RE: more grist

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

So the InfoSet takes the role of the 
SGML Declaration sort of (not tied 
to a specific syntax)?

Doesn't that push us back to precisely 
where this chapter of the markup adventure 
started?  I can see the utility of a 
comprehensive base infoset.  My worries 
come after that.  This seems to come down 
to:

1.  There will be privileged XML vocabularies 
that have special semantics in XML processors. 
True already.  The provisions made for HTML 
were considered 'pragmatic' but set a scandalous 
precedent.  The inventors of HTML 
make those decisions.  House rules and choice 
of means of choosing means prevailed:  choose 
W3C, choose Berners-Lee.  That's not personal. 
That is the way the means are chosen.  It has 
worked until this hiccup.

2.  The rules for extending the privileged 
vocabularies or adding to their number are 
not known at this time and may never be.  
Since XML is W3C property, I assume the 
same means to choose means prevail.

When Henry or whoever can publicly provide a plan here, 
I'd like to hear it.  Otherwise, the critics are 
right about XML Schemas.  It is just another 
vocabulary.  It is a post XML infoset language with 
its own property sets that force loopbacks to 
the XML property set through the sharing of 
properties with other languages such as XPath.   

There will be no end of this and we will satisfy 
the naysayers (the relentless march of abstraction) 
by showing quite dramatically tnat not only 90% of 
the XML specs fail, but that XML 1.0 itself fails.  
FUD or unavoidable precipice?

So, well-formed anybody?  InfoSets next?  Sheesh. 
Groves and grove plans just like the HyTime guys 
said it would be.  I don't think that avoidable 
unless someone wants to redo the base infoSet 
to accomodate a limited set of vocabularies, 
but I think the slope is becoming ice quickly. 

Is Rick J. right? Do it but don't call it XML? 
Does that 'renaming' save us anything but 
admitting we need something like groves and 
grove plans?  In my opinion, once you take 
away lexical unification (same syntax), 
by definition, by Draconian rule, it quits 
being XML.

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Ben Trafford [mailto:ben@legendary.org]

         However, tying it to something like the Infoset, and through that, 
the sort of things you might see in XML Schema, or TREX, or RELAX, might 
make sense. That way, the implementation is tied to no specific syntax, 
while maintaining the features that schemas offer.