New results from "dictionary word count" programs...
Gregory_Hill at tni.com
Tue Mar 21 08:53:40 MST 2006
> > Isn't the LOC thing kinda bogus anyway since some programs are
> > on external modules or libraries? I always kinda chuckle when
> > say they can write something in 5 lines of Perl, but they don't
> > that it's because they are using 3 CPAN modules that weigh in at
> > 1MB each. Yeah, that's probably exaggerated, but you get the idea.
> > I'd be curious to see the LOC comparison of programs written with no
> > external dependencies.
> There's a direct relationship between LOC and how long it takes to
> something. This can be partially addressed by using a good IDE, if one
> is available for the language. Most important modern languages are
> actually platforms, to fully evaluate the language you have to
> what libraries are available with the language, so using a library or
> module is plenty valid.
I agree that it's valid for the benchmark tests, and I agree that when
comparing languages for practical reasons (as in, what language should
my company use), accounting for available modules is valid. But saying
that you can write something in 5 lines, but behind the scenes you're
relying on 50,000 lines of someone else's code, compared to another
program that does the same thing in 20 lines but has no external
dependencies doesn't reflect the relative power of the languages, unless
you account for the extra 50,000 lines of code. Anyway, maybe I'm alone
in that thought, but I'd still be curious to see an LOC count of these
languages if no external dependencies are permitted.
More information about the PLUG