|
OBAMA BIN LAUGHIN posted:python is a really cool language its good and sensible and easy to write for and easy to debug and if performance got you down you can easily write critical parts in c/c++ (using weave if you're doing simple numerical stuff, otherwise using swig)
|
# ¿ Apr 25, 2012 23:18 |
|
|
# ¿ Apr 27, 2024 06:27 |
|
Police Academy III posted:someone tell spiders he should've used a defaultdict there you can just call get with a default argument pizza[tomatoes] = pizza.get(tomatoes, []) + [int(peppers), int(olives)] you dont need a whole nother type
|
# ¿ May 20, 2012 05:55 |
|
Char posted:also if you want to have a language understandable by management but usable by developers just use cucumber features. so this thing generates unit tests from actual (specifically-structured) written english? seems actually cool, definitely good for communicating with nonprogrammers
|
# ¿ May 20, 2012 15:18 |
|
|
# ¿ May 20, 2012 22:31 |
|
i like most every physics person and ee person ive met but i dont like reading their code, b/c they are the worst programmers physicists with spaghetti fortran and ee nerds with terse c functions that are 400 lines long (and single letter variables)
|
# ¿ May 21, 2012 00:37 |
|
Gazpacho posted:um you know function calls have overhead and it all compiles to a single binary anyway dont you??? my fist compiles into a single binary in your solar plexus
|
# ¿ May 21, 2012 01:24 |
|
rotor posted:some day software engineering will be actual engineering but not any time soon. right now we're still at that awkward phase that cars were back when each one was handmade and some ran on steam and some ran on kerosene and some ran on compressed air. i think we're more like carpenters everyone learns a detailed set of joint styles and has their favorite materials. quality varies widely by the experience and care of the person making the piece of furniture eventually someone will come along with the ability to mass-produce cheap versions of things that suit most needs, and the entire discipline will fade away except for in a few niche cases where the mass-produced versions wont work
|
# ¿ May 21, 2012 03:21 |
|
Sulk posted:Python isn't a better option than Java for an intro course, so that's a bad decision on their part. yeah, but when you think about it, its a much better option than Java for an intro course, so it's actually a good decision on their part
|
# ¿ May 21, 2012 14:00 |
|
Sweevo posted:i think its a rule that all Intro to Programming classes have to use a completely retarded language. in my day it was either pascal or modula-2 depending on which crusty greybeard teacher you got. then they switched to vb6 which is even stupider c++ when i took ap cs its just mean, bc after 2 semesters you can barely know any of the language even though you feel like youve really learned stuff
|
# ¿ May 21, 2012 14:03 |
|
i think cs professors are bad at teaching coding the same as most math professors are bad at teaching basic calc like they spend all day working on compiler theory or machine learning or what have you, they never think about how to teach someone whats an array or whats a string
|
# ¿ May 21, 2012 14:58 |
|
Internaut! posted:was it jane street, they're known for being very proud of their special snowflake tech which iirc in tyool 2005 was vb6 i know a few people there and they seem to like it. ive thought once or twice about learning ocaml just to work there, but on the other hand... ocaml other terrible 'learn a language get a job' ideas include scala for working at foursquare
|
# ¿ May 23, 2012 12:49 |
|
speaking of which, internaut, which would you rather work at -- a proprietary trading type place (trading internal money to get the partners rich), or a mutual fund type place (trading on gramma's money and trying not force her to eat cat food during retirement)? i'm thinknig i'd almost lean toward the former. even though you're not doing the public any good, your ability to do the public serious damage is smaller
|
# ¿ May 23, 2012 12:52 |
|
it seems like a lot of the math that funds do is high-dimensional integrals that they have to use mcmc / gibbs / whatever sampling for, and it seems like that stuff would be pretty amenable to cuda but maybe that's way off base
|
# ¿ May 24, 2012 02:38 |
|
Internaut! posted:excel calling out to matlab functions
|
# ¿ May 26, 2012 16:13 |
|
"my 1981 Yugo only has a top speed of 45 miles an hour" "thats ok, just replace the engine with the one out of this 1974 Gremlin with the broken head gasket, and you'll be driving in style"
|
# ¿ May 26, 2012 16:17 |
|
Internaut! posted:excel and matlab are noted for their poor numerical capabilities im sure it does make sense in the context of your work, im just baffled. i mean lots of things have good numerical capabilities and also work as decent languages
|
# ¿ May 26, 2012 17:09 |
|
Shaggar posted:lol thats awful how is the orange sky on your planet? do the three moons look nice today?
|
# ¿ May 29, 2012 17:11 |
|
if your changing what your comparing, you will need to name the new comparator something else, and youll have to find and replace anyway. shaggargument is specious
|
# ¿ May 29, 2012 17:22 |
|
BonzoESC posted:does python have both comparator sort and a schwartzian transform sort-by method shwartzian transform is considered "the old way" but yeah you can do it that way
|
# ¿ May 29, 2012 17:30 |
|
standardtoaster posted:Is this accurate? nah theyre more like daemons than spirits (at least in *nix)
|
# ¿ May 29, 2012 19:34 |
|
shaggar what is the best program ever written
|
# ¿ May 29, 2012 23:26 |
|
lol @ using computers im posting this via smoke signal from my campfire in montana
|
# ¿ May 30, 2012 16:14 |
|
Janin posted:This is completely different, good job being mutually incompetent in three languages I guess. well the spec was written in haskell and no non-spergs can read that
|
# ¿ May 30, 2012 17:11 |
|
Internaut! posted:those are just from the last page tbh i think a lot of people would actually benefit by taking basic real analysis. definitely if you do any numerical stuff. plus the cantor set is cool as poo poo
|
# ¿ May 31, 2012 15:05 |
|
i remember the pde stuff that was basically breaking stuff down to fourier components after separation of variables, but i would have a hard time solving an ode by hand irl if i got asked to analyze a system of differential equations i would immediately go to graphical methods
|
# ¿ Jun 1, 2012 03:48 |
|
fourier transform is useful as all hell i think ive learned it separately like 5 times. in pdes, for signal processing / wavelets, in real analysis (the graduate one), in a data course (useful because it also went over sampling issues and windowing functions). theres probably one more ive been in school a long drat time is what im sayin
|
# ¿ Jun 1, 2012 04:03 |
|
Hammerite posted:watching americans talk about these things is weird because they say things like "oh calculus 2 was my favourite calculus" and expect you to know what the hell calculus 2 is. are your mathematics courses really totally 100% uniform across every institution in the whole country or what lower level classes are more standardized. probably a combination of the fact that we have gen-ed requirements, ap exams, and people transfer after finishing gen-ed stuff fairly often
|
# ¿ Jun 2, 2012 14:14 |
|
All Hat posted:yah, jonnty's bang on the money here. when backing up a car with a trailer attached, you arent actually solving 2nd degree integrals. its all numerical approximation cerebellum, i hardly knew him!
|
# ¿ Jun 2, 2012 14:17 |
|
erlang is definitely the most yospos, but scala actually seems p cool and maybe even useful
|
# ¿ Jun 26, 2012 05:59 |
|
Ronald Raiden posted:using emacs would be like touching RMS, that's gross. Contributing would be like eating his scabs. if you hate rms you should not contribute to emacs and instead have children human children, not half-human half-parrot grotesque freak children
|
# ¿ Jun 27, 2012 19:19 |
|
i'm like a month behind in this thread. snyping to say hi to me from a month in the past
|
# ¿ Feb 5, 2013 23:19 |
|
Shaggar posted:has there been anyone who's tried the current asp.net web stack probably not
|
# ¿ Dec 10, 2015 16:06 |
|
Shaggar posted:everyone here whos tried it has switched and the only people who haven't tried it are scared that they'll get something that works instead of having to deal with garbo p-langs. ive heard that .net stacks are common in finance. do you work in finance? the only tech place I know with a .net stack is zocdoc, and they were started by ex-finance people, ha
|
# ¿ Dec 10, 2015 16:14 |
|
i thought all the stacks looked like react or whatever now, with the back-end stuff in jvm services?
|
# ¿ Dec 10, 2015 16:19 |
|
Destroyenator posted:.net is one of the best stacks if you can deploy to azure paas offerings or have people who care enough (and you trust) to maintain windows servers. luckily lots of enterprise shops do actively maintain windows server installs and startups can use azure easily so there's lots you can do in that space. hopefully the next asp (due in q1) will make mono a viable option. was just glancing through the documentation on azure; didn't realize it supported hadoop and spark and all that is there an equivalent to kafka and storm for queuing / streaming events?
|
# ¿ Dec 11, 2015 00:28 |
|
Bloody posted:hi someone point me to a deece learn you a scala thing thanks in advance odersky's coursera course is useful
|
# ¿ Dec 20, 2015 11:05 |
|
FamDav posted:scala is going to persist in ML so long as spark is a thing, though with dataframes not incurring a real performance penalty in python i could see scala being an implementation detail. pyspark and spark dataframes are the path to suicide only make rdds of case classes. also don't use mllib, it's mostly lovely wrappers for breeze on spark dataframes. just use breeze on the rdds.
|
# ¿ Dec 20, 2015 22:54 |
|
qntm posted:Our code reviews are part of the system, you can't commit without one. are the review pre-commit people on Gerrit? i like Gerrit OK, but github (and sometimes phabricator) seems more popular with it's review pre-merge
|
# ¿ Dec 21, 2015 15:57 |
|
Soricidus posted:at least I think that's the reasoning idrk probably originally it was "they can implement that themselves" and now it's more that java stuff gets used as a back-end language where people are using DI from a config or reading in parameters from the DB or whatever that works well with continuous deployment
|
# ¿ Dec 29, 2015 01:30 |
|
|
# ¿ Apr 27, 2024 06:27 |
|
tef posted:this awful post became this awful post http://programmingisterrible.com/post/139222674273/write-code-that-is-easy-to-delete-not-easy-to that was a fun read. thanks tef
|
# ¿ Feb 15, 2016 02:13 |