Re:Ask HN: Is there evidence regarding PG's theory about pow...
by kristianp on Tuesday April 12, 2016 @08:26PM
(#11484754)
Interesting that no-one has pointed out here the dynamic vs static or scripted vs compiled dimensions.
Facebook's use of php arguably allows them to build new features quickly, just as Viaweb's use of Lisp. But is it the dynamic nature of those languages that allows the faster development, not the 'power'?
It seems accepted that dynamic languages allow faster development, but static languages are more able to prevent bugs. It varies from language to language of course.
Apparently the research studies on static vs dynamic have many problems though, so you can argue that there is no evidence that one is "better" than the other: http://danluu.com/empirical-pl/
Maybe we should have a coding competition that has an equal number of (lisp, perl, php, ruby, python) experts vs (haskell, ocaml) experts and see how fast they can add features, vs how buggy those features are. Haskell and Ocaml are powerful, but 'very static', so it would be an interesting comparison.
Facebook's use of php arguably allows them to build new features quickly, just as Viaweb's use of Lisp. But is it the dynamic nature of those languages that allows the faster development, not the 'power'?
It seems accepted that dynamic languages allow faster development, but static languages are more able to prevent bugs. It varies from language to language of course.
Apparently the research studies on static vs dynamic have many problems though, so you can argue that there is no evidence that one is "better" than the other: http://danluu.com/empirical-pl/
Maybe we should have a coding competition that has an equal number of (lisp, perl, php, ruby, python) experts vs (haskell, ocaml) experts and see how fast they can add features, vs how buggy those features are. Haskell and Ocaml are powerful, but 'very static', so it would be an interesting comparison.