|
Bloody posted:lol scala is a crazy lang did you actually find a resource after everyone here XY'd it?
|
# ? Dec 21, 2015 21:35 |
|
|
# ? Jun 12, 2024 07:23 |
|
kinda yeah. part of scala for the impatient is free online
|
# ? Dec 21, 2015 22:02 |
|
JawnV6 posted:did you actually find a resource after everyone here XY'd it? if you mention
|
# ? Dec 21, 2015 22:09 |
|
Bloody posted:kinda yeah. part of scala for the impatient is free online scala for the impatient is a good starter for anyone who has developed in a language used by more than, like, 5 people. for functional scala, i don't really think there's a good resource for just that. the community for the most part has been moving away from scalaz' "haskell, but atrocious" junk and that library has even said operator soup just doesn't work in scala.
|
# ? Dec 22, 2015 01:15 |
|
has the old couriers course by odersky been mentioned? it's 100% about functional programming in scala if that's what you're specifically looking for.
|
# ? Dec 22, 2015 03:22 |
|
|
# ? Dec 22, 2015 18:04 |
|
im glad someone made that joke into a thing IRL
|
# ? Dec 22, 2015 18:25 |
|
countdown to some oblivious fucker sporting that sticker unironically.
|
# ? Dec 22, 2015 22:14 |
|
lold
|
# ? Dec 22, 2015 22:39 |
|
ultramiraculous posted:countdown to some oblivious fucker sporting that sticker unironically. haha yeah, mongoDB is actually really good!
|
# ? Dec 22, 2015 23:54 |
|
FamDav posted:scala for the impatient is a good starter for anyone who has developed in a language used by more than, like, 5 people. functional programming in scala is probably the closest fit. it's pretty good and the last chapter covers scalaz-stream which can otherwise be pretty opaque if you want to do stuff with types and run on the jvm there are worse choices. frege could be cool but its even less mature than scala. the scala fp crowd are for the most part still all there doing stuff, a bunch of people have been moving away from scalaz only inasmuch as they've moved to cats which (aims to) have all the same stuff with different names/package + module structure/etc and still has some of the "fun" operators with their attendant surprised and confusion. this is awesome because now other libraries get to work out how to support both cats and scalaz at the same time. Fullets fucked around with this message at 07:34 on Dec 23, 2015 |
# ? Dec 23, 2015 07:24 |
|
|
# ? Dec 23, 2015 07:26 |
|
how about kotlin
|
# ? Dec 24, 2015 01:11 |
|
looked at it briefly a while ago and it didn't seem to offer enough over java8 + judicious lombok to justify learning a new language that nobody else uses but I'm prepared to be convinced I missed something
|
# ? Dec 24, 2015 01:38 |
|
kotlin would be better if it didn't require an associated runtime jar, but I could see replacing application code with it
|
# ? Dec 24, 2015 03:28 |
|
it's more interesting on android where you're never getting anything past java 6
|
# ? Dec 24, 2015 03:29 |
|
Soricidus posted:looked at it briefly a while ago and it didn't seem to offer enough over java8 + judicious lombok to justify learning a new language that nobody else uses
|
# ? Dec 24, 2015 14:43 |
|
Sagacity posted:you end up with delomboked code I don't, though
|
# ? Dec 24, 2015 18:56 |
|
Alan Kay posted:The same way it should be easier to do your own composting, you should have the ability to deal with complicated ideas by making models of them on the computer.
|
# ? Dec 24, 2015 23:12 |
|
Gazpacho posted:OOP = poop
|
# ? Dec 24, 2015 23:50 |
|
hey perl 6 is out today? well https://rt.perl.org/Public/Bug/Display.html?id=123766 i guess there's nothing blocking it and they did this? https://perl6advent.wordpress.com/2015/12/24/an-unexpectedly-long-expected-party/ but there's no links to a release? i guess there's something on the page for rakudo perl http://rakudo.org/ which is a month old. like, what's been released? https://github.com/perl6/roast/commits/6.c the test suite has a 6.c branch? the specs haven't changed either https://github.com/perl6/specs anyone anyone
|
# ? Dec 25, 2015 20:17 |
|
perl6 was inside us all along
|
# ? Dec 25, 2015 20:39 |
|
Painfully Esoteric, Released Late
|
# ? Dec 25, 2015 20:47 |
|
actual perl 6 launch footage https://www.youtube.com/watch?v=7O4V7JfeTSU
|
# ? Dec 25, 2015 21:08 |
|
i remember caring about perl6
|
# ? Dec 26, 2015 00:01 |
|
you should care about perl 6 because the more you learn about it the more hilarious it gets for example the main data type Cool from which nearly everything inherit provides string methods and numeric methods to most of the type heirarchy, but doesn't have the Numeric role or the Stringy role (also the Stringy role doesn't provide any methods) anyway https://perl6advent.wordpress.com/2015/12/25/christmas-is-here/
|
# ? Dec 26, 2015 12:06 |
|
qntm posted:you should care about perl 6 because the more you learn about it the more hilarious it gets i remember when it was announced. i remember when hugs came out and suddenly people thought it might happen for real this time. i even met someone with a perl 6 "lambda camel" t-shirt working on it (nb, same place where i saw 3+ plan 9 users in the same room at same time, and there were more dylan programmers than perl 6 ones at the time too). quote:for example the main data type Cool from which nearly everything inherit provides string methods and numeric methods to most of the type heirarchy, but doesn't have the Numeric role or the Stringy role (also the Stringy role doesn't provide any methods) i have stared at the void for too long and can see the rationale, or invent one "everything will need to be printed eventually, so everything will have a string method, and then we'll tag the real strings with "Stringy" so you can avoid implicit typecasting to string by a role check" "i guess if we want to avoid implicit type casting, we need to do the same for numbers. hey, but we use scalars as the basic type. let's just have a scalar type and then explicit tags for number or string, and call it cool because i'm larry wall"
|
# ? Dec 26, 2015 14:47 |
|
i often joke that ruby is a language designed to look pleasing to read on slides, no excess punctuation, familiar structure, and the semantics seem to be bolted on to make a handful of syntax examples work. perl 6 seems to be what happens when this rationale is applied over fifteen years, and instead of in slideware, specware. ( i am sure the tests and specification are well written, larry is a very good and funny technical writer, part of why people picked up perl 5 in the first place was the docs)
|
# ? Dec 26, 2015 14:56 |
|
qntm posted:you should care about perl 6 because the more you learn about it the more hilarious it gets so i saw a "perl 6 is happening" talk at fosdem, it's as if perl has been in a nuclear bunker for the last 15 years. most of the focus was on "so, we found a thing you wanted to do and we couldn't do it in one or two readable lines so we added another operator and data type". i mean, i like roles, i like rules, but most of perl 6's useful ideas could and have been bolted onto a smaller language, and they already have docs, modules, libraries, package management, test frameworks. instead, perl has spent 15 years polishing a series of hard to google operators and syntax quirks, and a whole bunch of features to let people write code with a tiny line count. not so much "if we build it they will come" but anything that could be rewritten in perl 6 from perl 5 will have been turned into go, ruby, python by now, and the ones left won't be changing any time soon. it's not as if python 3 was a simple change but when you look at perl 6 it is hard to call python 3 anything other than wildly successful.
|
# ? Dec 26, 2015 15:14 |
|
I'm late to the party but the posts about Java pass by reference made me go back and look at some small examples.Java code:
Java code:
|
# ? Dec 26, 2015 16:38 |
|
pepito sanchez posted:I'm late to the party but the posts about Java pass by reference made me go back and look at some small examples. String is an object type, but value is an object reference, which is passed by value
|
# ? Dec 26, 2015 16:50 |
|
this was the first impression I had of perl 6 beyond its name: http://www.ozonehouse.com/mark/periodic/ at the time I couldn't tell if it was satire
|
# ? Dec 26, 2015 17:28 |
|
pepito sanchez posted:I'm late to the party but the posts about Java pass by reference made me go back and look at some small examples. let's look at something simpler. int a = 1 int b = a a = 3 # a != b now what if you do String a = new String("foo") String b = new String("foo") a != b but a.equals(b) or String a = "foo" a = "bar" what you're doing in those two examples above is changing what the variable stores. in java, variables can store either primitive values directly, or store the address of an object in memory elsewhere. as you're doing it to a function parameter, the value (an address or a primitive) is copied into a new variable for the function's environment, and then reassigned. it is worth bearing in mind that python and ruby and javascript are also call by value, but they do not have primitives, they can only store addresses of objects. but you can pass in mutable objects and change them, and it's as if you did this List a = new ArrayList(); List b = a a.append("foo") # a == b is true, they both store the same address in memory # a.get(0) will be b.get(0)
|
# ? Dec 26, 2015 17:33 |
|
Serenade posted:this was the first impression I had of perl 6 beyond its name: http://www.ozonehouse.com/mark/periodic/ satire is dead, perl is postmoden
|
# ? Dec 26, 2015 17:39 |
|
my first experience with perl6 was https://www.reddit.com/r/perl6/comments/r44do/is_perl6_code_unreadable/
|
# ? Dec 26, 2015 17:46 |
|
right. I didn't know that about java function parameters. But I never tried using void methods that way either. Because why would anyone? The rest makes sense. Thanks.
|
# ? Dec 26, 2015 17:47 |
|
tef posted:"i guess if we want to avoid implicit type casting, we need to do the same for numbers. hey, but we use scalars as the basic type. let's just have a scalar type and then explicit tags for number or string, and call it cool because i'm larry wall" after 15 years of work, they arrived at p much the perl 5 situation ([ask] me about writing xs extensions)
|
# ? Dec 26, 2015 17:47 |
|
One advantage of Perl 6 is people won't rant at you when your project supports Perl 5 but not Perl 6.
|
# ? Dec 26, 2015 18:02 |
|
Suspicious Dish posted:my first experience with perl6 was https://www.reddit.com/r/perl6/comments/r44do/is_perl6_code_unreadable/ [–]diakopter 2 points 3 years ago I agree it's inpenetrable and therefore unmaintainable to a non-expert of Perl 6. Otoh, not all code needs to be maintainable. Some code [like this example] exists purely for aesthetic purposes, or for educational purposes. Also, some code is purposefully written as tersely as possible to challenge the reader's parsing skills. Idiomatic Perl 6 is apparently maximally-terse Perl 6, or at least maximally-distinctive Perl 6, and writing in a language idiomatically is all the rage these days. Languages whose idioms differ greatly from most other languages are distinctive. To sum up, yes, very idiomatic code in any language is unmaintainable by someone not familiar with that language's idioms. . Whether or not highly idiomatic code should be used in production (for business or for large-scale personal projects) is another question. Your business doesn't need to commit to writing/maintaining highly terse Perl 6 if you choose to use an implementation of Perl 6 in production. Enforce yourself and your developers to use whatever coding standards you think are best: especially with Perl, you have ultimate flexibility in designing standards that are effectively DSLs or use whatever programming paradigm you wish.
|
# ? Dec 26, 2015 18:21 |
|
|
# ? Jun 12, 2024 07:23 |
|
i have monthly contributions to rakudo core for over a year while programming stuff in it for over 3 years now. it's pretty easy because things will be thrown out the window and completely changed at the whim of a fart. for instance: the entire module loading and byte code generating stage was revamped like 2 months before release, and by revamped I mean someone took the lead devs gist of pseudo code literally (as in his first step was cutting and pasting the pseudo code into a fork and patting themselves on the back in irc repeatedly) while never looking at the actual spec or current implementation. There are multiple modules, including Star bundled http useragent, that had to add 'no precompilation' to even work. also it will still generate the precompiled files but this is ok because no one ever used a read only file system ever. it was also important to merge this poo poo in its half assed form as soon as possible because of egos and attention. unsurprisingly it's still just the one person working on that part presumably because everyone who worked on the previous implementation also realized it just needed to be brought inline with the concepts of the aforementioned gist, not redone from scratch because empty functions make more sense to you than working code. all of these people are smart as hell but there really is no strong lead or direction beyond the parts Larry rules on. there is a recent PR from someone very new to programming that got merged, and then half a day later the pr author comments that it should be reverted (which should have been obvious to anyone) but hey who needs to code review anything from the guy who blogs about using poop6 to scrape MLP fan fiction and also has a github avatar of one. I'm mad about new perl and wrote all of this
|
# ? Dec 26, 2015 19:10 |