chromatic wrote:
On Monday 30 June 2008 15:03:14 David Cantrell wrote:
Surely you can at least check that all POD is "well-formed" without
running any code from the distribution in question?
Sure, but that's a very different question from "Did the author write useful
documentation for everything that needs public documentation?"
which is why I said "at least". I was under the impression that
Aristotle was claiming either that you couldn't do that (which I
believed to be incorrect, and I'm glad you can confirm that), or that
even if you could it was useless on its own (which I also believe to be
incorrect).
> as "Are there
any POD errors?" is statically determinable. "What do useful, proper,
public, and everything mean?" is not.
I think that "POD is well-formed" is clearly a Good Thing To Measure,
even if it falls short of being the perfect measure of POD, and should
stay as part of CPANTS. We're down to arguing about how to implement
that check - whether it should be "ooh look, there's a t/pod.t file
which says 'use Test::Pod'", or whether CPANTS should itself use
Test::Pod to look for badly-formed POD.
I favour the latter, although I find that having the Evil Cargo-Culted
t/pod.t in my distributions is rather useful for finding bugs before I
release my code so I'm gonna continue using it.
--
David Cantrell | Godless Liberal Elitist
It's my experience that neither users nor customers can articulate
what it is they want, nor can they evaluate it when they see it
-- Alan Cooper
