Re: [lojban] Re: Is Lojban a CFG? (was Re: [lojban-beginners] Re: Enumerating in Lojban)

[Robin Lee Powell <rlpowell@digitalkingdom.org>]

On Wed, Jul 12, 2006 at 02:16:06AM -0400, Jonathan Gibbons wrote:
> Forwarding the message I sent to the beginners list, because I
> took far too long to catch on to the moved-ness of this thread.
> Also continuing.
> 
> >le nu le broda brode brodi
> 
> Okay, now that one would be what I was looking for, I think. Okay.
> Tossing that out as ungrammatical would appear to make Lojban into
> a non-context-free language (being mildly nitpicky; the question
> of "Is Lojban a CFG?" is trivially no; very few languages (sets of
> strings) are context-free grammars (sets of production rules). "Is
> lojban a CFL?" is, however, what it seems intended as. And, much
> to my sadness, it seems to be a no; why it has been defined not to
> be I don't quite understand, but that does look like the
> counterexample; I still hold that it's a very odd way of forcing a
> language into unambiguity, making it non-context-free.)
> 
> Perhaps name the language I seem to be going to construct
> "narvablojban"? a language that is a proper superset of Lojban
> that's /actually/ context-free as well as semantically
> unambiguous, rather than just close.
> 
> -A very sad Jonathan

Well, I didn't want to make you sad.

Here's my shot at why it's like this:

1.  "Not a CFG" != "not formalizable".  Lojban is still formalizable
and easily machine parseable (witness the PEG formalism, for
example).

2.  Elidable terminators seem to make intuitive sense to humans.

3.  The lanugage would be *very* unwieldy as a CFG: way to many
syllables would be expended in required required terminators.

-Robin


-- 
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"
Proud Supporter of the Singularity Institute - http://singinst.org/