The World Wide Web Consortium

and Standards

C. M. Sperberg-McQueen,
Co-chair, W3C XML Schema Working Group
Co-chair, W3C XML Coordination Group
Domain Lead, W3C Architecture Domain

XML World 2001

San Francisco, California

27 November 2001


W3C is the birthplace of XML, but there's more to XML than one specification. The XML family of technologies is largely developed within the W3C XML Activity. Michael will provide an overview of the current and planned work of the W3C XML Activity, and address questions related to specific topics including XML Schema, XML Query, and XLink.

1. Overview

2. XML: the simple idea

At heart, XML is a simple idea:
  • Text is not just a sequence of characters.
  • Text is not just an arbitrary mixture of characters with commands.
  • Text has structure.
Information has structure.
Information formats should expose that structure.

3. XML: the simple idea

If information formats expose the structure of information, then:
  • Information can be richer, more variable, more subtle.
  • Software can be stupider simpler, cheaper, faster.
If we can't have intelligent software, let's at least have well-informed software.

4. XML: the simple idea

Nothing in XML is new.
It all (start-tags, end-tags, elements, attributes, tree structure, entities, document grammars) came straight out of SGML.
The original goals of the XML activity were:
  • make SGML usable on the Web
  • add structure to Web documents
  • be simple (25 pp.), clean (validity)

5. Growth of XML

XML has grown beyond its original goals:
  • make SGML and HyTime, and DSSSL usable on the Web
  • add structure to Web documents and data

6. Growth of XML

  • allow arbitrary data to be exposed on Web in XML
  • make XML usable in specific applications:
    • improve metalanguage for defining languages
    • query
    • authenticate with digital signature
    • canonicalize so signatures are more robust
    • use XML for program-to-program communication
    • use XML as the basis for Web Services

7. What's new at W3C?

8. Who is the W3C?

The World Wide Web Consortium is a member-supported organization which creates Web standards.
Our mission:
to lead the Web to its full potential.

9. W3C goals and operating principles

10. Organization of work in W3C

Domains and Activities:
  • Architecture Domain
  • Technology and Society Domain
  • Document Formats Domain
  • Interaction Domain
  • Web Accessibility Initiative

11. Architecture Domain

12. Technology and Society Domain

13. Document Formats Domain

14. User Interface Domain

15. Web Accessibility Initiative

16. The W3C Rec Track

17. The W3C Rec Track

18. XML Core WG

Chairs: Paul Grosso, Arnaud Le Hors
  • XML 1.0 Second Edition, ed. Eve Maler (XML 1.02e) (Recommendation)
  • XML updates for Unicode 3.x (XML Blueberry) (Requirements out, draft out)
  • XML Information Set (Rec)
  • XML Inclusions (XInclude) (Last Call completed)
  • Errata for existing specs
  • XML Namespaces 1.1
  • Classification of XML Processors

19. XML Linking

Chair: Henry Thompson
Two Recommendations:
  • XLink (standoff links, multi-ended links, better role labeling)
  • XML Base (relative URI support)
One Candidate Recommendation:
  • XPointer (addressing into XML)

20. XML Schema

Chairs: Dave Hollander, Michael Sperberg-McQueen
XML Schema, parts 0, 1, and 2 a W3C Recommendation.
  • entities/special characters work
  • formal description of XML Schema
  • work on test suite

21. XML Schema Highlights

Like DTDs, but:
  • in XML document syntax
  • coherent account of schema validation across multiple namespaces
  • modular construction of vocabularies
  • full validation, partial validation (black boxes, white boxes)
  • simple datatypes
  • inheritance

22. XML Query

Chair: Paul Cotton
Toward an XML Query language for all XML data:
  • XQuery 1.0 and XPath 2.0 Data Model
  • XQuery 1.0 and XPath 2.0 Functions and Operators
  • XML Query Use Cases
  • XQuery 1.0 Formal Semantics
  • XQuery 1.0: An XML Query Language
  • XML Syntax for XQuery 1.0 (XQueryX)

23. XML Query Highlights

  • element selection (XPath++)
  • strong static typing
  • dynamic type checking
  • keyword syntax and XML syntax
  • generate arbitrary XML results

24. DOM

Chair: Philippe Le Hégaret
Recommendations: Document Object Model (DOM) Level 2
  • Core
  • Views
  • Events
  • Style
  • Traversal and Range

25. DOM, continued

Working drafts:
  • Level 2 HTML DOM
  • Level 3 Core
  • Level 3 Views and Formatting
  • Level 3 Abstract Schemas and Load and Save
  • Level 3 Events

26. DOM Highlights

DOM provides
  • an intelligent API for document manipulation
  • browser-independence* of scripting code
  • crucial tool for power users and system integrators (this means you)
  • support for editors ...

27. XML Protocols

Chair: David Fallside
Goal: create a simple foundation for program to program communication using XML.
Recent documents:
  • Requirements document
  • XML Protocol Abstract Model
  • SOAP 1.2 Working Drafts
    • Part 1: Messaging Framework
    • Part 2: Adjuncts

28. XSL

Chair: Sharon Adler
XSLT a W3C Recommendation since November 1999
XSL (aka XSL FO) now also a Recommendation
Work underway for XPath 2, XSLT 2 (extensibility, XML Schema type support)

29. Semantic Web

Goals: define
  • enabling standards (RDF/XML)
  • and support technologies (software)
to facilitate the creation of machine readable metadata on the Web for more effective
  • discovery,
  • automation, and
  • integration
of networked information.
  • Semantic Web Activity headed by Eric Miller
  • Advanced Development headed by Ralph Swick

30. RDF Core WG

Chairs: Brian McBride, Dan Brickley
Resource Description Framework: a low-level model for semantic structures.
Basic model: labeled directed graph.
Current work:
  • RDF Model and Syntax (Recommendation)
  • RDF Schema (Candidate Rec)

31. RDF -- the basics

To describe a resource, you need a vocabulary in which to describe it.
The problem is that in a world-wide Web with universal access, we cannot rationally assume that we know in advance everything anyone might wish to say.
No closed-world assumptions!
So -- define a vocabulary ... but also specify a more fundamental extensible semantic system in which it is defined.

32. RDF -- the minimum

To say something about something is to describe some property/ies of your subject. You need at least:
  • a way to name the subject (URI)
  • a way to identify the property you are giving (label, property name)
  • a way to give the value of a property (literal, or URI)

33. Intellectual property

Does anyone like software patents?
Can anyone do anything about them?
And what do we do in the meantime?

34. W3C Patent Policy

The key components of the draft W3C patent policy:
  • disclosure (no* submarine patents)
  • commitment in advance to terms
  • essential technology
  • possible opt-out
  • viral effect
Other components:
  • royalty-free vs. `reasonable and non-discriminatory' terms

35. Some myths and facts

  • "W3C is loosening its policy."
  • No, W3C is tightening its policy.

36. Myth/Fact No. 2

  • "W3C used to grant royalty-free licences, and now you're planning to extract royalties."
  • No, W3C has never had licensing terms attached to implementation of W3C Recs.
    W3C documents and open-source software continue to be available on the same terms as before.

37. Myth/Fact No. 3

  • "Everyone knows the Internet is fundamentally royalty-free. Why change now?"
  • Many people do believe so, but not everyone agrees.
Without disclosure, the true state of technology is not RF, not RAND, but FUD.

38. Information