The road ahead

Well, it seems like we have an agreed-upon spec.  We need to do a
few administrative moves, and then we should invite public comments.
After that, what's the road ahead?

1) We can decide that we're done.  We have James's Javascript parser,
and I'll be rereleasing MicroLark fairly soon, all nice and conforming.
Uche has a Python implementation that he'll need to tweak in the same way.
We could declare victory and go home.

2) We could talk about a micro-sized validation language for MicroXML.
Obviously, this is not a requirement, because any XML validation language
(XML Schema, RELAX NG, Schematron) will work fine with MicroXML documents.
But a low-overhead, limited-use, easy-to-implement validation language
might be a Good Thing to have.  I've proposed MicroExamplotron, which
uses a sample instance plus optional extra markup as a schema.

It supports a Russian-doll-style schema, where by default elements are
mandatory and attributes are optional.  By putting in the right attribute
values or element content, a few simple types are declarable, and when the
content of an element is mixed, this means that you get not only mixed
content but an arbitrary number of child elements in arbitrary order.
With specialized attributes, you get quantifier control and define-refer
semantics.  See http://www.w3.org/community/microxml/wiki/MicroExamplotron
for details.

3) We could discuss a simple transformation language.  Again,
XSLT is sufficient, but having something smaller may be handy,
especially in the absence of namespaces.  I have a MicroAF proposal
which allows elements and attributes to be renamed depending on the
value of a specific attribute.  Character content can be preserved or
discarded, and child elements can be processed, skipped, or left alone.
See http://home.ccil.org/~cowan/microaf.html for details.

-- 
When I'm stuck in something boring              John Cowan
where reading would be impossible or            (who loves Asimov too)
rude, I often set up math problems for          cowan@ccil.org
myself and solve them as a way to pass          http://www.ccil.org/~cowan
the time.      --John Jenkins

Received on Monday, 24 September 2012 21:56:01 UTC