RE: [xml-dev] "Introducing MicroXML, Part 1: Explore the basicprinciples

[Liam R E Quin <liam@w3.org>]

On Sun, 2012-07-15 at 15:10 -0500, Len Bullard wrote:
> I agree.  It is a tooling issue (e.g., the ubiquity of Arbor Text, Microsoft
> and Adobe in mil-spec logistics organizations) and as you say, a DTD issue
> with large extant instances.

Sometimes I'm still sad that SoftQuad Author/Editor isn't there too.

> As I said my comment is advisory.  To be fair, the multiplicity of standards
> making and consortia sources and a lack of coherence in their directions
> makes this a wee bit more daunting than in the days of SGML when at least
> ISO was the dominant source.

It's a pain, I agree; I'm not sure how to break past the Balkan States
of Standardization Organizations, the fiefdoms of focus.

Going back to processing instructions, some of the uses are served just
as well with comments, some, like <?php...?>, won't work at all with the
current µXML proposal, some would be better with elements - in XML these
days one might simply say, <dl:page-break /> or <ptc:cursor-loc/> and
use NVDL and/or a pipeline validator to handle them.

If MicroXML is to go anywhere it probably needs to be rich enough for
. XSLT to be expressed in it
. Web pages, yes with PHP, and, as John notes, HTML compatibility
. be a close enough subset of XML and (for the data model) of XDM that
existing software can process them. This last was a requirement for XML,
to be a subset of SGML and to work with SGML tools; in the end we needed
some minor changes to those tools, and, had we known that would be a
possibility at the start, some other decisions might have been
different. The future is longer (we hope) than the past, and if there
are more people on the planet now than have ever lived, the statistical
odds of death are now under 50% - so maybe microXML has a rosy future.
Similarly,
. be formattable directly with CSS - but the CSS WG is open to new
proposals, so this is not too onerous provided that the naming/namespace
mechanism ends up not being too bizarre.
. have a clear easily-understood relationship to JSON (documents vs
data-binding), to XML, to Web services and the existing XML stack.

A W3C community group would be a fine place to discuss requirements and
use cases and get a draft hammered out. Or maybe I'm biased there :)

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org freenode/#xml