[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
At 05:59 AM 4/30/03 -0700, Jorge "Llambías" wrote:
la djan cusku di'e
> I should rather say that it sometimes produces the wrong answer, notably
> with the way it groups multiple sumti and the way it handles numbers,
> but most of the time its parse tree is quite sensible.
> It just is not *determinative*.
One particularly jarring case where it fails is in things
like {mi na broda gi'e brode}, which is parsed as
{mi (na broda) gi'e (brode)} instead of
{mi na (broda gi'e brode)}. This is particularly bad because
the grammar goes to pains to produce the wrong parse,
when it would be straightforward to produce the correct one.
In fact, I think the current parse should be the correct one,
but for some reason it was decided that the other interpretation
is correct, against the parser. Either the grammar or the
interpretation should be fixed in this case, the status quo
is nuts.
Not remembering the reasons why John .a I decided to go the other way, I tentatively agree with matching the parse.
lojbab -- lojbab lojbab@hidden.email 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