Re: [xml-dev] XML 2 so far

[Amelia A Lewis <amyzing@talsever.com>]

On Sun, 12 Dec 2010 20:42:50 -0500, Liam R E Quin wrote:

[big snip]

> What did I miss?

Namespace-related proposals.  :-)  Even if you don't care for them, 
they've been a significant part of the traffic.

> People have talked about every well-formed XML 2 document also being
> an XML 1 document, or (more controversially) about having a well-defined
> conversion.

I think that this design is a big win.  The reason?  You don't have to 
write a new parser in order to benefit.  For this to work, there has to 
be value to document creators (and low-cost entry to document 
manipulators, which is probably true-ish for the dynamic languages more 
so than for the strictly-typed crowd), but it would mean that the cost 
of adoption would be radically lower than the cost of wholesale 
replacement.  Moreover, with a single (smallish) transformer/converter 
dropped into the pipeline, existing tools would continue to work.  
Adoption would then be driven by folks finding value in the modified 
syntax, driving the parser and processor developers to consider 
adoption.  I think it's an attractive model for evolutionary 
development of a document format, set of APIs, and collection of 
processors that are in widespread, effective use.

The permitted changes *are* thus limited, though.  Things like 
namespaces, minimization, and probably some comment sorts of changes 
would be possible.  Changes to DTD support are less clearly possible.  
Basically, anything that can map one-to-one to XML 1.0 would be fine 
(note that it would *not* be necessary to map XML 1.0 to the new 
format--so this would mostly be wart removal and bug avoidance, not 
massive functionality enhancement).

Amy!
-- 
Amelia A. Lewis                    amyzing {at} talsever.com
  "Ruby fruit jungle?"
  "Yeah, women are thick and rich and full of hidden treasures and 
besides that, they taste good."
                -- Rita Mae Brown, "Rubyfruit Jungle"