The XML Processing Model Working Group is a Working Group of the W3C and follows the Working Group process described in section 6.2 Working Groups and Interest Groups of the Process Document.
The XML Specification defines an interchange format for structured information, but does not define how that format is to be used. Other specifications, such as XSLT, XML Schema, XInclude and XML Canonicalization define transformations that operate on and produce XML documents. The order in which these transformations are to be applied is not specified anywhere, even though applying them in different orders will in general yield different results.
The ability to specify the order, parameters, and expected results of such transformations in a standard way would be highly beneficial for interoperability between systems using different sets of processors for these transformations.
Two distinct perspectives can be identified: A scripting perspective, in which an independent XML document specifies the sequence of operations to be performed, and a self-describing perspective, in which a document is understood as signalling its own processing.
With respect to the scripting perspective, the group should define a language in which this can be expressed. This XML Processing language should allow users to specify the pipeline processing of XML documents, for example using multiple XSLT stylesheets as part of general data processing with XML.
With respect to the self-describing perspective, the existing
XML specifications which define transformations are manifested in
XML documents in many different ways, including the document
element (xenc:EncryptedData), internal elements
(xi:include), processing instructions
(<?xml-stylesheet…?>) and attributes
(xsi:schemaLocation). When encountering an XML
document including some of these features, a general-purpose
processor currently has no guidance as to which, if any, of the
corresponding transformations should actually be applied. So, for
example, a reference to part of a document may be ambiguous as to
whether the reference is applied before or after transformation.
The group will investigate and illuminate this issue.
The XML Processing Model Working Group will explore whether an XML-based system can determine whether the creator of any given XML document has indicated that operations on that document should be performed in a specific order for a particular result, and if so, how to apply those operations. This might take the form of reference to an external document in the XML Processing language, or some form of annotation regarding the self-describing aspects of the document. There is however no expectation that all documents will provide either form of indication, nor is there any expectation that such an XML Processing model as the WG produces will be the only way anyone will process XML documents.
The Working Group will take as input the position papers (Members-only) of the W3C Workshop on the XML Processing Model as well as the two related W3C Member Submissions (XML Pipeline Definition Language Version 1.0 and XML Pipeline Language (XPL) Version 1.0), and any other work in the area that the Working Group considers to be relevant.
As suggested above, the primary focus of the Working Group will be managing XML processing at the whole-document level, where each step in the processing of a document can be seen as a transformation as defined by a particular W3C Recommendation or comparable specification.
The goals of the XML Processing Model Working Group are to develop two Recommendation Track documents:
It is also expected that the Working Group will take into consideration potential consequences of processing XML documents represented by alternative (e.g. efficient XML interchange) serializations, or bundled together as compound documents or with other packaging methods.
Any specification of processing inevitably raises security issues: the Working Group will seek input on this from relevant sources of expertise inside and, if appropriate, outside the W3C.
The two main goals of the WG are clearly related, and will be pursued in parallel, but the situation with respect to the XML Processing Language is much further advanced, with a number of systems already in use. Accordingly a more aggressive timetable will be possible for that goal, which is expected to draw heavily on already deployed and widely-used systems. The questions surrounding the XML Processing Model are more complex, and may require more extended exploration and discussion before a conclusion is reached.
The XML Processing Model Specifications are expected to go through Last Call Working Draft and reach Candidate Recommendation or Recommendation during the lifetime of this Charter, but milestones after the first Public Working Drafts may need to be adjusted in the light of the reaction to the draft documents both from other Working Groups and from the public.
| Specification | FPWD | LC | CR | PR | Rec |
|---|---|---|---|---|---|
| Note: The group will document significant changes from this initial schedule on the group home page. | |||||
| XProc Language | N/A | N/A | N/A | April 2009 | June 2009 |
| XProc Model Requirements | June 2009 | N/A | N/A | N/A | N/A |
| XProc Model | September 2009 | November 2009 | December 2009 | N/A | N/A |
The expiration date of this charter was originally 31 January 2010, extended to 30 September 2010.
XSLT is a commonly used transformation mechanism, and XSL-FO a frequent (and the original intended) target of such transformations.
The XML Query Language can be used to select data from arbitrary collections of XQuery 1.0 and XPath 2.0 Data Model instances; the XQuery 1.0 and XPath 2.0 Data Model specification is maintained jointly by the XSL and XML Query Working Groups, and is a possible starting point for an XML Processing Model.
A W3C XML Schema document can provide default document content which could (for example by introducing XInclude directives or default attribute values) substantially affect the results of other transformations.
The XML Processing Model must interact in a well-defined manner with the PSVI.
The XML Processing Model Specification was originally within the scope of the XML Core Working Group, and was taken out for logistical reasons.
Interactions with core specifications such as xml:base and XInclude must be carefully reviewed.
Any suggested default XML processing model would need to be carefully checked for impact on the work of these groups.
The work on default behaviors for the XML Processing Model will need coordination with the TAG. The XML Processing Model should also be aligned with the Web Architecture, including the role of Internet media types.
Formal liaison between the XML Processing Model Working Group and other W3C Working Groups, as well as organizations outside of the W3C, shall normally be accomplished through the XML Coordination Group.
Participation is expected to consume up to one day per week of each participant’s time, although the time commitment for the chair and editors may be higher.
The level of participation expected for this Working Group suggests that representatives be experts in the development and use of XML-related specifications, as well as being W3C Members or Invited Experts.
The initial Chair of the XML Processing Model Working Group is Norman Walsh.
The initial W3C Team contacts for this Working Group are Henry S. Thompson and Liam Quin.
It is expected that this Working Group would consume about 15% of an FTE, including administrative logistics.
The Working Group will have weekly teleconferences. Face-to-face meetings will be held as needed.
The XML Processing Model Working Group shall communicate among its
participants using the public-xml-processing-model-wg
publicly-readable mailing list and with the other XML Working Groups
through the w3c-xml-plenary
mailing list. Both lists are archived.
Public comment on the work of the XML Processing Model Working Group may be submitted via the public-xml-processing-model-comments public mailing list.
The proceedings of this working group are publicly readable, subject to exceptions made by the chair. In particular, the archive of the mailing list used for technical discussion and the Group home page are publicly readable.
This Working Group operates under the W3C Patent Policy (5 February 2004 Version). To promote the widest adoption of Web standards, W3C seeks to issue Recommendations that can be implemented, according to this policy, on a Royalty-Free basis.
Information about how to join the Working Group is available on a separate “How-to-Join page”.