Keywords: DOM, SAX, XML, XPath
Biography
Gareth has worked with XML for the past five years, acquiring a deep knowledge of XML related technologies, including XML Schema, XPath and DOM. Gareth has contributed to a number of major Open Source projects. He sits on the Apache Project Management Committee for XML Products and has commit access to the Apache Xerces project. He also manages the Pathan project, DecisionSoft's Open Source XPath 1 and XPath 2 implementations. His experience led him to found Parthenon Computing, a company specializing in providing XML and Open Source expertise. In addition Gareth participates in XML standardization working groups and has contributed as a technical reviewer to several key XML publications.
Biography
Alberto Massari started working on XML technologies in late 1998, working for Object Design in the team that built eXcelon, the first XML Database. He now works for the Stylus Studio team managing the XQuery and XSLT processors and the DTD/XMLSchema editors bundled inside this IDE for XML development. He has contributed to a number of major Open Source projects, and has commit access on the Apache Xerces project.
Biography
Lucian works for DecisionSoft, well known for their cutting-edge XML expertise. Since joining he has taken responsibility for designing and coding an implementation of the DOM Level 3 Validation interfaces as part of a larger XML Schema management tool, and is currently Technical Architect on their XBRL Toolkit product.
Biography
Neil Graham has worked on XML parsing technologies since 2000. He has contributed code to both the Java and C++ versions of the Apache Xerces Project, and serves on the Project Management Committee of the Apache XML project. He also serves as one of IBM's representatives on the JSR 206 (JAXP 1.3) Expert Group. He has been Manager of XML Parser Development at IBM since mid 2004.
It's been 7 years and three "levels" since the first W3C DOM activity. XML and the way it is used has changed vastly over that time. DOM itself has
moved from an API to access and manipulate an in-memory tree with no
concept of namespaces, to an end to end XML technology, where parsing,
modification of the tree (with the ability to check for validity with a
schema as you go) and serialization are all specified.
At the other end of the spectrum, SAX has scarcely changed since moving
into the world of namespaces in 2000. A few extensions have been added and
very recently formalized, after nearly two years of a seemingly endless
beta, but other aspects of the evolution of XML technology have appeared to
have simply passed SAX by.
Now that the W3C DOM activity is coming to a formal close and SAX seems to have stopped for all intent and purposes, this Town Hall aims to look at the benefits and problems that these APIs have provided us with. In
particular, we will look at whether DOM really is a cross platform and language API, if DOM has got too big and covers too much, or conversely, not enough, and whether continued development on this, or a similar standard would be useful to the community.
We will also focus on if there is a future for a push-based event API
in modern XML parsing, or whether there needs to be more focus on
standardizing and extending pull-based API's, as well as whether
the multiple data models specified by the myriad of XML standards
are required, or whether more consideration should be paid to
interoperability to prevent the half-hacked solutions that companies
have created to build wrappers for the different XML views. The latest round of DOM specs delve into the XML Schema and XPath worlds, and this will only exacerbate this situation.
Since this was a Town Hall meeting, it was not possible to prepare a paper for the proceedings.
XHTML rendition made possible by SchemaSoft's Document Interpreter™ technology.