Re: policy summary
Matthew Vernon <matthew@sel.cam.ac.uk> writes:
> I still maintain that undocumented(7) serves a useful function (as my
> objection said), but perhaps it should be reworded somewhat; maybe we
> should encourage people to check the BTS and submit a bug report, or
> simply "do submit this as a bug if there isn't an appropriate one
> open", since it seems that there often isn't the bug open that
> undocumented(7) currently says there is.
I think the problem is really that lintian is too picky about lack of man
pages. At least, it's impossible to get a package by lintian without setting
undocumented links for every binary without a man page, which is a sure way to
get maintainers to do so.
Put another way, undocumented is superfluous now because we do in fact know
about every missing man page because lintian tells us.
Having a man page that says "there's no man page" isn't great. I would prefer
to see this built into man as a default page that the user or admin could turn
off and that would still provide an error return instead of a successful
return with a man page saying it wasn't successful.
It's a bit late to deal with this now for the freeze though, and there are
much more important problems. The current situation isn't so bad really.
--
greg
Reply to: