[bug #62233] [doc] Stop referring to Roff as a group of languages

[G. Branden Robinson]

Update of bug #62233 (project groff):

                  Status:                    None => Need Info              
             Assigned to:                    None => gbranden               

    _______________________________________________________

Follow-up Comment #1:


[comment #0 original submission:]
> Our Texinfo manual uses the term _"roff languages"_ (plural) instead of
_"[the] roff language"_ (referring to a specific entity).

Yes.  This term is, I think, my coinage, and if it's not, I've done a lot to
propagate it through our documentation.

Bernd used the same term as an umbrella under which to group the lexica of
macro packages; I don't find that a worthwhile usage since none of these
differ in _syntax_.  I rewrite these when I happen across them.

> The former wording is misleading and inconsistent with docs that (correctly)
refer to Roff as a solitary programming language,

What documents are these?

> albeit one that's been extended several times by various implementations.

I'm resistant to making such a change.

I guess the main reason is that I cannot see that anyone anywhere has formally
defined "the roff language".  I've never seen a BNF-style grammar for it, and
the section "Text" I wrote for our Texinfo manual is the closest thing I've
seen to an _informal_ survey of the syntax.  (One can distill such a thing
from CSTR #54, but that document doesn't try to present the syntax per se,
instead admixing it with a comprehensive survey of features.  Admittedly, my
approach isn't pure either, because it's larded with hand-holding pedagogy.)

By saying "roff languages", I do not suggest the existence of a single formal
specification of the grammar (which doesn't exist--wait, here comes somebody
to tell me that Unix V7 troff C source files with a hand-written recursive
descent parser constitutes one--no, it doesn't, and if it did it would
invalidate the identification of groff as a "roff language", which is probably
the point of the intervention, so go away), and I create uncertainty in the
reader's mind regarding the existence of same.

That's good!  Or, at least, it fails to pretend, which is better than lying.

Do you see the method behind my madness?  Where does there exist a formal spec
of "the roff language" that will admit Unix V7 troff, Heirloom Doctools troff,
and GNU troff?


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?62233>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/