Re: Namespaces, schemas, Simon's filters.

[Ronald Bourret <rpbourret@rpbourret.com>]

Peter Piatko wrote:
> So far on this list I've seen 2, well 2.5, interpretations:
> 
> (1) A ulabel should have a 1-1 mapping to its "meaning" (a word I am being
> intentionally vague about).
> 
> (1.5) Believes in (1) and introduces local elements in the hopes of some
> modularity.
> 
> (2) A ulabel can map to multiple meanings.  Other mechanisms (context, a
> schema, calling up someone on the phone) might also be employed to fully
> disambiguate the meaning.
> 
> When groups with different interpretations meet, sparks fly, harsh words are
> spoken

Which I think is the whole point of XML. Anarchy at the global level,
with locally non-chaotic bits where people meet to do work, but with no
technical basis for _everybody_ being able to get together except
parsing.

>, and Simon creates a filter to bridge the semantic gap.  ;-)  Maybe
> such filters are the best we can hope for.

Quite possibly -- both a strength and a weakness of XML.

-- Ron