Aaron Van Couwenberghe <[EMAIL PROTECTED]> writes:
> On Thu, Feb 03, 2000 at 12:23:06AM -0800, Chris Waters wrote:
> > It's a bug report against lintian, not policy, but maybe it's
> > something we should discuss and decide on this group anyway: should
> > lintian report a warning for the use of undocumented(7)?
> >
> > (I note from rereading that Gecko seemed to think I was trying to
> > force a policy change -- in actuality, while I'd like to see policy
> > change, I'd like to see the lintian warning in any case.)
> Actually, this bug *should* be against policy. I agree with you on this
> point, but you have to follow procedure to get your idea acted on:
No, I don't actually think it *should* be against policy. I'd *like*
to see policy change, yes. But that's my preference, not a project
necessity. Frankly, I'd be happy if we merely eliminated *some* of
the abuse of undocumented(7).
And there already is a proposal, but it seems to be stalled. Everyone
agrees that it's silly to require man pages and then have half the
system use undocumented(7). But nobody agrees what to do about it.
> I'll be one of your seconds if you do this.
I've already seconded the existing proposal. Feel free to do the
same, or make your own proposal if you'd prefer.
At this point, I think a lintian warning would be more useful.
If we need an excuse, here's a simple one: policy requires that you
have a bug report on file if you use undocumented(7), but lintian
can't check this. (Not easily, anyway.) So, instead of erring on the
side of abuse, lintian should err on the side of caution, and issue a
warning to remind you that you need a bug report on file.
This gives the added benefit that people can hunt for man pages to
write by searching the lintian reports on www.d.o.
Plus it reminds people that they have a bug report on file (if they
do, which they'd better), which may prod them into fixing the problem.
And best of all, no policy change required. (Sometimes I think we're
spending a little too much time modifying policy, and not quite enough
time following it.:)
Anyway, I think it would be nice. Any dissent? If not, I'll reopen
the lintian bug as wishlist, since there has been at least *some*
support for the idea.
cheers
--
Chris Waters [EMAIL PROTECTED] | I have a truly elegant proof of the
or [EMAIL PROTECTED] | above, but it is too long to fit into
http://www.dsp.net/xtifr | this .signature file.
