Re: * RL * code review and strategy for macros set?

[Mike Bianchi]

On Wed, Aug 12, 2020 at 06:07:21PM +0200, Marc Chantreux wrote:
> hello Mike,
> 
> first of all: pardon my lack of culture but i don't get what "RL" means
> in this case. i suppose it's a way to say you replied intentionally to
> me and not to the mailing list but have no confirmation of that�.

RL has to do with the mail filtering I have.
It means nothing to anyone else.

 
> it is sad that the rest of the mailing list can't read this very
> interesting answer.

That was a goof on my part.  I'll add the groff mailing list to this response.

 
> On Tue, Aug 11, 2020 at 10:44:53AM -0400, Mike Bianchi wrote:
> > I would suggest you turn the list upside down and start with something that
> > _works_ even if not to your liking.
> 
> i gave roff a try in circa 2000 and i hated it. i started again since
> the begin of the year because i love the man command and wanted to write
> my own manuals.
> 
> it quickly appeared to me that
> 
> * compared to the other typesettings systems (including the very
>   overated TeX), troff isn't that bad.
> * my only sine qua none is fixed by the -k flag
> * now understand that some syntax "limitations" actually really
>   ease many automations
> * nothing compares when it comes to be fast and lightweight.
> 
> so i started using mm, me and ms and got results that are far enough
> for me but if i want to go further, i should be abble to tune the
> documents to comform the visual chart of my employees for exemple.
> 
> if i can be good enough at it, maybe i can introduce roff in the
> communities i work for.
 
> > The *roff commands are an _assembly_ language with lots of hidden states,
> > arcane rules and interactions. It's value is that it works and that _lots_ 
> > of
> > smart people mastered it to such a degree that some _really_ clever things 
> > have
> > been done with it. Macro sets are the means of doing the clever things.
> 
> to explicit one question i have: should a high level macro set rely on
> lower level ones like ms (as it seems ms is't as rich and opiniated than
> me, mm and mom)?

I strongly recommend you pick one macro set, read and become fluent in the
the concepts in the command macros manual for that set and use the built-in
tuning features to make them look the way you want.  You are _not_ the first
person to want it "just like that, but different" and you will find that the
macros sets are built with fine-tuning in mind.  I have used the mm package
since it's invention and have always used a set of its fine-tunings to make
it specific to my needs.  

         export GROFF_TMAC_PATH=${HOME}/lib/tmac
         groff  -Kutf8  -GtpeR  -U  -rW6.5i  -mm  -mFm  ...files...

where  Fm  refers to the "Foveal macros" that tune the  mm  macros.  They are
found in  ${HOME}/lib/tmac/Fm.tmac .

There I have things like:

        \#      A mark list of checkoff boxes.
        .de CheckList
        .ML "\s+8\(sq\s0  " 7
        ..
        .de CheckListEnd
        .       LE 1
        ..


> > Said another way, many have gone before you and pushed the rock far up the
> > mountain.
> 
> i'll take my time on it as i'm convinced mastering troff is worth it.
> 
> > Skim and then re-read the *roff command manual of your choice.
> > I suggest groff.
> > ... [you describe a interesting path there ] ...
> 
> thanks for this widsom. I just copied m.tmac as ike.tmac and will
> modify/tune it until the documents i already written fit the look
> i acheived in my demo.

I advise _against_ modifying the standard *.tmac files, especially with the
sets like  me ,  ms  and  mm .  They rarely change, but when they do you loose
the benefit of the fixes.


> > When you think you have the basics down, go to the macro package of your 
> > choice
> > and attempt to understand the simple macros for the simple concepts.  I use
> >  mm  so I would start with
> > 
> >     .P      Paragraphs
> >     .SP     line SPacing
> >     .HU     Header Unnumbered
> >     .R      Roman font
> >     .B .I   Bold  Italic
> 
> i'll do that one by one.
> 
> > ((From someone who has been writing nroff/troff/groff since the late 
> > 1970s.))
> 
> impresive :)

No.  Just old.


> in a sense, i envy you: it feels to me that computers were about for and real
> hobbists back then so the digital culture was much more inspiring than 
> nowadays.

I was fortunate enough to be at Bell Labs when the UNIX Programmer's Work Bench
was happening and in one of the first groups to do our PL/I code development on 
a UNIX PWB machine that took the place of a card puncher, punch card reader and
printer.  Said another way, it was a work environment making use of a new
Bell Labs innovation.  It led into a career.


> �: The New Hacker's Dictionary definition of RL is "Real life"
>    (http://www.catb.org/jargon/html/R/RL.html)

-- 
 Mike Bianchi
 Foveal Systems

 973 822-2085

 MBianchi@Foveal.com
 http://www.AutoAuditorium.com
 http://www.FovealMounts.com