[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[lojban] Re: morphology paper announced
coi xorxes
the idea of fu'ivla is that any string that follows the phonotactics
of brivla and cannot be confused with cmavo and/or lujvo is a valid
fu'ivla form. If the brkwords algorithm makes {.iglu} invalid, then
there is a problem with the algorithm, not with {.iglu}, which cannot
break into cmavo and/or lujvo.
I think the idea behind is that recognition algorithm must be as easy as
possible and must not require a complex analysis. You argue about {iglu}
because you SEE that {glu} is not a brivla. Nobody mistakes one-syllable
piece of text for a word. What about {adjgadja}? Do you still SEE that
{djgadja} is not a brivla? You suggestion is to cut off an initial cmavo
and then recursively apply word recognition algorithm to the rest.
(Well, I'm lying. Not recursively - some simplified version of the
algorithm is enough. In particular, the procedure in question - cut off
cmavo and test the rest - doesn't apply, so there is no recursion.
Nevertheless the applying part is still complex enough and includes such
non-obvious beasts as slinku'i test). On the other hand, my approach is
to apply a short sequence of simple patterns (specified in brkwords).
Just for an illustration: in my program my approach is implemented in
two lines of code, while your requires six (and four more auxiliary, but
it is rather poor design than task requirement, so I don't count them).
> There are only two potentially infinite initial cluster forms:
> (d[jz])+ and (t[cs])+, but who have decided to forbid them?
I did, on the grounds that it makes no sense to allow inintial
"tctctctc" while disallowing "sb". I think disallowing the affricates
from participating in further clusters makes sense.
Oh, I'm just whining... I've spent hours writing down explicit syntax
for initial cluster, and you suggest me throw it out. Cruel, cruel you!
Since the onset cannot be empty,
Oops! I've missed this. Thank you!
--
Cyril Slobin <slobin@ice.ru> `When I use a word,' Humpty Dumpty said,
<http://45.free.net/~slobin> `it means just what I choose it to mean'
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.