[lojban] Re: jbovlaste natlang word voting question

[Robin Lee Powell <lojban-out@lojban.org>]

On Sun, May 13, 2007 at 04:08:35PM +0200, Arnt Richard Johansen
wrote:
> On Sat, May 12, 2007 at 08:55:39PM -0700, Robin Lee Powell wrote:
> > 
> > Someone pointed out that you couldn't acutally derive that
> > second place of sampli from its components, so I've corrected
> > that.  Try http://jbovlaste.lojban.org/natlang/en/ball instead
> > for an example of multiple natlang usages; if the dictionary
> > said something like "ball - first place of bolci, second place
> > of selboi", would that be a bad thing?
> 
> _In this particular case_ it would be a bad thing, because if we
> already know that ball is the first place of bolci, then it is
> kinda sorta tautological that it is the second place of selboi.
> 
> But I do think it is a good idea to have the _possibility_ of
> multiple Lojban word for one natlang word. A good example of why
> this is necessary is:
> 
> http://jbovlaste.lojban.org/natlang/en/science fiction
> 
> Here, there are two "live" definitions for "science fiction". More
> are still languishing in Noralujv. All are subtly different, and
> should be pointed to by the natlang entry for "science fiction".
> But it does not make sense to split "science fiction" into
> different senses, because it's not really polysemous.

Having thought about it more, both of these cases can, and IMO
should, be handled via modification of the definitions.  That is, if
ball == x2 of selboi shouldn't be in the dictionary, it shouldn't be
in the definition either, because it's useless.  That loses some
information, I know.  My question is, is that loss of information
worth the interface streamlining of not having to vote for
natlang words?  Because the two-stage voting is, IMO, a PitA.

-Robin


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.