True Scala complexity

Post by gclaramunt » Wed, Jan 11, 2012, 1:55 PM UTC

Peaker wrote: post #149

gclaramunt wrote: post #128

Earlier quoted context omitted.

I guess it depends on the definition of "powerful". Scala unlike Haskell, has first class modules ( aka OO ) and unlike Coq (And Agda, I presume) is turing-complete. ( Not that those things really matter to make an awesome language :) ) But for mature statically typed langs running JVM, Scala is the most powerful ...

Well, Scala has advantages and disadvantages over Haskell, but I wouldn't say it is more "powerful" than Haskell. Consider the implementation of a properly lazy "const" function in Haskell: const x _ = x And in Scala: http://apocalisp.wordpress.com/2010/04/21/a-proper-constant-... Not sure about Coq, but Agda is Turing Complete (you just need to turn off termination checks) and Epigram also supports general recursion…

Coq is not turing-complete, but IMHO, completeness is overrated :)

I was trying to point that the "more powerful" criteria is pretty subjective.

The OP said "more powerful than the alternatives", that usually means "runs in the JVM"... and AFAIK, Jaskell or CAL are far from mature.

Post by Yeroc » Thu, Jan 12, 2012, 8:13 PM UTC

soc88 wrote: post #86

jerf wrote: post #22

Earlier quoted context omitted.

"As a pragmatic user of Scala I don't see why the fact that you _can_ do some amazing but cryptic things with the language is a detriment." Because you hit the cryptic too soon, and you can't really avoid it. And I generalize this to any such language that has this characteristic, not just Scala. You can't use C++ for very long without having to know huge swathes of it if you want to use the libraries created by othe…

This is not my experience. I don't give a dime about higher-kinded types, type bounds, type views, type constraints, CanBuildFrom ... and I'm happily using the language.

You must be one of those lucky people that only write new code all the time and never have to maintain code written by someone else. For the rest of us poor suckers who have to do a mix of writing new code and maintaining existing code bases written by other sadistic (!) developers then it's a very valid concern.

Re: True Scala complexity

Re: True Scala complexity