|
Dessert Rose posted:compiler goes, "hey maybe you don't want to do that" turn on warnings
|
#
?
May 14, 2015 20:30
|
|
|
|
| # ? May 24, 2024 20:00 |
|
|
b0lt posted:turn on warnings then -Werror
|
|
#
?
May 14, 2015 21:13
|
|
|
Valeyard posted:i dont think ive ever seen two places agree on what "referential transparency" means, in regards to haskell, and how it differs from "pure" essentially you can treat them as interchangeable an expression is referentially transparent if all the functions involved in it are pure. technically an expression can be referentially transparent if it has an impure function as part of it, as long as the impure functions side effects have no impact on the behavior of the program (although you will almost never run into a situation where this is actually the case) so functions are pure, expressions are referentially transparent. in practice they often mean the same thing
|
|
#
?
May 14, 2015 21:22
|
|
|
fleshweasel posted:hi, just popping in to be the nth person to say do the interview bitch
|
|
#
?
May 14, 2015 21:22
|
|
|
Arcsech posted:impure functions side effects have no impact on the behavior of the program (although you will almost never run into a situation where this is actually the case one impure data structure for which is this is the case (by design) is in the ST library. in essence it's the same as State except that the mutations happen in-place and it gives destructive updates on mutable data. This lets you do stuff like doing various algorithms in place and so on. for a practical example see this paper: Lazy Depth-First Search and Linear Graph Algorithms in Haskell the core idea is that the main dfs algorithm is implemented using mutation to give vastly increased speed, and then there's lots of utility and other methods all implemented in pure code on top of this. actually using the module, you couldn't tell at all that it uses impure code internally.
|
|
#
?
May 14, 2015 21:28
|
|
|
still marking courseworks: quote:The results are presented in table ?? latex strikes again!!
|
|
#
?
May 14, 2015 21:46
|
|
|
Brain Candy posted:then -Werror we have warnings and warnings as errors turned on but lol if you think that matters apparently
|
|
#
?
May 14, 2015 21:52
|
|
|
oh, i see what you did there
|
|
#
?
May 14, 2015 21:53
|
|
|
Dessert Rose posted:we have warnings and warnings as errors turned on but lol if you think that matters apparently valgrind
|
|
#
?
May 14, 2015 22:07
|
|
|
Arcsech posted:essentially you can treat them as interchangeable well that does help a little and makes sense
|
|
#
?
May 14, 2015 22:07
|
|
|
Subjunctive posted:valgrind windows 
|
|
#
?
May 15, 2015 00:48
|
|
|
gonadic io posted:also most of them are overfitting to the test data so much down to picking the best seed for their RNG might as well go ahead and fill out their acceptance letters into the ml phd program then
|
|
#
?
May 15, 2015 01:01
|
|
|
it's only overfitting if there's another universally accepted data set that someone can use to call you out
|
|
#
?
May 15, 2015 01:11
|
|
|
why waste an excellent opportunity to add more data to your training set
|
|
#
?
May 15, 2015 02:39
|
|
|
i really wish i could program in java, c#, or swift every day and not ruby
|
|
#
?
May 15, 2015 03:01
|
|
|
c# is great
|
|
#
?
May 15, 2015 03:08
|
|
|
MALE SHOEGAZE posted:i really wish i could program in java, c#, or swift every day and not ruby i could have had a daily c# job but i decided to go gently caress myself and write c++ instead really what i want is a clojure job
|
|
#
?
May 15, 2015 03:23
|
|
|
in the end, at least you aren't writing php or js
|
|
#
?
May 15, 2015 03:23
|
|
|
I write verilog kill me
|
|
#
?
May 15, 2015 03:26
|
|
|
my reasoning was basically, i'm already pretty good at c#, so i should get that good with a different, more difficult language be careful what you wish for
|
|
#
?
May 15, 2015 03:28
|
|
|
yeah i dont really feel like i'm missing out by eliminating pointer issues from the class of problems i have to deal with like i am excluding myself from all the really cool low level stuff and eventually i'll want to get into it but there are plenty of interesting problems to solve w/o having to manage any memory
|
|
#
?
May 15, 2015 03:39
|
|
|
Modern C++ really doesn't have that much manual memory management, and what you do have can usually be handled automatically by a smart pointer. I actually really like C++ because it's not particularly dogmatic about it's OO, meaning you can have useful stuff like free functions. Plus I think RAII is a far-superior system to garbage collection.
|
|
#
?
May 15, 2015 03:46
|
|
|
Dessert Rose posted:really what i want is a clojure job same except common lisp. unfortunately, i haven't seen a shop here using anything but php.
|
|
#
?
May 15, 2015 04:11
|
|
|
MALE SHOEGAZE posted:yeah i dont really feel like i'm missing out by eliminating pointer issues from the class of problems i have to deal with lol if your only complaint with c++ is pointers
|
|
#
?
May 15, 2015 05:02
|
|
|
Bloody posted:lol if your only complaint with c++ is pointers i've never used it so... lol if you have any complaints about c++ besides pointers
|
|
#
?
May 15, 2015 05:03
|
|
|
pointers are good and cool
|
|
#
?
May 15, 2015 05:05
|
|
|
Dessert Rose posted:compiler goes, "hey maybe you don't want to do that" welcome to the big leagues, bucko!
|
|
#
?
May 15, 2015 08:08
|
|
|
Bloody posted:I write verilog kill me change it up and write Haskell to write Verilog instead
|
|
#
?
May 15, 2015 08:11
|
|
|
eschaton posted:welcome to the big leagues, bucko! this is what it feels like, yeah which is, I'll admit, part of the reason I decided to go for it
|
|
#
?
May 15, 2015 08:40
|
|
|
eschaton posted:change it up and write Haskell to write Verilog instead CλaSH (pronounced ‘clash’) uuhhhhh no thanks
|
|
#
?
May 15, 2015 08:51
|
|
|
https://wiki.haskell.org/Beta_reduction When it says "free occurrence" , surely it means "bounded occurrence", right?
|
|
#
?
May 15, 2015 12:19
|
|
|
terrible programmers thread: monad is like burrito
|
|
#
?
May 15, 2015 13:01
|
|
|
bobbilljim posted:terrible programmers thread: monad is like burrito i like to eat burritos
|
|
#
?
May 15, 2015 13:09
|
|
|
Captain Foo posted:i like to eat burritos i hate that you cant get good mexican food in denmark, it loving sucks
|
|
#
?
May 15, 2015 13:32
|
|
|
Snapchat A Titty posted:i hate that you cant get good mexican food in denmark, it loving sucks You'll have to make it your self
|
|
#
?
May 15, 2015 13:37
|
|
|
Snapchat A Titty posted:i hate that you cant get good mexican food in denmark, it loving sucks i recall this from my time in copenhagen
|
|
#
?
May 15, 2015 13:38
|
|
|
MALE SHOEGAZE posted:You'll have to make it your self btw i make a p good chili, pro tip: french onion powder also for some reason amsterdam has a lot of really great argentinian steakhouses
|
|
#
?
May 15, 2015 13:48
|
|
|
Snapchat A Titty posted:btw i make a p good chili, pro tip: french onion powder
|
|
#
?
May 15, 2015 13:54
|
|
|
Snapchat A Titty posted:btw i make a p good chili, pro tip: french onion powder this is also a good thing to mix into your hamburger beef
|
|
#
?
May 15, 2015 13:58
|
|
|
|
| # ? May 24, 2024 20:00 |
|
|
gonadic io posted:CλaSH (pronounced ‘clash’) that aside it honestly looks p interesting
|
|
#
?
May 15, 2015 14:55
|
|