Re: mandoc shortcomings

["James K. Lowden" <jklowden%schemamania.org@localhost>]

On Mon, 7 Dec 2015 15:53:49 +0000 (UTC)
christos%astron.com@localhost (Christos Zoulas) wrote:

> > /usr/share/man/man7/groff.7.  
> 
> I am actually amazed that it does this well given that it is not
> implementing the full roff macro set, and this is really not a man
> page since it uses low level roff macros (instead of restricting
> itself to the -man or -mdoc set).

I agree with Mouse: there has *never* been a constraint that man pages
must use -man or -mdoc macros exclusively.  I'm sure there was never a
release for which that was true of every man page.  

What is the official position, then?  It has always been my
understanding that correct man pages were part of NetBSD, and that a
page that fails to render correctly -- whether because of a bug in the
formatter or in the markup -- was considered a bug.  

Thor seems to think I'm making a mountain out of a molehill.
After all, what's one man page between friends, especially while groff
is still in base? My question is, how many incorrectly rendered pages
constitute a bug now, if not one?  

It seems to me we have two valid choices, neither of which we've
adopted:

1.  Use mandoc, and ensure that everything in base and pkgsrc is
compatible with it.  

2.  Use something else (probably groff for now) that renders all man
pages in base and pkgsrc.  

Blindly adopting mandoc and letting the bits fall where they may is not
a recipe for utility, let alone greatness.  

--jkl