From: SonOfLilit Date: 2007-02-13T05:58:43+09:00 Subject: Re: [OT] the funniest thing ever ------=_Part_60829_13220491.1171313899484 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline On 2/12/07, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > I have to disagree with the paper a bit. It seems that the author is > evaluating languages in a vacuum without practical considerations... > typical for academia. > > If he likes OCAML so much, is there a mod_ocaml for apache? are > there MIME, SMTP, IMAP libraries available? Is there a a nice > framework like Rails to build with? (just picking on it because the > author likes it so much. I've never used OCAML) > > The pragmatic side of me says I don't have time to look at a language > that can't get the job done. (even if you add "yet" to that). > > More importantly, the scalability of a language depends more on the > programmer than the language. I've seen scalable and brittle code > in almost all of those languages. His only point is that there is correlation between scalability and applicability of a language. The rest is evaluation of scalability of different languages. Does the lack of a mod_ocaml make it less sacalable? No. It makes it more applicable. Which isn't the same. Does the hampered (is that a word?) scalability of C++ make it less applicable? IMHO, it does. The scalability of a language isn't dependent on the programmer. The scalability of a language is the inverse of how much the language gets in the way of writing scalable code (I think it's more fitting than "how much the language helps the programmers write scalable code", though that can be used too if we talk paths of least resistance). Aur Saraf Just my NIS.05 ------=_Part_60829_13220491.1171313899484--