p-lang thread: (now (have you (problems two)))

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »

fart simpson: Jul 2, 2005; DEATH TO AMERICA

FamDav posted:

i like that jvm has a gay rear end c++, a gay rear end ml, and a gay rear end lisp now.

scala, ????, and clojure. whats the gay rear end ml

# ? Feb 10, 2014 19:35

Adbot: ADBOT LOVES YOU

# ? May 12, 2024 11:09

Notorious b.s.d.: Jan 25, 2003; by Reene

to put it another way, there is a fork in the road:

if you are committed to being C#-compatible, you will spend your dev time chasing whatever microsoft thinks is a good idea
if you are not committed to being C#-compatible, you do whatever the gently caress you want. that brought us scala and groovy.

# ? Feb 10, 2014 19:35

fart simpson: Jul 2, 2005; DEATH TO AMERICA

unless you mean java itself is the gay rear end c++, and scala is the gay rear end ml

# ? Feb 10, 2014 19:36

FamDav: Mar 29, 2008

MeramJert posted:

unless you mean java itself is the gay rear end c++, and scala is the gay rear end ml

yar, though honestly i know nothing about scala. should i? is scala more c++ than ml?

teach me

touch me

# ? Feb 10, 2014 19:39

Sapozhnik: Jan 2, 2005; Nap Ghost

not yet it doesn't

also wake me up when java 8 has properties because getter/setter boilerplate is so tedious that i use naked data members for classes that are literally just dumb tuples with no methods on them

try it sometime, it's liberating.

# ? Feb 10, 2014 20:23

Notorious b.s.d.: Jan 25, 2003; by Reene

Mr Dog posted:

not yet it doesn't

also wake me up when java 8 has properties because getter/setter boilerplate is so tedious that i use naked data members for classes that are literally just dumb tuples with no methods on them

try it sometime, it's liberating.

scala has everything you just wanted

properties on regular classes
dumb tuples with no methods -- even a literal syntax to define on the fly
"case classes" -- shorthand to define classes so simple/dumb they are just structs

edit: also all these things interoperate with regular java, so if a scala class produces a tuple or something as a return value, that will not be a pain in your teammates' balls

# ? Feb 10, 2014 20:28

Notorious b.s.d.: Jan 25, 2003; by Reene

it's almost as if thinking people sat down and considered what a more powerful java-like language would look like, then worked on implementing their ideas

# ? Feb 10, 2014 20:29

gonadic io: Feb 16, 2011; >>=

when I had to be in .net I used F# and it was okay. I guess I'll try scala if I'm ever on the jvm

# ? Feb 10, 2014 20:30

Notorious b.s.d.: Jan 25, 2003; by Reene

AlsoD posted:

when I had to be in .net I used F# and it was okay. I guess I'll try scala if I'm ever on the jvm

one of the great things about scala vis a vis F# is that even if you never, ever touch any of the FP stuff, it is still a useful improvement over java

the syntax is extremely similar to java, but a repl, type inference, and improved collections/primitives are super great even if you never want to get functional

i'm not sure anyone would ever work in F# unless they explicitly wanted F#'s FP stuff. (this might be true only because C# has moved forward a lot faster than java)

# ? Feb 10, 2014 20:32

Sapozhnik: Jan 2, 2005; Nap Ghost

eugh, no.

If you want to create a new language, create a new freestanding language, don't build one on top of the Indian burial ground that is the JVM and Java class library. Can you really tell me that the constraints of the JVM had no effect on Scala's design? I seem to recall something about reams and reams of generated interfaces.

I know it's easier said than done since the JVM is the product of 10 solid years of R&D into the best managed VM ever created, but jeez couldn't they have done it on top of LLVM or something? at least that one doesn't come with a huge class library that has Opinions.

poo poo, Ruby and Python just use bytecode interpreters and it seems to work out ok for them.

# ? Feb 10, 2014 20:34

gonadic io: Feb 16, 2011; >>=

Notorious b.s.d. posted:

i'm not sure anyone would ever work in F# unless they explicitly wanted F#'s FP stuff.

i'm pretty sure this is correct.

the trouble is though that a lot of the fp stuff is compiler related. does scala have tail call recursion? What about pure-code related optimisations like being really aggressive with constant folding and immutable collections and stuff?

# ? Feb 10, 2014 20:36

raminasi: Jan 25, 2005; a last drink with no ice

Notorious b.s.d. posted:

one of the great things about scala vis a vis F# is that even if you never, ever touch any of the FP stuff, it is still a useful improvement over java

the syntax is extremely similar to java, but a repl, type inference, and improved collections/primitives are super great even if you never want to get functional

i'm not sure anyone would ever work in F# unless they explicitly wanted F#'s FP stuff. (this might be true only because C# has moved forward a lot faster than java)

yeah a lot of the good stuff from f# leaks into c#. algebraic data types probably never will though

# ? Feb 10, 2014 20:37

Notorious b.s.d.: Jan 25, 2003; by Reene

AlsoD posted:

the trouble is though that a lot of the fp stuff is compiler related. does scala have tail call recursion? What about pure-code related optimisations like being really aggressive with constant folding and immutable collections and stuff?

scala has a really fuckin fancy pants compiler. it has to, in order to work at all.

it supports some very limited TCO. (the jvm doesn't have a working longjmp/goto instruction at the bytecode level, so nothing beyond very simple local TCO is even possible.)

coming from java, scala's collections library is a thing of wonder.

there is a set of basic types describing behaviour. (see chart)
for each basic type, there is at leas one mutable and one immutable implementation sharing the interface
the concrete implementations can be converted to parallelized types easily. "collection.map(foo)" is single-threaded. "collection.par.map(foo)" converts the collection and runs the map in parallel.

# ? Feb 10, 2014 20:49

HORATIO HORNBLOWER: Sep 21, 2002; no ambition,
no talent,
no chance

two words i loving hate: entity and role. ban these words forever please

# ? Feb 10, 2014 22:43

MononcQc: May 29, 2007

I want to put arbitrary bans on words like 'data', 'object', and 'thing' in code discussions and see how quickly the conversation grinds to a halt.

# ? Feb 10, 2014 22:45

Notorious b.s.d.: Jan 25, 2003; by Reene

MononcQc posted:

I want to put arbitrary bans on words like 'data', 'object', and 'thing' in code discussions and see how quickly the conversation grinds to a halt.

"ok, this is simple. when Alice frobs the fizzbat held by Bob, Charlie's whatsit will grok the updated fizzbat status, alright?"

jargon is better than nothing.

# ? Feb 10, 2014 22:48

gonadic io: Feb 16, 2011; >>=

MononcQc posted:

I want to put arbitrary bans on words like 'data', 'object', and 'thing' in code discussions and see how quickly the conversation grinds to a halt.

what about if i make it clear that i'm talking about the data keyword?

# ? Feb 10, 2014 22:51

Notorious b.s.d.: Jan 25, 2003; by Reene

AlsoD posted:

what about if i make it clear that i'm talking about the data keyword?

we'll deprecate it and replace it with a unary $$..> operator

then the forbidden words can go in peace and people can try and explain "double dollar sign double dot gee tee" annotations

# ? Feb 10, 2014 22:53

MeruFM: Jul 27, 2010

MononcQc posted:

I want to put arbitrary bans on words like 'data', 'object', and 'thing' in code discussions and see how quickly the conversation grinds to a halt.

you prefer people using the structure type?

Bio and chem have long names up the rear end deep into the colon. Now there's a billion acronyms and it's almost impossible for anyone to join the conversation until you read at least 2-3 landmark papers on the subject. And it's still not a good bullshit filter because anyone with half a brain can memorize acronyms, see HR and recruitment.

# ? Feb 10, 2014 22:55

MononcQc: May 29, 2007

If we can't bother being able to name the poo poo we're using in a conversation properly instead of using overly vague keywords, we shouldn't expect our audience/interlocutors to get a decent understanding of the meaning we carried, at least without examples or further indications to help clarify the intent behind said ~~key~~buzzwords we used :colbert:

# ? Feb 10, 2014 23:07

FamDav: Mar 29, 2008

MononcQc posted:

If we can't bother being able to name the poo poo we're using in a conversation properly instead of using overly vague keywords, we shouldn't expect our audience/interlocutors to get a decent understanding of the meaning we carried, at least without examples or further indications to help clarify the intent behind said ~~key~~buzzwords we used

suck my dick canuck bitch

# ? Feb 11, 2014 00:23

FamDav: Mar 29, 2008

actually i like being explicit better this is fun tia

# ? Feb 11, 2014 00:23

Damiya: Jul 3, 2012

Scala owns and is a blast to write and poo poo like implicit classes rules and implicits rule.

:getin:

# ? Feb 11, 2014 00:41

MononcQc: May 29, 2007

FamDav posted:

suck my dick canuck bitch

I won't play into your fantasies, sorry.

# ? Feb 11, 2014 01:08

fart simpson: Jul 2, 2005; DEATH TO AMERICA

FamDav posted:

yar, though honestly i know nothing about scala. should i? is scala more c++ than ml?

teach me

touch me

idk i thought you were making a reference to the hugeness of scala. i don't think scala's really like ml though. like it clearly has been influenced by ml to some degree, but everything seems way more java. this is the type signature for map:

code:

def map[B, That](f: A => B)(implicit bf: CanBuildFrom[Repr, B, That]): That

and you dont actually need to touch any fp stuff at all and can use it as just a better java.

although i guess there's scalaz, which i never used but seems like a library for people that really cant cope with scala not being an ml

# ? Feb 11, 2014 01:32

gonadic io: Feb 16, 2011; >>=

MeramJert posted:

although i guess there's scalaz, which i never used but seems like a library for people that really cant cope with scala not being an ml

code:

Apply[Option].apply2(some(1), some(2))((a, b) => a + b)

# ? Feb 11, 2014 01:38

fart simpson: Jul 2, 2005; DEATH TO AMERICA

AlsoD posted:

code:

Apply[Option].apply2(some(1), some(2))((a, b) => a + b)

welp,

# ? Feb 11, 2014 01:42

echinopsis: Apr 13, 2004; by Fluffdaddy

god u guys talk some gibberish at times

# ? Feb 11, 2014 01:47

Damiya: Jul 3, 2012

Scalaz is scary.

We don't go there.

Mostly we just do some map stuff and a lil bit of implicit references and pass some first class functions around.

It's better that way.

# ? Feb 11, 2014 01:47

HORATIO HORNBLOWER: Sep 21, 2002; no ambition,
no talent,
no chance

MononcQc posted:

If we can't bother being able to name the poo poo we're using in a conversation properly instead of using overly vague keywords, we shouldn't expect our audience/interlocutors to get a decent understanding of the meaning we carried, at least without examples or further indications to help clarify the intent behind said ~~key~~buzzwords we used

right on man right on

# ? Feb 11, 2014 02:11

MononcQc: May 29, 2007

you need to have data to prove the things you object to :smuggo:

# ? Feb 11, 2014 02:13

Notorious b.s.d.: Jan 25, 2003; by Reene

i warned you guys the type signatures got ugly

# ? Feb 11, 2014 03:42

MeruFM: Jul 27, 2010

MononcQc posted:

If we can't bother being able to name the poo poo we're using in a conversation properly instead of using overly vague keywords, we shouldn't expect our audience/interlocutors to get a decent understanding of the meaning we carried, at least without examples or further indications to help clarify the intent behind said ~~key~~buzzwords we used

I'll take good distinct examples over more precise jargon.

Or jargon that doesn't have ambiguities.

# ? Feb 11, 2014 05:28

Nomnom Cookie: Aug 30, 2009

i like java better than scala. but at least scala doesn't have to do a hash lookup just to find a variable so it's miles better than a p-lang

# ? Feb 11, 2014 08:10

Innocent Bystander: May 8, 2007; Born in the LOLbarn.

if you were a langauge what would you put in your toy box?

# ? Feb 11, 2014 09:15

FamDav: Mar 29, 2008

whole lotta dicks

# ? Feb 11, 2014 09:17

Innocent Bystander: May 8, 2007; Born in the LOLbarn.

IF YOU WERE A LANGUAGE!

# ? Feb 11, 2014 09:18

Damiya: Jul 3, 2012

I just wrote some code that uses the |>> operator and some other code that uses |>>> and &>>

You're welcome Notorious B.S.D.

# ? Feb 11, 2014 14:47

motedek: Oct 9, 2012

is there an ide to write all the scala boilerplate? i'm curious about it but gently caress if i'm going to write stuff like this:

MeramJert posted:

code:

def map[B, That](f: A => B)(implicit bf: CanBuildFrom[Repr, B, That]): That

# ? Feb 11, 2014 17:49

Adbot: ADBOT LOVES YOU

# ? May 12, 2024 11:09

Nomnom Cookie: Aug 30, 2009

motedek posted:

is there an ide to write all the scala boilerplate? i'm curious about it but gently caress if i'm going to write stuff like this:

you're starting to see why i prefer java

# ? Feb 11, 2014 17:56

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > p-lang thread: (now (have you (problems two)))

«‹›1784 »