Why isn't Haskell popular in industry? • Hacker News Board

Re: Why isn't Haskell popular in industry?

#141

Post by applicative » Tue, Aug 03, 2010, 6:00 PM UTC

vicaya wrote: post #136

jrockway wrote: post #48

Earlier quoted context omitted.

Interestingly, I came to the exact opposite conclusion. Haskell is easier than C++ (less code and more safety), and it is equally fast. Haskell may not be the perfect language for your generic FFT library to be distributed with your OS, but it's a great choice for building applications that would otherwise be C++ or Java. Java and C++ had no trouble catching on, despite being slower than C, and Haskell isn't even tha…

You know, it'd be more convincing if you can come up with reproducible benchmarks to backup your claims. According to This guy's data: http://www.codexon.com/posts/debunking-the-erlang-and-haskel... Haskell is even slower than python (which is not known for its speed) for implementing simple servers. Haskell code that performs close (within an order of magnitude in terms of run time and space) to its C/C++ equivalent…

If you read a little further, you will discover that there is some discussion of his 'data'.

Re: Why isn't Haskell popular in industry?

#142

Post by codexon » Tue, Aug 03, 2010, 7:42 PM UTC

applicative wrote: post #133

applicative wrote: post #131

Earlier quoted context omitted.

This is febrile raving, codexon, why are you insisting on it? It is like a kind of madness, right on its face. A range of solutions were given on http://haskell.forkio.com/gmpwindows of which the writer, Sigbjorn Finne, said "nothing too magic or new here". None of them seem particularly appetizing, of course, which is presumably why people continued to labor.

I was trying to think why I found this so disturbing: "I simply have no ulterior motive or incentive against organized groups who have their livelihoods and PhDs based on Haskell." The thing is, it cannot be honest. A simple reflection, a Google search, e.g. http://www.google.com/search?num=100&hl=en&lr=&a... and some time spent reflecting on the modules linked, will show that Don S. could make a far better 'liveliho…

He does it because he is biased towards something he has already spent years on his PhD candidacy and many hours on writing libraries already. If he completes his PhD, it is likely that he will try to get a Haskell job. Just because he wrote a thousand Haskell packages doesn't mean he is as suitable in another language. Sure he could probably be a C programmer, but his pay grade won't be as high as if he landed a Haskell job to take advantage of his experience and degree.

That you cannot see this is because you also have the same bias as Dons.

Re: Why isn't Haskell popular in industry?

#143

Post by loup-vaillant » Wed, Aug 04, 2010, 6:30 PM UTC

codexon wrote: post #123

loup-vaillant wrote: post #99

Earlier quoted context omitted.

Err, I was only objecting to your attack on Don's on-line behaviour. Actually I'm (still) an outsider. Consider "what I ever saw" as anecdotal evidence. Now it's a pity that companies are scared of the GPL. As far as I know, most software is custom or private, is never released[1], and thus can't possibly infringe the GPL. I think the real problem here is corporate culture. No company would object using the glibc in…

Why do you object? I just proved to you the legitimacy of my arguments while you still stand by Don's complete denial?I think it is very off-putting that anything remotely negative about Haskell is usually met with snide comments questioning their intelligence or being labeled as a troll. It also doesn't help that my comment was upvoted a few times, and when Dons & Co. replies, it drops to -3. I've also observed this…

> It also doesn't help that my comment was upvoted a few times, and when Dons & Co. replies, it drops to -3.

Interestingly, I saw it happen almost everywhere. It looks like people are more likely to down-vote comments which are contradicted by a recognized authority. This is of course not a good thing.

> You aren't really an outsider considering your participation in the Haskell Cafe mailing list and being credited in Real World Haskell.

I didn't post more than a few messages, 2 years ago, and I posted about 3 minor comments in Real World Haskell. I've read a few papers, but seldom have written anything (except http://www.loup-vaillant.fr/articles/assignment of course). I'm not an outsider, but hardly what I consider to be an Insider.

> You also don't have any proof that most software is custom or private.

I don't. But in this (huge) niche, my point still stand: being afraid of the GPL is silly. Corporations that are should stop being.

> you cannot deny that the shrink-wrap industry is huge

I cannot and I won't.

> The answer is to avoid GPL code in the language […]

Yes, assuming you want the (proprietary) shrink-wrap industry to use Haskell. I want it gone. Therefore, I see the GPL as the solution.

Re: Why isn't Haskell popular in industry?

#144

Post by BMeph » Tue, Aug 10, 2010, 9:16 PM UTC

ww520 wrote: post #110

wisty wrote: post #35

Earlier quoted context omitted.

achieving anything practical using purely functional code remains extremely difficult ... Excel is pretty much a functional language (though it is a little disabled), and people do amazing things in it. The problem isn't that functional languages are hard. They are a bit different, but good code is often fairly functional anyway. The problem is that most of the community seems to be obsessed with showing that functio…

It's interesting to see Wisty's post and the replies immediately prove his point. BTW I like Haskell.

I disagree. His point is not merely that Haskell advocates cannot clearly explain simple interfaces like Monad (let's be honest here; this is more about Haskell folks being too researchy, and not being better skilled in the dog-and-pony-show environment of "practical" marketing), but that the lack of skill in clear explanation is due to a reluctance to admit that "Haskell needs the ability to do imperative things in order to be useful".

The thing to "get" about Monads are that they are a means to describe imperative things going on, in a way that blends in with not necessarily imperative things. Which is part of why it's so tough to explain them - we're all talking past each other regarding what "the point" of using Monads are. The point to them in Haskell is mainly twofold:

  1) Haskell users can use a Monad to describe net access, DB access, file access, and other "once done, the results are out of your control" effects in the same way as describing using (Writer) computations that keep a log of what's being done, (Error) exceptions, (Reader) environment variables, (List) functions with multiple - or no - results, and more.

It's really easy...and kind of mind-blowingly complex, all at the same time.

Re: Why isn't Haskell popular in industry?

#145

Post by cmars232 » Sat, Aug 28, 2010, 8:58 PM UTC

silentbicycle wrote: post #18

cmars232 wrote: post #13

Earlier quoted context omitted.

Why doesn't Haskell support objects? Seems like that would be quite useful. They don't need to be mutable or anything. They could be like records, that scale.

Because it has typeclasses, which solve the same problems without creating the new problems that conventional objects do. Why doesn't Java support pattern matching?

Typeclasses are a fine substitute for polymorphism in behavior (methods). But does Haskell throw out the baby with the bath water? Haskell seems to lack an ability to express a taxonomy of data models. Records are not even close to a substitute.

Why can't Haskell make it easier to express data structures? Something "turtles all the way down" introspective, like a MOF (meta-object facility, http://en.wikipedia.org/wiki/Meta-Object_Facility) would be nice.