Live data from Hacker News

Poll: Haskell or Go?

news.ycombinator.com

11–20 of 35 posts

Re: Poll: Haskell or Go?

#12

What kind of "server side" application? As a systems developer many of my applications would be more suited to Go. The biggest problem with Haskell's lazy evaluation system in real world analyses is that performance is unpredictable. While I would strongly consider using Haskell for an academic publication (e.g., my master's thesis), I would be more cautious about using it in a systems capacity. Then again, my assump…

Haskell with a cleaned up library and eager evaluation would be a dream. Alternatively, adding type classes to OCaml (and removing the object system) would yield a similarly awesome language.

Re: Poll: Haskell or Go?

#13
Haskel is beautiful language, but I voted Go because side effects in Haskel are pain and if you want language to the server I suppose it is because of doing side effects (like working with db or files, changing memory, serving webpages etc).

Plus Go has easy conurrency via message passing, which is nice to have these days.

Re: Poll: Haskell or Go?

#14
post #13

Haskel is beautiful language, but I voted Go because side effects in Haskel are pain and if you want language to the server I suppose it is because of doing side effects (like working with db or files, changing memory, serving webpages etc). Plus Go has easy conurrency via message passing, which is nice to have these days.

> Plus Go has easy concurrency via message passing, which is nice to have these days.

Indeed. I voted Go because learning it made writing concurrent software FUN! (As it should be)

Re: Poll: Haskell or Go?

#15
This is a rather limited poll, but if the choice is between Haskell and Go I'd pick Haskell for a web application, and Go for small things that need to be fast (like receiving and sending messages in a chat application). Go is too low level and generally not a very well designed language. It's a nice improvement over C, but it keeps far too much from C for my tastes.

Re: Poll: Haskell or Go?

#16
For me, Go has replaced C++ and Java for systems work. All of my projects at work for the past 6 months have been done in Go. I don't think we'll be going back.

But, really, good systems are based on well-defined protocols, which makes the language choice moot. (If you're passing Java or C++ objects over the wire, shame on you!)

Re: Poll: Haskell or Go?

#17
Depends a lot on the need.

Haskell has lots of fundamental weaknesses when it comes to long-running real world programs, laziness being the most obvious one.

Re: Poll: Haskell or Go?

#18
post #16

For me, Go has replaced C++ and Java for systems work. All of my projects at work for the past 6 months have been done in Go. I don't think we'll be going back. But, really, good systems are based on well-defined protocols, which makes the language choice moot. (If you're passing Java or C++ objects over the wire, shame on you!)

Could you possibly share a little bit more about the kind of work you do and why Go has become the clear winner over C++/Java?

I, for one, would really love to learn more about Go.

Re: Poll: Haskell or Go?

#19
post #14
post #13

Haskel is beautiful language, but I voted Go because side effects in Haskel are pain and if you want language to the server I suppose it is because of doing side effects (like working with db or files, changing memory, serving webpages etc). Plus Go has easy conurrency via message passing, which is nice to have these days.

> Plus Go has easy concurrency via message passing, which is nice to have these days. Indeed. I voted Go because learning it made writing concurrent software FUN! (As it should be)

Haskell has easy message passing as well. You can plop "forkIO" in front of any IO operation to make it concurrent

Re: Poll: Haskell or Go?

#20

What kind of "server side" application? As a systems developer many of my applications would be more suited to Go. The biggest problem with Haskell's lazy evaluation system in real world analyses is that performance is unpredictable. While I would strongly consider using Haskell for an academic publication (e.g., my master's thesis), I would be more cautious about using it in a systems capacity. Then again, my assump…

Haskell with a cleaned up library and eager evaluation would be a dream. Alternatively, adding type classes to OCaml (and removing the object system) would yield a similarly awesome language.

FYI, Haskell was originally created as an extension of Miranda, basically a testbed for lazy functional languages. Removing lazy evaluation seems almost like killing the language.
Post reply on HN