Re: Type-assignment (was Re: Are we losing out because of grammars?)

["K.Kawaguchi" <k-kawa@bigfoot.com>]

> To distinguish the two types of ambiguity, we need two terms.  How about  
> ambiguity of datatype assignment and ambiguity of interpretation?

I strongly suggest to use "ambiguity of datatype" and "ambiguity of
labels", because when we say "interpretation", it should refer to
"things" that we get (datatypes, labels, or whatever)


> Kawaguchi-san's algorithm for detecting ambiguity of interpretation looks 

To be precise, in RELAX, when a grammar is unambiguous in terms of
labels, it automatically ensures unambiguity of datatypes.

In other words, unless the grammar is proven to be unambiguous in terms
of datatype, the grammar cannot be said to be unambiguous in terms of
label.

Am I right?


> Is it possible to detect ambiguity of datatype assignment by 
> examining a TREX pattern?

This's another interesting question... Let me see...



regards,
----------------------
K.Kawaguchi
E-Mail: k-kawa@bigfoot.com