Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

If Haskell is so great, why hasn't it taken over the world? (2017)

pchiusano.github.io

Post reply on HN
Print view View original

101–110 of 154 posts

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#101

Post by _8j50 »

Y_Y wrote: post #17
_8j50 wrote: post #9

It takes an insane amount of time to compile ghc. I can't stand apps that use it because of that. Could take hours on some machines, would have considered riding on that bandwagon if it wasn't for that limitation.

I use lots of versions of GHC across several architectures, and can't remember the last time I compiled GHC. I think you must be doing it wrong. It's not without its flaws, but `stack` can do a lot of the work for you too.

On some platforms that build from source haskell apps pull ghc as a dependency and build it from source as well. I try to avoid those apps but it is getting harder.

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#102

Post by jjtheblunt »

Animats wrote: post #57

The author notes the success of Go. Go's developers knew when to stop. Go is a mediocre language, but it has all the parts needed for server-side software. That's a big market. I suspect that part of the success of Go is simply because its libraries are heavily used. You know that the library is serving a zillion requests a second in Google data centers. Major bugs have probably been found by now. Go has one solidly…

Wait.

What facility does Go have for immutable values?

(I'm presuming you mean the val vs var in Scala.)

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#103

Post by lgas »

pitaj wrote: post #56
Smaug123 wrote: post #47

Earlier quoted context omitted.

It's not an uncommon meme that programmers and mathematicians possess a weird twist in the mind that allows them [in the strong form of the meme] to do, and [in the weak form of the meme] to enjoy doing, what they do. If you grant that, then it's a small step to admitting the existence of different sub-twists that incline you towards different parts of the abstract space. If you don't grant that, then consider it a r…

Why the word "defect" though? There must be other words that are less controversial. Maybe "abnormalities" or "specialities"?

"Interests"? "Appreciations"? "Proclivities"? "Personalities"?

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#104

Post by JoshMnem »

swebs wrote: post #67
reikonomusha wrote: post #10

Because the industry isn’t dominated by well thought out solutions. Most programmers I talk to love the idea of spending time and using the perfect tools to design outstanding solutions to problems. They discuss how much they supposedly like to learn. But when push comes to shove, your 9–5 career growth is probably going to be best optimized by cobbling a bunch of Python together and shipping it. Especially if you wr…

But it doesn't seem too popular in the open source world either. The only large project I can think of that uses it is Xmonad.

The Elm compiler is written in Haskell.

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#105

Post by dirkt »

tanilama wrote: post #66
Animats wrote: post #57

The author notes the success of Go. Go's developers knew when to stop. Go is a mediocre language, but it has all the parts needed for server-side software. That's a big market. I suspect that part of the success of Go is simply because its libraries are heavily used. You know that the library is serving a zillion requests a second in Google data centers. Major bugs have probably been found by now. Go has one solidly…

In other words, Go is designed for practicality. It shows design consideration to favor engineering while Haskell, as far as I can tell, barely cares.

Haskell actually doesn't care by conscious choice: It's meant as a research language, and has "avoided success at all costs" (to use the tongue-in-cheek quote) to retain the freedom to play around with things. Which in turn lead to quite a few inventions that are now being copied by other modern languages (like Rust), though they rename them to make them less scary (apparently programmers are easily scared by names ... like the one that starts with M.).

That's a bit sad. On the other hand: Imitation is the sincerest form of flattery.

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#107

Post by dfacastro »

opnitro wrote: post #14
reikonomusha wrote: post #10

Because the industry isn’t dominated by well thought out solutions. Most programmers I talk to love the idea of spending time and using the perfect tools to design outstanding solutions to problems. They discuss how much they supposedly like to learn. But when push comes to shove, your 9–5 career growth is probably going to be best optimized by cobbling a bunch of Python together and shipping it. Especially if you wr…

Interestingly, the comp-sci program at University of Maryland does start with a list-esque system. They start with a very basic language and slowly add new features throughout the first course.

The master's computer engineering program at the Faculty of Engineering of the University of Porto, Portugal, starts off with Scheme in the first semester. The fourth year includes logic programming with Prolog.

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#108

Post by wwwater »

robotmay wrote: post #31

From my own experience, Haskell is hard to learn. I have no grounding in maths, but that hasn't stopped me learning most languages. It's a pretty massive barrier to Haskell though, and I find that a lot of the community really struggle to explain things in terms that non-haskellers will understand. The biggest problem I've found is a real inability to explain _why_ the things it does are cool, or what real-world appl…

I think it's hard to explain why the things it does are cool, because it's related to the aesthetic sense of beauty and it's very difficult to explain that one. It's like you take a person who is doing number theory, and you ask them to explain why they are so fascinated that you can approximate prime numbers distribution with a particular function.

I think it's related to aesthetics of seeing how things are connected into a "bigger picture". It gives you an immense sense of beauty, once you are able to see that things, that you thought are not related, are connected into a bigger picture, and are actually very much related.

In Haskell a lot of things are about this "bigger picture", while in many popular languages the things are just a bunch of practical tools, that have nothing to do with each other.

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#109

Post by wwwater »

danharaj wrote: post #51

I've done a 9-5 in mostly Haskell since 2013. If you have specific questions about using Haskell in production, I can share my experience. If you've had a bad experience with a piece of tech and draw a broad conclusion, I think it's reasonable if unfortunate that it left a bad taste in your mouth. But I also think there's a brighter picture to paint :)

I have a specific question: do you hire remotely? :)

Re: If Haskell is so great, why hasn't it taken over the world? (2017)

#110

Post by tuvok »

As a mathematician who works with category theory, Haskell is a piece of cake. Much simpler than java, python, ruby, etc. It just requires a kind of thinking that is not tought to computer scientistis or on any programming courses. I recommend the book Category Theory (Steve Awodey ,2006).
Post reply on HN