Mission statement and Knuth-Plass reconsidered

[G. Branden Robinson]

At 2023-04-30T02:42:33-0500, Dave Kemper wrote:
> On 4/27/23, Douglas McIlroy <douglas.mcilroy@dartmouth.edu> wrote:
> > "Semantic newline" warnings are relatively innocuous. The occasional
> > pitch on this mailing list for paragraph awareness is far less so.
> 
> If you're referring here to changing groff's line-by-line processing
> to paragraph-at-once processing (via Knuth-Plass or similar
> algorithm), that's a central part of groff's 2014 mission statement
> (http://www.gnu.org/software/groff/groff-mission-statement.html, where
> it's billed as an "exciting challenge") and had widespread support
> when the mission statement was being assembled some 10 years ago.

Maybe we should revisit/amend/re-ratify the mission statement since it's
been about 10 years.  I've seen at least one prominent subscriber to
this list express reservations about whether Knuth-Plass is so great
after all.  (I don't have a citation handy.  It's somewhere in the list
archives in the past 5 years.)

I don't think that view was necessarily an expression of sour grapes.
If I remember correctly, the observation was that K-P can cause slightly
irregular line lengths from one paragraph to the next.

Myself, I wonder if K-P couldn't be implemented above the formatter
itself, using a diversion.  We could then put the implementation in an
auxiliary macro package.  Since I have plans to attack our facilities
for diversion re-processing anyway, it might be a good time to identify
any feature gaps we have that would make doing K-P this way more
difficult than it needs to be.

I don't have any serious concerns with the Mission Statement as it
stands, though I don't interpret it as some people do.  It says:

"Future work on manpages will entail improving the semantic clarity of
the man(7) macros, decoupling them as much as possible from low-level
presentational requests. The aim will be to ease conversion of manpages
to markup languages that do not rely on groff for display and printing,
e.g. XML, while preserving the full presentational richness of manpages
processed with groff."

It goes on to say:

"Concurrent with work on man(7), mdoc(7) will be actively supported and
its use promoted."

...which some people seem to interpret as "mdoc(7) will be promoted to
the detriment of man(7), which we deprecate".

But that interpretation is absurd given the immediately preceding
paragraph.

Lord, save us from mdoc(7) partisans.

Probably my biggest grievance with our Mission Statement is that it
says "manpages" instead of "man pages".  ;-)

Regards,
Branden

signature.asc
Description: PGP signature