RE: [lojban] the formal grammars' utility

["And Rosta" <a.rosta@dtn.ntl.com>]

Lojbab:
> At 02:16 AM 07/15/2001 +0100, And wrote:
> >Lojbab:
> > > At 04:40 PM 07/14/2001 +0100, And Rosta wrote:
> > > >1. What use is the EBNF grammar, given that it can't be used instead of
> > > >YACC?
> > >
> > > Some people find it easier to understand than the YACC grammar because the
> > > rules are compressed into a more language-grammar-like form.
> >
> >Am I mistaken in thinking that the EBNF in the book contains less info
> >than the YACC?
>
> According to the guy who did it (originally Cowan, but later another person
> - Clark Nelson, I think his name is) it SHOULD contain the full language.

You do realize that it's not that I think I have cleverly spotted subtle
lacunae in EBNF. Rather, I simply couldn't find the details of, say,
YACC's vocative_35 in EBNF. (But I can now.)

> >  For instance, trying to answer my own question of earlier
> >today about DOI, I find that "vocative415 = (COI [NAI])...& DOI", but
> >find nothing about the expansion of DOI, whereas I do find this info
> >under vocative_35 and DOI_415 in YACC.
>
> I think it is note (6) before the BNF text that says that the symbol "&" in
> A & B means
> A
> or
> B
> or
> A B
>
> so the rule for vocative415 allows
>
> COI(NAI)
> DOI
> COI(NAI)DOI

I understand this. But COI and DOI expressions contain cmevla or selbri
or relative clauses. However, I now realize that I should have been
looking at EBNF free32; I hadn't noticed that the EBNF is more than a
notational variant of YACC.

> > > >2. Is there a downloadable version of YACC ordered alphabetically (or
> > > >in any way such that one knows whereabouts in the rule list to find the
> > > >expansion for a given node)?
> > >
> > > The YACC grammar in the Book has an index that provides this.
> >
> >No, the index lists for each phrase the phrases it occurs within, but
> >it doesn't tell us whereabouts in the YACC grammar text itself the
> >rule can be found; one has to flick through the pages hunting fairly
> >randomly.
>
> I'm not sure what you are looking for then.  All of the rules have numbers
> on them and the YACC grammar rules are in numerical order, so that finding
> the definition of a given rule is trivial.

It's trivial only if either (a) you have the nous to notice that the YACC
grammar rules are in numerical order or (b) someone points it out to you.
I turn out to be someone who has required (b). I reckon there can't be
many Lojbanists as stupid as me.

My stupidity having been established, i would add that it then took me
ages to work out where rules between 500 and and 800 are.

--And.

> Finding what other rules use
> that given rule takes an index.
>
> lojbab
> --
> lojbab                                             lojbab@lojban.org
> Bob LeChevalier, President, The Logical Language Group, Inc.
> 2904 Beau Lane, Fairfax VA 22031-1303 USA                    703-385-0273
> Artificial language Loglan/Lojban:                 http://www.lojban.org
>
>
> To unsubscribe, send mail to lojban-unsubscribe@onelist.com
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>