Like wiggity wack yo dude! dont be dissin my shizzle yo man!
Name:
Anonymous2014-06-30 10:19
The main problem of Go is statical typing, which have long been proven to be pseudo-scientific crackpottery, like strong AI.
Name:
Anonymous2014-06-30 11:33
>>4 The last I heard from SPJ was that Haskell is (he said this playfully) "useless" in the sense that it's academic and not suited for a lot industry of areas. It's still useful, the Real World Haskell movement is a good thing, especially because people are learning to adapt or create other useful programming languages to allow better correctness guarantees.
The true utility of Haskell is to show how useful certain concepts are. People have gone beyond that to make the guarantees stronger. In truth Haskell is a focal point for a certain body of knowledge. It's not bad, it's not like we'd be better off not knowing what Haskell and related languages have to say. You can write monads in nearly any language.
>>7 In that case I agree, I am learning myself haskell anyway
Name:
Anonymous2014-06-30 15:35
>>12 Yes. Haskell is good to know, and often good to use.
Writing code in something like Haskell isn't necessarily harder, in fact I think it's easier if you approach it tabula rasa. It's often seen as hard for the experienced programmers because it's not closely related to anything in their language tree. You can't reach for the same tools, but you do get other ones, about half of which are there to prove the other half actually works and are being used correctly.
Right now, Rust is trying hard not to become Haskell (OCaml really) for systems programming. It's a shame in a sense, but they're really terrified of looking threatening to C++ programmers. That's valid in a marketing sense, but as someone who fights viciously for the future he wants to live in I would rather be a little more daring in hopes that C++ programmers will acclimate via network effect if nothing else. I don't want it to become Haskell, but a zero-overhead and more practical OCaml with idioms intelligible to C++ programmers where applicable sounds about right.
Haskell is seen as hard, because it takes more lines of code to do useful things, like stateful code. Moreover, it is impossible to write efficient code in Haskell.
Name:
Anonymous2014-06-30 15:47
The Haskell hype is similar to the OO hype of some years ago. If you don't know it, you can't criticize it. If you DID know it, you wouldn't criticize it. If you DO criticize it, well clearly you just don't "get it". And of course there's a special "vocabulary" you have to learn to use appropriately -- call a 'method' a 'procedure' and you've grouped yourself with the incompetent wanna-be's.
Name:
Anonymous2014-06-30 15:48
>>15 You're wrong on both counts. 1) Haskell is seen as hard because of its heady libraries and loads of compiler extensions. The core of Haskell is really simple, though: lambda calculus and pattern matching.
2) It is totally possible to write efficient code in Haskell, as has been demonstrated lots of times. It's just hard to reason about performance with all that laziness and compiler aggression and nonstandard runtime workings going on. But if you do devote some time to optimization, you'll get efficient, Java or C efficient (which is good enough in the vast majority of cases).
Name:
Anonymous2014-06-30 15:58
>>14 Sure, you can write Fortran in anything but I mean in the context of writing idiomatic code. I don't think it would be all that hard to learn how to write that kind of code in Haskell if that's all you wanted to do.
>>15 I assume you're trolling because the efficiency thing is obvious bait, and Haskell sources aren't all that long unless you're comparing to something like J.
However the "useful things, like stateful code" quip is exactly what I'm talking about. Most of your code doesn't need to be stateful, even in languages that aren't seen as primarily functional. Fluent-style code in any language has a very low density of stateful operations. When you stop thinking about using interim side-effects as a means to get interim results and start focusing on expressing transformations between function input and eventual output, functional languages try to give you all the tools you need to do so painlessly. It is a different way of thinking, and that can be hard to grasp at first.
Name:
Anonymous2014-06-30 15:59
>>16 But it's true: the vast majority of criticism is that "ZOMG it haz no effects", "whatcha mean I can't mutate me some shit in a dope for loop, shizzle my nizzle?" and "it's academia and academia is useless, all useful things are created by braindeads like teh Guido".
>>13 Design decisions in Rust are driven primarily by the needs of Servo, which as you can imagine is a huge project with lots of people working on it. Hypothetical popularity isn't a big deal to them. They mostly want their program to work. So, many of the changes have been a direct result of trying certain features and observing whether they improved the development process of Servo, or made it worse. Very empirical.
Name:
Anonymous2014-06-30 16:09
MOZILLA QUALITY, SERVO CERTIFIED™
Name:
Anonymous2014-06-30 16:37
>>20 That's true in a lot of ways, but not entirely.
The developers have periodically expressed fear of becoming too much like Haskell or looking too much like OCaml. Familiarity to C++ programmers is often cited as motivation when modifying syntax or arguing against features. This has nothing to do with Servo, you can't measure something you don't have.
A good example is there was a pyramid of deeply nested matches in Servo (or maybe it was rustc) and separately IoResult handling was also really icky. Instead of solving the problems generally (do-notation and HKTs were popular requests) the half-measure of writing specific workarounds for each was taken. They were both very ugly, just less ugly than not using them.
There still hasn't been a language that makes C89 not worth using for systems programming and just about everything else.
Name:
Anonymous2014-06-30 19:42
>>25 I think C99 is a significant improvement, and C11 would be great if MS would shut up about C++ and implement it already. Who am I kidding? no VS user is going to try to compile my code.
If Rust had the libraries I'd bring it up. Right now it's pretty much only good for writing compilers and video games.
Name:
Anonymous2014-06-30 19:50
>>25 BLISS was pretty nice and it had way better macros than C.
Name:
Anonymous2014-06-30 20:04
>>26 C11 is a abomination but it has some useful stuff Same goes for C99 I use both
you can simplify this to "unemployed need not apply", because I've never seen an employed functional programmer.
Name:
Anonymous2014-07-01 10:08
>>31 So you only hire those who are already employed?
Name:
Anonymous2014-07-01 10:08
Why functional programmers always get fired?
Remember, there are potential negative consequences of retaining a poor performer after you've exhausted all the options available: - You'll annoy other members of your team, who may have to work harder to "carry" the poor performer. - You may promote a belief in others that you're prepared to accept mediocrity – or, worse, underperformance. - You may waste precious time and resources that could be better used elsewhere. - You may signal that some employees deserve preferential treatment. - You may undermine the whole idea of finding the best person for the job.
Name:
Anonymous2014-07-01 10:40
>>30 You mean like libraries for windowing, 2D and 3D graphics, font raster, asset management, linear algebra, physics, scripting, controller input and HMD support with speed in the ballpark of C? [ Check. ]
>>30 And the real world are retards who don't know what they are doing and cause people to die unlike the academics/professionals.
Name:
Anonymous2014-07-01 18:30
I thought functional programming went against the grain against ``enterprise'' but you guys also seem to make fun of that. Retarded noobs, functional is the future (quantum) and too hard for you low IQ mongrels. In the future there will be no such thing like libraries or 3rd party support, just the pure technical performance of the language (speed, efficiency, elegance, performance, expressiveness, safety, it must be at the top in all good qualities) will be the judge and ensurer of its popularity and use.