In message <3655.1070268454@splode.eterna.com.au>, you wrote:
> 
> so, if you initially work on the non-generated man pages that
> would been a good thing.  after that's done we can think about
> the generated ones...

Hello!

I have just started working on the non-generated man pages.
To check that all is ok, I wrote two patches (one for gzip(1)
and the other for znew(1)) that are available at:

    http://condmat1.ciencias.uniovi.es/~sobrado/man2mdoc.html

Can someone, please, try those patches?  Those patches can be
applied to a local copy of the man pages and checked by running:

    $ nroff -mdoc gzip.1 | more

and

    $ nroff -mdoc znew.1 | more

Patch for gzip.1 is a bit large, sorry!

*** Some important tips ***

I made some important changes to the gzip(1) man page:

1. The line:

       file: already has .gz suffix -- no change

   is wrong because both it accepts other extensions when identifying
   previously compressed files, and it says "unchanged" instead of
   "no change".  For example:

       $ cat > test.txt
       I will rename this file before compressing
       ^D
       $ mv test.txt test.txt.Z    (now it looks as a compressed file)
       $ gzip test.txt.Z
       gzip: test.txt.Z already has .Z suffix -- unchanged

2. Ascii should be ASCII.

3. Changed

       for csh: (setenv GZIP -q; tar -xfz --block-compr /dev/rst0
   to
       for csh: setenv GZIP -q; tar -xfz --block-compr /dev/rst0

   in the CAVEATS section (it is a typo, am I wrong?)

On the other hand, I do not now if gzip should be typed as a command
name (.Nm) when it appears in "in gzip format".  I believe that it is
not a reference to the gzip binary itself.

If all is ok, I will continue working on gzip.  I want to start
working in postfix soon.

All feedback on current patches will be really important.

Best regards,
Igor.

-- 
Igor Sobrado, UK34436 - sobrado@acm.org