ISO/IEC JTC 1/SC34 N409

ISO/IEC JTC 1/SC34

Information Technology --
Document Description and Processing Languages

Norwegian National Body comments on contributions to the SC34 meeting in May 2003

These are the official comments of the Norwegian National Body on documents submitted prior to the SC34 meeting in London in May 2003. We regret that we haven't been able to comment on every document. The position of the Norwegian National Body is further reflected in postings made by Lars Marius Garshol and Steve Pepper to the sc34wg3 mailing list. In addition we have contributed directly to the documents relating to TMQL and TMCL.

1. General comments

In the light of the current lack of consensus in WG3 the Norwegian NB urges greater respect for ISO process and draws particular attention to the following:

• Since the adoption of ISO 13250 in January 2000, ownership of the "Topic Maps" brand is vested in ISO under the responsibility of SC34. Only standards and documents approved through the ISO process can lay claim to being authoritative in terms of defining and explaining the concepts of the Topic Maps paradigm.
• Editors of SC34 standards and parts of standards are appointed by the committee and only by the committee.
• The names of standards (and parts of standards) are decided by the committee. Editors should not change these names without the approval of the committee.

2. The structure of ISO 13250

The London meeting should revisit the Roadmap defined and agreed upon in N278 and N323 in the light of the approval of N358 and the National Body comments in N388. The structure of the new multipart Topic Maps standard should be clarified and a timetable drawn up for further progress.

The Norwegian NB proposes the following structure for 13250:

ISO/IEC 13250-1: Topic Maps -- Overview and Basic Concepts
ISO/IEC 13250-2: Topic Maps Data Model
ISO/IEC 13250-3: XML Topic Maps Syntax
ISO/IEC 13250-4: HyTime Topic Maps Syntax
ISO/IEC 13250-5: Topic Maps -- Canonicalization

The editorial assignments for 13250 should be reworked to match the split of the standard into several parts. The Norwegian NB nominates Mr. Steve Pepper as editor of Part 1.

3. Comments on individual documents

3.1 N396 (SAM)

The Norwegian NB believes that the Standard Application Model is now at a stage where it can be progressed to CD ballot. In order to enable further progress on ISO 18048 (TMQL) and ISO 19756 (TMCL) it is important that this be done immediately after the London meeting. The Editors should be instructed to resolve all outstanding issues in N396 on the basis of discussions in London, ensure that the form of the document meets ISO requirements, and submit it for CD ballot.

The Norwegian NB proposes that the name of this part of the revised ISO 13250 be changed to "Topic Maps Data Model".

3.2 N398 (XTM)

The Norwegian NB believes that the XTM 1.1 Syntax Specification is similarly at a stage where it can be progressed to CD ballot. The Editors should be instructed to resolve all outstanding issues in N398 on the basis of discussions in London, ensure that the form of the document meets ISO requirements, and submit it for CD ballot.

3.3 N391 (HyTM)

The Norwegian NB regrets that it is not qualified to review and comment on the HyTime-related aspects of the HyTM Syntax Specification and urges that HyTime experts be requested to do a full review. Since HyTM needs to conform to the HyTime standard, processing of HyTM documents should be described in terms of deserialization to a grove that is described by a property set.

In addition to considering the form that the HyTM Syntax Specification should take, the Norwegian NB suggests that the following major issues should be discussed and resolved at the London meeting:

• facets
• mnemonics
• locator handling
• scope, addthems and addthms
• identity attributes

3.4 N395 (CXTM)

The Canonical XTM specification should make even clearer that CXTM is not intended to provide a general purpose methodology for canonicalizing every conceivable topic map, but merely as a way of enabling conformance testing of topic map processors. A call for trial implementations should be made in order to test the feasibility of the approach currently taken.

3.5 N393 (RM)

The Norwegian NB still seeks a clearer explanation of what the Reference Model is supposed to do and why the work on it is important. If the principal purpose of the RM is to allow the definition and interoperability of multiple "Topic Map Applications" (TMAs), examples of such TMAs should be given, along with their definitions in terms of the RM and a clear practical demonstration of how the RM leads to increased interoperability. Examples of mappings from the SAM and RDF to the RM would be particularly useful in this respect.

The position of the Norwegian NB is that the RM in its current form (N393) extends ISO 13250 and the concept of Topic Maps in ways that go far beyond a restatement of the existing standard, which is what was proposed in N358 and approved by National Body ballot in N388.

One indication of this is the amount of new terminology it introduces: Of 34 terms in the glossary (discounting abbreviations), only 8 are recognizable from ISO 13250:2000 or the XTM 1.0 specification (fully merged, merging, reified, role, role player, subject, topic, topic map). The remaining 28 are new. A specification where more than 75% of the terminology is new can hardly be regarded as a "restatement".

In our opinion, the RM represents a further generalization of the concept of Topic Maps that may or may not prove useful in the future. However, for the time being there is no industry demand for such a standard. The Norwegian NB is prepared to support continued work on the RM, in the hope that it may lead to useful results, but only on the following conditions:

(1) The work is pursued as a separate work item, not as part of ISO 13250.

(2) A new name is chosen to reflect the fact that the RM goes beyond Topic Maps as defined in ISO 13250 and to avoid confusion in the marketplace as to the nature of Topic Maps.

To the extent that the RM can be shown to embody concepts useful to Topic Maps as defined in ISO 13250, the Norwegian NB is open to having those added to the SAM during the editorial process between ballots, provided they do not seriously delay the progress of the revised standard. (One such concept that could usefully be given greater prominence in the SAM is that of the SLUO, which should be renamed "colocation objective" in order to convey greater meaning to a wider community.)

The Norwegian NB proposes that a new work item proposal (NP) be prepared for an international standard whose goal is to describe a model for achieving interoperability between various knowledge representation formalisms, including Topic Maps and RDF. We suggest Dr. Steven R. Newcomb as Editor and propose "Knowledge Aggregation Model" as the title of the standard.