|
| XML
Charter of the XML Schema Working Group
April 2002
C. M. Sperberg-McQueen
and Dave Hollander, chairs
C. M. Sperberg-McQueen, staff contact
$Revision: 3.6 $ of $Date: 2002/08/21 16:33:54 $ by $Author: liam $
The XML Schema Working Group is a
Working Group of the W3C and follows the working group process
described in the W3C Process
Document. Except as outlined elsewhere in this charter, the
Working Group follows the Common
Procedures for XML Working Groups.
This is the
charter for this Working Group.
It extends and supersedes the Working Group's
previous charter.
The XML 1.0 specification defines the concepts of
well-formedness and validity; the former
characteristic is very simple to check, while the latter requires more
work to check, and allows the user to define more powerful constraints
on document structure. XML validity requires that a document follow
the constraints expressed in its document type definition,
which provides the rough equivalent of a context-free grammar for a
document type.
In some contexts, there is need for constraints tighter than,
looser than, or simply different from those which can be expressed
using document type definitions as defined in XML 1.0. There is also a
widespread desire for an XML-based syntax for expressing constraints
on document types, in order to allow tools for XML documents to be
used on the constraint specifications.
Using a term borrowed from database terminology, the expression of
constraints on a document type is called (in this document, at least)
a schema. (Note that the term schema has two
plurals, schemas and schemata; we use the
former.)
The XML Schema working group will maintain and revise the XML
Schema specification developed beginning in 1998 and published as a
W3C Recommendation on 2 May 2001. The XML Schema specification
addresses several topics relevant to users of XML:
-
primitive data typing: XML Schema defines
a type system for integers, dates, and the like, based on experience
with SQL, Java primitives, etc.
- structural constraints: XML
Schema provides a mechanism roughly analogous to DTDs for constraining
document structure (order, occurrence of elements, attributes). XML
Schema differs from DTDs in expressing these constraints in the form
of an XML document, rather than in a special-purpose ad hoc
notation. Unlike DTDs, which contain declarations both for the logical
structure of elements and attributes, and for the physical structure
of entities, XML Schema provides declarations only for constraining
elements and attributes; entity declarations continue to use DTD
syntax. In addition to DTD-like functionality, XML Schema:
- is integrated with namespaces;
- allows incomplete constraints on the content of an element type
in the form of wildcards, so-called ‘black box’ validation, and
so-called ‘white box’ validation;
- is integrated with primitive data types;
- defines inheritance for datatypes;
- provides various hooks for safe adaptation and reuse of
schemas.
- conformance: XML Schema defines the relation of schemas to
XML document instances and specifies the obligations of schema-aware
processors.
The goals of the work group in the period described by this
charter are five-fold:
- to provide continuing maintenance and support for
the XML Schema 1.0 specification, in the form of
an Errata document containing corrections of identified errors
and clarifications of points which experience shows are
obscure to careful readers,
and to complete the document XML Schema: Formal Description,
which provides a formal treatment of schema-based type assignment
and validation in terms of a Gentzen calculus
- to prepare a version 1.1 of the XML Schema Recommendation,
which corrects known errors and may make modest improvements to
the language, while retaining compatibility with version 1.0
- to clarify various related problem areas, and do technical work
leading possibly to Recommendations, possibly to Notes, and possibly
to the determination that no publication is necessary; these
related problem areas include:
- the formalization of XML Schema in a manner suitable for use
by XML Query and others
- the definition of a free-standing specification describing how
to name or refer to arbitrary components in an XML Schema;
in existing discussions these are sometimes referred to as
normalized universal names or NUNs
- the application-programming interface (API)
for the post-schema-validation (PSV) information set; this
work is undertaken in response to comments on the Last-Call
draft of XML Schema, and will be performed in consultation with the DOM WG.
It is expected to be an extension of their generic interface to
content models and related information
- useful ways of handling the functionality provided in XML 1.0
DTDs by entities (with special reference to the use of entities to
provide names for special characters)
- the systematic development of test cases for XML Schema
- provision of a sample library of common types, for illustration
of the construction of type libraries using XML Schema and, as
appropriate, for reference from schemas for W3C specifications
- definition of operators on XML-Schema types, in collaboration
with XML Query and XSL. (This work is currently driven by the
other two WGs; XML Schema participates in order to provide
expertise and help avert problems.)
- the study of various proposals for new architectural approaches
to schema definition and schema validation, with an eye to their
potential inclusion in future versions of XML Schema; these proposals
include but are not limited to:
- an even sharper separation between the level of abstract schema
components and the XML transfer-syntax level than is present in
XML Schema 1.0, possibly including separation of the abstract and
concrete levels into separate specifications
- possible use of multiple transfer syntaxes which 'compile' to
the same set of abstract components
- multiple inheritance
- abstract simple types
- various proposals postponed from the work on XML Schema 1.0
- schema annotation and the specification of mappings which
allow software to translate data from XML document instances which
conform to a given schema into other formalisms or data models, such
as application-specific data structures, generic data structures (e.g.
relational tables), the entity-relationship model, and RDF. This
work follows on from the discussions reported in the Cambridge
Communiqué of 1999, with the aim of showing how the
goals described there can be achieved.
- interoperability with other schema languages such as RELAX-NG
and Schematron
- to cooperate with other W3C working groups and assist them in
exploiting XML Schema for their work and adding schema-awareness
to their specifications where appropriate; this will include
work on the problem of schema-annotation and the problem of mapping
from the XML vocabularies defined by a schema into target data models
The XML Schema WG has repeatedly experienced tension between
the desire to explore some proposal for a relatively major change in
the design of the XML Schema language, and the desire to have a
non-proprietary specification available for use as soon as possible.
In order to allow adequate time for consideration of possible
major changes in approach,
the WG may (time permitting) do technical work on proposals intended
for future versions of XML Schema (such as those listed above)
in parallel with its work on XML Schema 1.1.
The XML Schema WG has three main deliverables:
- A formal description of XML Schema.
- Errata list for XML Schema version 1.0
- XML Schema version 1.1; this will be a maintenance release
of the spec. It may differ from version 1.0 in incorporating:
- fixes for errors in version 1.0
- clearer (and if possible briefer) exposition
- improvements to the XML transfer syntax, to improve
usability based on reports from users and implementors
- minor improvements to the specification of the
abstract component level (if changes can be made without
inconvenience to users of existing schemas or developers of existing
implementations)
In addition, there are several other deliverables:
- collections of XML Schema use cases and test cases
- list of XML Schema test suites and report on coverage
- possible documents stemming from the various ‘related problem areas’
mentioned above:
- application-programming interface (API)
for the post-schema-validation (PSV) information set
- entities (especially for special characters)
- test cases for XML Schema
- sample library of common types
- work on schema annotation
Expected schedule:
For details of dates and locations see the WG home page.
A schedule of expected milestones in the development of
requirements documents, use cases, working drafts, etc.,
will be maintained and
linked to from the
XML Schema WG home page.
Note that public working drafts will be made available at least once every
three months, in accordance with the
W3C Process.
The expected duration of the working group under this charter is
through September 2003.
XML has become a strategic technology in W3C and elsewhere. The deliverable
of the XML Schema WG must satisfy the dependencies from other W3C Working
Groups. Some dependencies
to and from other W3C Working Groups will require close cooperation during
the development process; the requirements posed for the Schema work by these
WGs may change during the development process, which means the interdependency
of the Schema work with these WGs must be managed actively:
-
XML Query WG
-
The XML Query WG expects to use schema-related information in the
processing of queries; we will discuss points of mutual interest in an
effort to ensure that XML Schema provides the information Query
needs. In particular, the XML Query WG has expressed particular
interest in the work on a formal description of XML Schema, and
in the definition of the post-schema-validation infoset (PSVI).
The XML Schema WG will collaborate with XML Query and XSL in
work on operators for XML Schema datatypes and (as appropriate) on
the integration of support for the XML Schema type inventory
into future versions of XPath.
-
XSL WG
-
In collaboration with the XML Query WG, the XSL WG expects to
make future versions of XPath and XSLT support schema-related
information, in particular types.
The XML Schema WG will collaborate with XML Query and XSL in
work on operators for XML Schema datatypes and (as appropriate) on
the integration of support for the XML Schema type inventory
into future versions of XPath.
-
DOM WG
-
The DOM WG may wish to provide access to details of the data types and
structural relationships (e.g. inheritance) among element and
attribute types for schema-validated documents. We expect to work
together with the DOM both on their definition of a generic interface
to DTD- and schema-related information, and on our possible definition
of an API to XML-Schema information and the PSVI.
- HTML WG
-
It is a goal of the XML Schema work to ensure that the structural
schema language we define will suffice for the requirements of current
and future versions of the HTML specification.
-
XML Core WG
-
The XML Schema work will define methods for specifying constraints on XML
documents; those constraints should apply to objects and properties identified
as significant by the XML Information Set specification.
The XML structural schema specification
will use namespaces to combine
fragmentary structural specifications; it will also define schema-validation
for documents using namespaces. (In short: the XML Schema work will be
name-space-aware.)
-
XForms WG
-
The XForms WG is responsible for a datatype-aware replacement for HTML
forms; we will discuss points of mutual interest in order to ensure
the alignment of the two specs and the utility of XML Schema for the
work of XForms.
-
Internationalization WG
-
Since XML Schema is expected to be an important tool for the definition
of markup vocabularies, it may have follow-on effects on a larger number
of people than create XML Schema documents directly.
The XML Schema WG and the Internationalization WG will work together
to clarify and resolve internationalization and localization issues
in XML Schema, and will jointly ensure that it satisfies W3C goals for
international access to the Web.
-
RDF Core WG
-
The RDF Core WG is preparing a new version of RDF Schema, which will
use and build upon XML Schema datatypes. The RDF Core WG is also
chartered to "provide an account of the relationship between RDF
and the XML family of technologies (particularly Schemas ...)".
-
Web Services Description and XML Protocol WGs
-
The XML Protocol WG and the Web Services Description WG in the Web
Services Activity use XML Schema extensively. The XML Schema WG will
ensure that
the schema language is sufficient to meet the requirements of deliverables
from these Web Services WGs. The XML Schema WG will also discuss points of
mutual interest and will review deliverables from the Web Servicesb
WGs and provide them with feedback.
Some other work groups should be mentioned here; there are no requirements
for co-development of features with these WGs, but there are points of contact
between their work and that of this WG, and thus logical dependency between
their deliverables and those of this WG. Requirements from these WGs are
expected to be well suited for communication via documents.
-
WAI Protocols & Formats WG
-
Reuse of common constructs greatly facilitates accessibility; the WAI PF
WG will review work on structural schemas to be sure cost/benefit design
decisions are informed of the benefits of accessibility.
-
XML Linking
-
The XML Linking and Schema WGs will consult with each other on the question
of declaring element types and attributes as links or addresses.
Formal liaison between the XML Schema Working Group and other W3C
working groups, including the other XML working groups and the WAI
(Web Accessibility Initiative) group, as well as organizations outside
of the W3C, shall normally be accomplished by the exchange of
documents (requirements, reviews, etc.) transmitted through the XML
Coordination Group.
When approved by the XML Coordination Group, liaison with other W3C
Working Groups can be accomplished through direct exchange of
documents or by joint task forces. It is
expected that this be required for liaison with at least the XML
Query, XSL, and Internationalization Working Groups.
WG members
Participation is expected to
consume one to two days per week of each WG member's time, though the time
commitment for the chair and editors may be higher.
The list of members
and details about joining are on the group home page.
The level of participation expected for this WG suggests that representatives
be experts in the development or exploitation of schemas for XML-based
markup languages. In addition to the time commitment noted above,
each member of the WG will be expected to show evidence to the chairs,
each six months, of active work in the general topic area. Such active
cultivation may take the form of:
- development of at least one non-trivial schema, using XML Schema 1.0
or a draft under development
- development of at least a small piece of software which processes
XML Schema documents, or participation in the development of larger
software packages which support XML Schema.
Chair
The initial co-chairs of the XML Schema Working Group are
C. M. Sperberg-McQueen, W3C, and
Dave Hollander, Contivo.
Later chairs, if any, will be named by the Director.
The term chair in this document refers to either
co-chair.
The XML Schema WG (principals and alternates) shall communicate among its
members using the
w3c-xml-schema-wg
mailing list and with the members of other XML
WGs through the
w3c-xml-plenary@w3.org
mailing list. The WG mailing list is archived; the
archive is visible
to the public.
It shall communicate with the general public only through its
Chair.
Members must treat all Member-only documents as confidential within W3C and use reasonable efforts to maintain this confidentiality and not to release this information to the general public or press.
excerpt from section
1.1.3 -
Member Confidentiality of the W3C
Process
The proceedings of this working group are member-confidential,
subject to exceptions made by the chair.
In particular, the XML Schema WG mailing list archive and the Group
home page are accessible by members only. (N.B. published documents
may of course be discussed publicly.)
W3C, and all W3C Working Groups, are accountable to the Web
community as a whole for the quality of W3C technical work. In
support of this public accountability, this working group will
periodically (e.g. monthly) make public a summary of all
technical decisions made since the last public summary, and
the rationales for these decisions.
Face-to-face meetings will be held as needed; it is expected that
such meetings will be held about every two to three months.
To be successful, we expect the XML Schema WG to have 15 or more
active principal members for its duration.
The W3C shall be represented by one principal and (optionally) one alternate
WG member. The W3C staff resource commitment is no more than the minimum
level of participation required of WG members in good standing by the W3C
Process Document.
The W3C staff contact for this WG is, initially,
C. M. Sperberg-McQueen. It is expected that this WG would consume
about 60% of the contact's time including administrative
logistics.
Communications resources for press and media relations and speaking
appearances or meeting planning resources are amortized across the
working groups in the XML Activity, and concentrated in the XML
Coordination Group.