[lojban-beginners] Re: nitcu le denpa lerfu fi ma

["Jorge Llambías" <jjllambias@gmail.com>]

On Dec 27, 2007 10:45 AM, Michael Turniansky <mturniansky@gmail.com> wrote:
>
>   ienaicai! The whole point of la'o is that NOTHING is parsed inside of the
> quoted string.  It's foreign.  You can use sounds and tones that don't exist
> in lojban and, yes,  even pauses in the middle of words.

What do you mean by "is not parsed"? Its syntax is not parsed, but its
morphology necessarily has to be treated by the parser. The parser
has to identify each unit of input either as one of the Lojban phonemes
or as a non-Lojban character/sound.

> That's why the ONLY
> thing that resumes the lojbanic parsing is the close quote, which is the
> pause and the  identical sounds used at the beginning.  Everything is a
> black box as far as lojban is concerned, and the whole la'o X Y X just
> ="[some non-lojban name]"

That's the semantics, yes, but we are discussing a more basic stage,
identifying X, Y and then X again.

> But I guess we're going around in circles, here.
> I understand your point.  It's IF we were parsing inside the la'o quote, so
> that "cmavo" would fall off the front of words that would be considered
> gismu, lujvo, fu'ivla or cmavo clusters, then we can't use those as guard
> words, and  using cmevla as guard words is okay, too, since if they were to
> be interpreted as guard words, they would have to have a pause AFTER them as
> well. Simliar reasoning applies to using brivla as guard words, I suppose,
> too (I haven't worked out all the permutations).

Yes, any word can be used as delimiter, be it cmavo, brivla or cmevla. The
only exception is the "word" {y}, although I think it is better not to think
of it as a word.

> But I contend none of this
> is germaine because we are NOT parsing inside the la'o except lsitening for
> a pause and the sounds.

I think you are not treating lexing as part of "parsing", but the Lojban parser
has to deal both with morphology and syntax. The method you are thinking
of requires that you can't do the lexing independently of the syntax. You
need to know that you are within a ZOI-quote and do lexing stuff differently
than if you were outside. That's fine for humans, because we do mix syntax
and lexing together, but it's much more complicated from the fomal grammar
point of view.

> The grammar says ya can't put the sound (or string
> if written) anywhere in the quoted string, and I'm holding to that.

That's what CLL says, yes. We don't have an official full formal grammar
yet, so it may happen that we end up having something like that officialized
too, but I doubt it because it is much more complicated than doing the
lexing indpendently of the syntax. In fact the formal grammars that appear
in CLL take for granted that the lexing has already been resolved
independently of the syntax.

> I think
> this discussion has reached an end, and certainly any more would not belong
> on the beginner's list

We can continue somewhere else if you want, I enjoy these discussions
and they are in fact issues that will have to be decided at some point.

mu'o mi'e xorxes