21:35 <jfhbrook> my boss is insisting that my team works in python and not scala, so anything we take over gets ported to python
21:35 <jfhbrook> aaaand yeah, now I'm trying to port a bunch of spark jobs from scala to python
21:35 <jfhbrook> with a chance of taking over the ingestion pipeline and the app if we do a stellar job, I guess
21:35 <jfhbrook> it's stupid, I feel like claire's trying to hand off the part she doesn't like
21:36 <jfhbrook> buuut this is also legitimately difficult to work with, the spark jobs I mean
21:36 <jfhbrook> claire also wants to move off EMR and onto hand-managed ec2 boxes, which is fine, ok
21:36 <jfhbrook> I'd probably feel better about this if the spark stuff wasn't legitimately hard
22:54 cloudhead joined
23:37 meowrobot joined