XML Conference Report: XML Seminar

The conference started with a half-day seminar from Tim Bray, 'The XML Standard: A Top to Bottom Tutorial'. Tim covered the design goals of the W3C Working Group and Editorial Review Board (ERB) in developing the specification which is available at www.textuality.com/sgml-erb/WD-xml.html. He then led the attendees through the XML syntax.

For most SGMLers, the biggest mental hurdle is the optionality of a DTD (see "XML: Questions and Answers," Newsletter Vol 3 Issue 1, January 1997). While it is possible to use one, and certainly encouraged in many business situations, it is possible to have XML documents that do not have a DTD. These XML documents must, however, be 'well-formed'. The XML covers the definition of well-formedness in section 2.2:

'A textual object is said to be a well-formed XML document if, first, it matches the production labelled XML Document, and if for each entity reference which appears in the document, the entity name has been declared in the document type definition.

Matching the "XML Document" production implies that:

It contains one or more elements.
There is exactly one element, called the root, or document element, for which neither the start-tag nor the end-tag is in the content of any other element. For all other elements, if the start-tag is in the content of another element, the end-tag is in the content of the same element. More simply stated, the elements, delineated by start- and end-tags, nest within each other.

As a consequence of this, for each non-root element C, there is one other element P such that C is in the content of P, but is not in the content of any other element that is in the content of P. Then P is referred to as the parent of C, and C as a child of P.'

The document XML - Questions & Answers published last quarter puts this a bit more understandably, if less precisely, with the explanation that '...a well-formed document is one that can unambiguously be parsed to create a logical tree in memory. . . to have a well-formed document, it's only necessary to start with a single root element and nest the child elements sensibly.'

Given the optionality of a DTD, two schools of thought emerge:

you need a DTD to create documents, but well-formedness is sufficient for delivery (i.e., control is needed during authoring, but searching and delivery mechanisms only require access to the instantiated tree, not all possibilities of that tree)
you don't even need a DTD for authoring (i.e., let an author create well-formed structures; if a DTD is required for some other process, let it be imputed from the well-formed instance).

Each of these could be valid in different business circumstances. The distinguishing factor is that XML provides the flexibility to use or not use a DTD as appropriate, rather than mandating one method over the other. For example, an organisation may require a DTD for the development of their maintenance procedures, but may allow well-formed documents for informal postings to their intranet.

In creating the XML profile of SGML, the developers have tried to remove 'all the SGML features that make SGML client software difficult to implement'. Whilst some users may miss the ability to use SHORTREF (such as for mathematical notation), the absence of CONCUR, RANK, DATATAG, etc. may fail to trigger a protracted mourning process, and the exclusion of exceptions (although extremely helpful in some areas) will probably result in far fewer SGML users having to troubleshoot unexpected results. In all, the general feeling (at both the announcement at SGML '96 and in this tutorial) was that the creators of this profile of SGML have chosen wisely in their constraints, especially considering that XML is specifically aimed at web delivery.

Further development will include work on a more robust linking model than that supported in HTML and a model for display processing. XML and Hypertext Links in XML (http://www.textuality.com/sgml-erb/WD-xml-link.html) will be presented at the W3C conference in April. The display processing work is expected to be drafted by June of 1997, and the Working Group's mandate expires by July 1997. Yes, everything is happening in web-time.

[Next]