Re: Reversible sound change applier
From: | Alex Fink <a4pq1injbok_0@...> |
Date: | Thursday, May 11, 2006, 2:23 |
On Wed, 10 May 2006 07:05:29 -0700, Aidan Grey <taalenmaple@...> wrote:
>Actually, check out IPAZounds, by Jamie Norrish. Not only does it do
reverse application, but it also uses IPA or XSAMPA, can group rules (you
write the rule for nasalization once, and then simply refer to Nasalization
whenever you need it again), can associate dates to rules (these rules all
happen before date 4, which = Middle Bobbic era), can associate dialects
too, and will shortly be able to do reverse by date and dialect as well.
I've found IPAZounds, which I dimly recall having known about, though I
never looked into it properly. It looks quite fully featured, yes. So
much for not having reinvented the wheel...
perhaps my only contribution with rsca is the generality of group
references, then.
I wonder exactly how it reverses, though? Having a strong theoretical bent
(could you tell, the way I rave about transducers?), one of the things I
like about rsca's approach is that the forward and reverse applications of
any transformation are (with a few exceptions) precisely inverse relations
on strings. A cursory glance at the source suggests that IPAZounds takes a
regular expression approach, and it seems to me much harder to guarantee
this property with regexps when the environments of changes can overlap
their domains of application.
Alex