Arbortext comments on XPointer requirements

Arbortext believes that, though we all want to provide important linking
and querying functionality to XML users, it is important to allocate the
various functionality properly among XPointer, XLink, and what the upcoming
XML Query work will develop.  To that end, we have written up a
"minimalist" XPointer proposal that introduces a set of additional design
principles that we feel belong in the XPointer requirements document.

We can support XSL/XPointer syntax and semantics unification, but believe
that this does not preclude a minimal approach for XPointer, which, as a
URL fragment-ID language for documents with XML MIME types, has a role
quite dissimilar from XSL.

The full proposal can be found at:

  http://www.w3.org/XML/Group/1999/03/ati-xptr-proposal

The following is excerpted and adapted from that proposal.

			*		*		*

Our motivations for developing the proposal are as follows:

- In our experience and the experience of several other active XPointer and
TEI extended pointer notation users (Tim Bray, David Megginson, Panorama
annotation software, etc.), actual pointer creation appears to use a small
fraction of available functionality.  The practical benefits of a
minimalist, easy-to-understand language outweigh the theoretical benefits
of a comprehensive, intensionalist language, particularly when it is
acknowledged that the minimalist subset still provides major benefits over
current HTML addressing methods.

- We believe that the query-related requirements that have been expressed
should be attended to in a query language, not an addressing language, and
anticipate that they will in fact be attended to by the proposed Query
Working Group.

- Similarly, we believe that linking-related requirements that have been
expressed, for example, aggregate links, should be attended to in the XLink
specification.

- Finally, since XPointer is the universal fragment-identifier language for
addressing into any Web resource with a MIME type of */xml, we believe that
its functionality is best kept to a minimum.

The specific design principles we propose are as follows; our sense is
that, with further discussion, these principles or similar might be adopted
by the WG, clarifying the design direction of XPointer considerably:

- The goal of XPointer is to address into documents, which means that the
primary method of addressing should be to start at the root of the document
tree.

- Most intensional addressing (that is, "addressing" of content whose
current location is likely to be unknown at traversal time) should be
handled by a query language. It is unreasonable for XPointer to optimize
for document volatility.

- The target of addressing should be single objects. It is reasonable to
address ranges that may correspond to user selections, but this
functionality should have single-node addressing at its base.

	Eve Maler, Principal, XML Linking WG 
	Paul Grosso, Alternate, XML Linking WG; Arbortext W3C AC rep

Received on Tuesday, 23 March 1999 11:53:00 UTC