> Consider the case where a word from vocab A calls a word form vocab B with a
> quotation calling a word from vocab A. So you have A -> B -> A

I have considered that case. The point you're making would be a deal-breaker 
for my proposal if this were the only case. But it is not the only case. In 
the scenario you present, it would be important to consider "what words in 
this vocabulary are expressly designed for being called by other unrelated 
vocabularies". Those words are candidates for having a more controlled 
encoding environment, rather than relying on inheritance. However, Factor is 
the king of languages with "helper" words. Helper (or private) words which 
are only used by other words in that vocabulary can be designed to inherit 
encoding settings safely. If some dufus comes along and uses something in my 
nuclear-powerplant.private vocabulary, he deserves to get radio-active 
unicode bits spewn all about himself.

> What you want is a vocab-local variable

I'm too conservative to go that far at this point. I currently think that 
would be unnecessary.

At any rate, we can discuss theoretical "A->B cases" till we're blue in the 
face. The Factor codebase is filled with real use cases. Each place where the 
current api is in the wrong, I'm going to demonstrate this.

This is the sort of discussion that should have happened (publicly) over the 
course of months; examing real, existing code; rewriting it using proposed 
api's, getting feedback, and more importantly, gaining consensus.

Ed

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Factor-talk mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/factor-talk

Reply via email to