Re: new util method proposition for Maketext

At 15:30 2002-09-05 +0200, Guillaume Rousse wrote:
>[...] The goal is to be able to have correct output for such 
>phrases:  "voulez-vous créer un nouveau X", where X is a noun of unknown 
>gender and first letter. [...]

I want to give this some very careful thought, as it involves a part of 
language that's very important, but where it's notoriously hard to find 
patterns that are intuitive AND applicable across languages AND don't 
overcomplicate the the simple common cases for the sake of handling rarer 
messier cases.

I'm very slow in managing to think about these things, because I'm too used 
to the methodologies in formalist (as opposed to functionalist) 
linguistics, where everyone wants every theory to work for every part of 
every language -- but clearly what's called for with the Maketext approach 
is a good mix of intuitive interfaces/notations and of "80/20" 
solutions.  (And that's to say nothing of the modern formalist fear of 
notations that are "too powerful"!)

Just now, I'm terribly busy with a grand rewrite of the basic Pod 
parser/formatter classes, but I'm hoping that in a week or two I'll have 
something clever to say about your suggestions.

--
Sean M. Burke    http://www.spinn.net/~sburke/