Re: [lojban] Re: "la" rule

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

On Mon, Nov 13, 2006 at 12:48:05PM -0800, John E Clifford wrote:
> (I don't think  the {lAItl} can be read as a sumti {lai tl}).

Of course it can.

> > > Clearly, humans sort these out, shouldn't the parser do so as
> > > well? 
> > 
> > ...
> > 
> > Short answer: "No; machine recognition of ambiguous languages is
> > AI-hard."
> 
> But here the range of ambiguity is so small as to allow a short
> dirty run through alternatives  -- or so it seems intuitively.

If we just allow la willy-nilly, there is *no way* for a parser to
tell if a speaker who says {laSTIvn.LAItl} means {la stivn laitl} or
{la stivn lai tl}, except for knowing the semantics of the language.
With the current rule, only the latter is correct.

-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/