RE: [dita] microXML and the future of DITA

["Grosso, Paul" <pgrosso@ptc.com>]

> -----Original Message-----
> From: Don Day (LbW) [mailto:donday@learningbywrote.com]
> Sent: Monday, 2011 January 10 0:26
> To: dita@lists.oasis-open.org
> Subject: [dita] microXML and the future of DITA
> 
> This appears to be an emerging standards discussion that will make
> future use of Web standards in DITA dicey if not managed reasonably:
> 
> http://blog.jclark.com/2010/12/microxml.html (and check the follow-up
> post as well)

There is currently no such thing as microxml.  It's just a
"marketing phrase" (or perhaps more accurately a philosopical
direction proposed by some people) to date.

Ideas related to this concept and the relationship of XML to
HTML5 have been discussed in the W3C XML Core WG and at various
levels of the W3C.  A new W3C task force has been formed to look
into issues related to the coordination of XML and HTML5. 

There is a public mailing list for the task force,
public-html-xml@w3.org, with public archives at
http://lists.w3.org/Archives/Public/public-html-xml/
All of the task force business will be conducted in public, there 
are no member-only mailing lists or archives. You are free to join 
the list and participate in the discussions there.  (You can
subscribe by going to the archive URL given above and click
on "subscribe to this list".)

The chair of the task force, Norm Walsh, is co-chair with me of the
W3C XML Core WG as well as chair of the OASIS DocBook TC.  However,
to date, the apparent power brokers on the task force appear to
be the HTML5 folks, most of who don't seem particularly interested
in bending over backwards to accommodate XML in general much less
a given complex application of XML.

> 
> For one, it would be nice if future HTML5 editors that were also
> microXML compliant could actually support DITA topic authoring.

There is no such thing as microXML or microXML compliance.

If you follow the discussions, you'll probably note that there is
very little interest in the HTML5 community of supporting what most
of us think of as real XML.

HTML5 is a (proposed, though already implemented in some browsers) 
language, not a standard for editing tools, so one can't say what 
HTML5 editors would support.

The question probably comes down to what it would take to transform 
HTML5 content into DITA content (which is almost certainly not 
something an HTML5 editor would care about).  And that probably has 
little to do with what HTML5 turns out to be as long as it turns
out to be something XSLT processors can read.  And that has more to
do with XSLT processors than HTML5.  (And for what it's worth, XSLT
processors will have a lot more interest in being able to read HTML5
than HTML5 people will have in it, so I do expect XSLT processors to
have modes that both read and write HTML5.)

> Likewise, it would be nice if HTML5 could evolve to support a default
> "topic" that DITA 2.0 could live with as a common base type. It would
> be nice if class-based extension could be retained, of course, although
> how we'd use that without a DTD or Schema beats me--I think specialized
> content literally on the Web is doomed outside of XML 1.0. Finally, it

Never going to happen.  HTML5 barely considers XML worth considering.
DITA is not even a blip on their radar.

> would be nice if microXML would support a formal content reference
> mechanism so that conref and topicref processing ideals could be used
> to aggregate content dynamically--I don't see anything in James's
> discussion that supports that use of microXML.

The point of microXML is to strip down XML to make it even simpler
than it already is.

DITA adopters are complaining that DITA is already too complex.

Asking microXML (whatever that turns out to be, if it turns out
to be anything) to add complexity to support DITA is almost
certainly going to be a non-starter.

> 
> Have any of the DITA TC been party to this discussion on the XML-dev
> list? I'm thinking that creating a liaison/intervention with the W3C
> might be useful for the DITA TC at this point, at least so that we can
> ensure some degree of common architecture when the DITA 2.0 effort
> begins. What do you think?

Anyone can join the public-html-xml@w3.org list.  (Warning, there is
a lot of boring email.)

I don't think DITA will have much leverage in determining HTML5.

I don't expect the relationship between HTML5 and DITA to be anything
much more than two markup languages on which--if we're lucky--we'll be
able to use XSLT processors to convert between the two.

paul

> --
> Don R. Day
> DITA and XML Consultant, 512 244-2868
> Co-Chair, OASIS DITA Technical Committee
> "Where is the wisdom we have lost in knowledge?
> Where is the knowledge we have lost in information?"
> --T.S. Eliot