|
gonadic io posted:honestly the biggest issue i have is that rust has "match foo {" and scala has "foo match {" omg yes every time. also typing `case Some(x)` instead of `Some(x)`
|
#
?
Feb 19, 2019 00:23
|
|
|
|
| # ? Sep 9, 2024 16:56 |
|
|
DONT THREAD ON ME posted:perhaps if performance mattered to you you should choose a language with zero cost abstractions, like rust
|
|
#
?
Feb 19, 2019 00:26
|
|
|
Fatty Crabcakes posted:Weren't you defending clojure in another thread? i didnt defnd it i just said it was better than groovy.
|
|
#
?
Feb 19, 2019 00:33
|
|
|
my point was dumb though anyhow. it's odd to me though because i've never used a dynamic lang for anything that wasn't mostly IO bound. if i reach a point where i'm swapping out semantically equivalent language constructs due to the performance of their implementations, then i kinda want to use something...closer to the metal. DONT THREAD ON ME fucked around with this message at 00:55 on Feb 19, 2019 |
|
#
?
Feb 19, 2019 00:34
|
|
|
DONT THREAD ON ME posted:i didnt defnd it i just said it was better than groovy.
|
|
#
?
Feb 19, 2019 00:36
|
|
|
DONT THREAD ON ME posted:my point was dumb though anyhow.  anyone who cares about quality posts containing well-reasoned arguments won't be reading this forum without me on ignore
|
|
#
?
Feb 19, 2019 00:38
|
|
|
cinci zoo sniper posted:about twice faster iirc. i sometimes have this problem at the scale of 10^7 lists, where that is an appreciable improvement yeah on this kinda scale you can just surrender to numpy arrays and pray one case I kinda remember chain being faster was when I was using it together with itertools' combinatoric generators, it was faster than doing a double comprehension fake edit: here's a super dumb unreadable example of what I mean $ python3 -m timeit 'import itertools; (3, 3) in [x for sg in (itertools.permutations(range(i)) for i in range(7)) for x in sg]' 10000 loops, best of 3: 76.3 usec per loop $ python3 -m timeit 'import itertools; (3, 3) in list(itertools.chain.from_iterable(itertools.permutations(range(i)) for i in range(7)))' 10000 loops, best of 3: 67 usec per loop
|
|
#
?
Feb 19, 2019 00:43
|
|
|
stuff like pypy usually optimize lists better than iterators so maybe that's where you got the performance boost actually
|
|
#
?
Feb 19, 2019 00:46
|
|
|
DONT THREAD ON ME posted:my point was dumb though anyhow. yeah, I totally agree with you, I just wanna be pedantic here. yospos birch
|
|
#
?
Feb 19, 2019 00:48
|
|
|
Symbolic Butt posted:yeah on this kinda scale you can just surrender to numpy arrays and pray oic. my problems usually are getting to the point where i can operate on arrays, so rip
|
|
#
?
Feb 19, 2019 06:21
|
|
|
please don’t write comprehensions with more than one for in them. the weird backwards syntax is bad and confusing and your code will be bad and confusing. thank you
|
|
#
?
Feb 19, 2019 09:53
|
|
|
c database structure s: let foo be a json containing object baz with N named objects with M identically named objects in each N. the query im looking at right now, which used to be simple left join on 3 fields in the old database version, now has me parse foo.baz so that i sort Ns by name in ascending order, find last name(N)::int that is <= input A, and then find underlying name(M)::int that is equal to input parameter B, and fetch value(M) for that
cinci zoo sniper fucked around with this message at 11:07 on Feb 19, 2019 |
|
#
?
Feb 19, 2019 10:22
|
|
|
some days i just get my faithful box of tissues, bottle of hand moisturiser, and think what it’s like to have a team that actually does things to achieve specific goals in a manner that considers not only themselves, and not just to show that they are “doing work”
cinci zoo sniper fucked around with this message at 11:08 on Feb 19, 2019 |
|
#
?
Feb 19, 2019 10:24
|
|
|
just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code
|
|
#
?
Feb 19, 2019 15:00
|
|
|
Maximum Leader posted:just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code are you sure you're not having a stroke?
|
|
#
?
Feb 19, 2019 15:10
|
|
|
Maximum Leader posted:just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code what
|
|
#
?
Feb 19, 2019 15:14
|
|
|
Maximum Leader posted:just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code I honestly have no idea how you can prevent VS Code from indenting code.
|
|
#
?
Feb 19, 2019 15:22
|
|
|
AggressivelyStupid posted:are you sure you're not having a stroke? Beamed posted:I honestly have no idea how you can prevent VS Code from indenting code. its surreal, i think he's begrudgingly going to make an attempt at indentation though
|
|
#
?
Feb 19, 2019 15:27
|
|
|
I always mash ctrl alt f in vscode
|
|
#
?
Feb 19, 2019 15:34
|
|
|
Maximum Leader posted:just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code this but with new lines
|
|
#
?
Feb 19, 2019 15:51
|
|
|
Maximum Leader posted:its surreal, i think he's begrudgingly going to make an attempt at indentation though is he super old or something, I think the closest I've seen that kinda attitude was with some cobol programmer once
|
|
#
?
Feb 19, 2019 15:56
|
|
|
gonadic io posted:i'm interviewing at a company that claims to be doing HFT ...in f# im really curious about the company
|
|
#
?
Feb 19, 2019 16:04
|
|
|
yeah I recently caught up with an old friend who is maintaining legacy pacbase (antiquated thing that generates cobol code and was originally designed for punch cards) code for an insurance company and he works on 80*25 terminals That's a world where indention was probably a luxury like the code is divided up into columns with fixed widths and stuff, it sounded pretty nightmarish
|
|
#
?
Feb 19, 2019 16:06
|
|
|
Beamed posted:I honestly have no idea how you can prevent VS Code from indenting code. i assume you can maybe define ‘’ as custom tab separator?? but why????
|
|
#
?
Feb 19, 2019 16:27
|
|
|
Maximum Leader posted:just had an encounter with a programmer that doesn't use indentation, refuses to use indentation and claims its "normal" to write code like this. not only that but actually zero indentation (no spaces, no tabs) is supposedly the default in visual studio code you can configure the indent size by language in vscode, this programmer must have manually set it to 0
|
|
#
?
Feb 19, 2019 16:36
|
|
|
gonadic io posted:i'm interviewing at a company that claims to be doing HFT ...in f# you can absolutely do high frequency trading with f# the tricky bit is doing it fast enough
|
|
#
?
Feb 19, 2019 17:06
|
|
|
Zlodo posted:yeah I recently caught up with an old friend who is maintaining legacy pacbase (antiquated thing that generates cobol code and was originally designed for punch cards) code for an insurance company and he works on 80*25 terminals our closest partner company maintains an as/400 db schema that dates all the way back to the 80s and has been growing ever since last time i looked at it their primary schema alone (out of a few hundred schemas) had about 70k tables with close to 600k columns in total the kicker: eight chars for table names, seven for columns. hope you're consistent with your aliases for ZVDAST0F
|
|
#
?
Feb 19, 2019 17:26
|
|
|
Maximum Leader posted:its surreal, i think he's begrudgingly going to make an attempt at indentation though I think you met Rob Pike raminasi posted:im really curious about the company I'm currently interviewing with them but if you email me at contact@gonadic.io I'll tell you the company name etc I don't have any internal details yet, still early in the process. Boiled Water posted:you can absolutely do high frequency trading with f# 
|
|
#
?
Feb 19, 2019 17:58
|
|
|
Set up a pre-commit hook which automatically formats everybody's code consistently and have done with it?
|
|
#
?
Feb 19, 2019 21:18
|
|
|
no, just fail the commit and don’t say why
|
|
#
?
Feb 19, 2019 21:36
|
|
|
Symbolic Butt posted:is he super old or something, I think the closest I've seen that kinda attitude was with some cobol programmer once I get the impression that he's self-taught and didn't have any technical guidelines or supervision during his 10 years of experience. not relevant but he looks like an ex-criminal and is shirtless on his linkedin profile picture
|
|
#
?
Feb 19, 2019 21:39
|
|
|
Maximum Leader posted:not relevant but he looks like an ex-criminal yo this is a really weird thing to say what does an “ex-criminal” look like shirtless on LinkedIn profile is pretty lol tho
|
|
#
?
Feb 19, 2019 21:58
|
|
|
like a reformed citizen I suppose
|
|
#
?
Feb 19, 2019 22:10
|
|
|
https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf are these openai guys actually doing something interesting, or is it just a glorified markov bot? if my academese is still fresh, they appear to claim that they can summarize and translate things, but only because their training set full of reddit posts contains lots of tl;dr and quotes like "as they say in french, blahde blah blah, which means 'some english phrase'" otherwise it looks like a pile of whatever-implementation-details-who-cares that just blindly tries to maximize the probability of its chosen next word given the previous words, then repeats until done
|
|
#
?
Feb 19, 2019 22:14
|
|
|
eschaton posted:no, just fail the commit i added commit hooks to our main repos and now I keep getting asked why commits don't work because reading the output window in vs or the error message is apparently too hard
|
|
#
?
Feb 19, 2019 22:31
|
|
|
balena.io (formerly resin.io) have a cool thing where when you push to their remote (that's how you send them your code to build and flash onto devices) it'll run the docker build step as part of the commit hook and show you the stdout honestly i'm pretty impressed with them. they started out as just an arm port of docker but now they offer a (v expensive) full rpi level hosted thing. they take care of: deploying code, updating devices, deploying secrets, OS setup (you base your dockerfile from one of their arm base images), ssh, public ip, logs, restarts, etc saved us probably hundreds of manhours setting all this poo poo up. i think they offer some kind of self-hosted poo poo too to cut down on costs but idk anything about that
|
|
#
?
Feb 19, 2019 22:40
|
|
|
Powerful Two-Hander posted:i added commit hooks to our main repos and now I keep getting asked why commits don't work because reading the output window in vs or the error message is apparently too hard i wish github let you add hooks on their end. i bet gh enterprise does this, right?
|
|
#
?
Feb 20, 2019 00:07
|
|
|
gonadic io posted:balena.io (formerly resin.io) have a cool thing where when you push to their remote (that's how you send them your code to build and flash onto devices) it'll run the docker build step as part of the commit hook and show you the stdout it’s ok. we used them but it was a bad idea because it flat out doesn’t function if you break the model of “100% of devices are or can easily be connected to the internet”. also it’s super expensive.
|
|
#
?
Feb 20, 2019 00:18
|
|
|
Blinkz0rz posted:i wish github let you add hooks on their end. i bet gh enterprise does this, right? idk we use gitlab and it let's you add all sorts of stuff it's OK but it utterly loving sucks at merges so I always end up doing them locally and pushing then it gets all pissy about branches being unmerged edit: to be fair I suck at merges too Powerful Two-Hander fucked around with this message at 01:20 on Feb 20, 2019 |
|
#
?
Feb 20, 2019 01:08
|
|
|
|
| # ? Sep 9, 2024 16:56 |
|
|
Just accepted my first programming job outside of academia! I'm really excited, and seriously a huge part of getting to this point was hanging around this thread and reading all of your thoughts and getting my questions answered in a friendly and helpful way. Just wanted to pop in and say thanks to everyone here. You all are great 🍻
|
|
#
?
Feb 20, 2019 01:37
|
|
