Re: [lojban] Re: ambiguity in lojban

[John E Clifford <clifford-j@sbcglobal.net>]

For most of its history, Loglan/Lojban's claim of unambiguity was cashed out as claims that 1) the
speech stream was uniquely decomposed into words, 2) the word string was uniquely parsed and 3)no
words are polysemous.  Both of the first two claims were false when first made but served as
goals.  They are now pretty much true (I haven't seen a clear case of a problem in quite a while,
the issue of cmavo in names aside.) The third is always suspect for a living language and, insofar
as it is true, points to Lojban's deadness, which no doubt will be solved some day.  And various
pressures from an Academy may well keep polysemy to a minimum. 
When Loglan started, language data processing was barely fetal, let alone in its infancy.  And the
first problem to solve was parsing (you can't figure out what a sentence means until you figure
what it is).  Even today, the best you can get for a natural language sentences is an array of
possibilities and perhaps some attached probabilities -- even with content and context taken into
account.  Of course, this is often more than enough for practical purposes, but not having to go
through the processing involved would speed thing up considerably, as would avoiding the hard
cases. Hence Lojban's apparent virtue. No one in his right mind ever claimed that Lojban was
ambiguity-free in a broad sense, nor even that it was significantly better at extention issues
than other languages (bracketting the possibility that Lojban words are -- or will be when the
language is filled out -- narrower than those in other languages.  This actually seems improbable
to me, a firm believer in what you gain in one place you lose in another). It is good to be
reminded of this, though the relative percentage of problem in each of these areas is open to some
question (and prgamatic problems are left out here altogether).

--- Chris Capel <pdf23ds@gmail.com> wrote:

> On 6/20/06, Yanis Batura <ybatura@mail.ru> wrote:
> > I think no one has ever pretended that Lojban can 10 times ease the
> > understanding of the language by a machine.
> 
> I think some of the older texts did imply that Lojban would be a *big*
> help in machine language comprehension. I haven't seen anyone really
> advocate this recently, though. I bet it's been many years since this
> position has really been plausible--maybe in the early 90s it was much
> more defensible. But I wanted to throw it in there anyway. :-)
> 
> Chris Capel
> -- 
> "What is it like to be a bat? What is it like to bat a bee? What is it
> like to be a bee being batted? What is it like to be a batted bee?"
> -- The Mind's I (Hofstadter, Dennet)
> 
> 
> To unsubscribe from this list, send mail to lojban-list-request@lojban.org
> with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
> you're really stuck, send mail to secretary@lojban.org for help.
> 
>