SGML/XML 98 Paris

[Peter Murray-Rust]

A fairy godmother made it possible for me to attend the latter half of the
Paris meeting and this is a brief report for other XML-DEVers who weren't
so fortunate. I only get about one chance a year to meet real-life XML
people - a year ago in Barcelona, and occasionally some visiting London.
This report is not comprehensive - I missed anything before Wed a.m.

The conf was wound up by Tim Bray who asked 'What is a document?' This is
important as Tim  now categorises himself - and most of us - as indulging
in 'document computing'. Tim threw up slides of objects that may or may not
be documents - music, a book with no words, a signed baseball, etc. The
essential message was that in working with XML we are working with the
material of human culture - in many forms - and that we can both enjoy it
and have a responsibility. The responsibility is the trust that is put in
us to manage information for the benefit of humankind.

There is no doubt that 'XML has arrived' and is here to stay. Unfortunately
I missed Jon Bosak's plenary (and I missed Jon). His plenary was highly
praised and - I believe- again stressed the responsibility that we have to
make XML work by always bearing in mind that we are part of a greater
community.

In a later session Jon outlined the next stages of the XML process. I date
not attempt my transcription here as Jon chooses his phrases very carefully
and with great meaning. He explained how the namespace proposal had come
from the requirement of many W3-members and the narrow path that the WG had
to tread. It had been essential to be simple at the outset to avoid
committing to something that later might not be found to be workable.
He stressed the need for the different W3 groups to work together (e.g. he
did not feel a separate - and therefore potentially isolated - group for
XML-data would be a good idea.)

For many people the joint plenaries on Wednesday (Jean Paoli, Microsoft,
and Bernard Feinmann, Netscape) were the key piece of take-home evangelism.
Both represented their companies as committed wholeheartedly to XML. JP
summarised some of the themes:
	- 'Data should be free' (Charles Goldfarb)
	- 'Give the user [power] (Jon Bosak) [I forget the exact words].
and emphasized the critical power of text-lovers and free speech on the
WWW. He summarised 
the many initiatives of vendors and others as showing that XML had really
arrived (including CML :-). He talked about XML and HTML interoperating
(see a URL at w3.org/TR/NOTE-xh-19980511.html if I got it right) and how
<SCRIPT> and <XML> had been proposed as HTML tags. The MS  dream was:
	- let users craft their documents
	- let searches find anything
	- store everything in text
	- have readable, accessible formats
He finished with a live demo of prototypes of the Office suite of tools
based on XML. This included Word, Powerpoint and Excel (** all based on
native XML **). This included a multi-dimensional numeric table from Excel
with its complete representation in XML. This - the first preview of XML in
Office - heralded XML in MS products 'this year'.

BF outlined the NS philosophy, including that:
	- source code should be free
Within 48 hours of the public Mozilla release, James Clark's XML parser had
been integrated and there had been 25000 downloads of the source kit on the
first day. Linux was the fastest growing OS and was the reference platform
for Mozilla. JavaScript 1.3 was now shipped fully. The source is free and
components can be depackaged - i.e. you can extract the bits you need for
any task. NS was committed fully to DOM, RDF, XML. RDF was being developed
for site maps and was a fundamental support tool client-side. Bookmarks,
History, Filestore etc were all now built on RDF, and B gave an interactive
demo of NS using these facilities. He showed how XML-based sites could be
cut-and-pasted at any desired granularity. 

There was an editorial/advisory board including: Jon Bosak, Tim Bray, James
Clark, Dan Connolly, and several other luminaries (sorry - didn't capture
them). NS had 'complete and total commitment' to XML and all things XML.

The main session I attended was the HyTime one. Charles Goldfarb outlined
the history, including need to develop HyTime as an abstract approach to
support music markup. As a result of that - and Eliot Kimber's presentation
- I am intellectually convinced that HyTime and AFDRs are a right and clean
way to tackle XML problems. Eliot wasn't happy about namespaces - HyTime
does it all, simply and cleanly, and namespaces don't. I cannot comment
with authority. He displayed his VB HyTime engine PHyLis (caps not
guaranteed) and showed how AFs could be handled. Later I had a long
masterclass sitting on the hotel floor and have made steps along the path
to enlightenment. My position is that if there is an implementation of AFs
in java, with a small number of examples, I will take the time to explore
it. There appears to be some prospect of this.

Masatomo Goto presented a very impressive demonstration of his Xlink engine
(announced last week on XML-DEV). He has used HyTime as the fundamental
infrastructure and implemented Xlink on top. Eliot described it publicly as
'very cool' and I wholeheartedly agree. If it could be made available to us
for testing Xlink it would be a great benefit.

The conference was very well attended and emphasised the commercial
realisation of XML as well as the theory. In a straw vote, delegates felt
that there was room for 'more technical material' in sessions next year. I
left with a feeling of invigoration, sadness that I couldn't spend more
time talking to people and an awareness of the need to keep the community
spirit as long as possible. 

	P.




Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)