XLink and XPointer prepare to move forward
12:21, 12 Dec 2000 UTC | Eric van der Vlist

XLink and XPointer seem ready to move forward after Daniel Veillard released the decisions of the XML Linking Working Group regarding the comments on the XLink and XPointer CRs.

The comments cover a wide range of both editorial and technical issues that are often at the border between several specifications.

One such clarification about the scope of XPointer applies to XML documents on which a XPath data model can be constructed. This scope covers multi-rooted documents that can be used as external parsed entities, but on which an infoset cannot be defined and:

"This resulted in a question asked to the XML Core WG to extend the Infoset capabilities to external parsed entities. This was accepted and is been added into XML Infoset."

On the other hand, "XML 1.0 well formed documents that do not have infosets, such as those that contain names with colons that are not namespace-conformant" may still be considered as having a mime type XML but XPointer will not apply to them and therefore there will be no possibility to define fragments on these documents.

The concerns about the ability of XPointer to define ranges which have a finer granularity than that which can be defined by styling technologies (including XSLT and CSS) have also been rejected.

Other suggestions from the XSL/XML Linking task force to add attributes to describe the rendering of XLink have been postponed, while XLink will be amended to state that the rendering of several ambiguous situations will be application-dependent.

The issue of using relative URIs for semantic attributes has given birth to "a minority opinion," but the decision has been to forbid them.

The patent held by Sun over a technique used by the XPointer specification was another issue that needed to be resolved before the specification could move forward.

The terms and conditions proposed by Sun would probably make the future recommendation the first shrink-wrap W3C document whose simple download would have important legal implications, since you would be protected from any legal action from Sun under their US Patent, No. 5,659,729, and in return would agree:

not to sue Sun or any third party for infringement of any patents which now or in the future are (a) owned by You; (b) under which you have authority to sue; or (c) to which you have authority to grant rights; for the making, using, selling, offering for sale, otherwise disposing of or importing a fully compliant implementation of the [current or future] XPointer Specification or a subset thereof.

Related stories:

xmlhack: developer news from the XML community

Front page | Search | Find XML jobs

Related categories