ISO/IEC JTC 1/SC34 N0271

ISO/IEC JTC 1/SC34

Information Technology --

Document Description and Processing Languages

Comments on N266 Proposed Requirements for Level 1 Model

This document responds to document N266.

The "Level 1" Model is intended to be a specialized application of the Core Model that will establish a rigorous common understanding of the XTM 1.0 and existing HyTime-based ISO 13250:2000 syntaxes for Topic Map interchange. Some features of N266 appear inconsistent with this intention:

1. N266 describes itself as providing requirements for "the foundational model" of Topic Maps. The Topic Maps standard cannot have both a "Core" and a separate and different "Foundation". As far as we know, there is no dispute that the draft Core model described in N243 and diagrammed in N244 is more "foundational" than the draft "Level 1" or "Infoset-based" model described in N261.
2. N266 says, "The foundational model shall describe the structure of topic maps and the process of parsing them in such a way that conformant implementations will be interoperable." The purpose of the Core Model can correctly be perceived as being to render the problem of interoperability moot, at least at a very general level. N266 would be a more useful requirements document if it explicitly concerned itself with the interoperability of the topic maps that employ functionalities, semantics, and syntaxes that the Level 1 Model builds on top of the Core Model.
3. N266 says, "Anything that is not explicitly part of the model shall not considered to be part of topic maps." This statement could be misconstrued to mean that the Level 1 model defines what Topic Maps are, and there is no need to refer to the Core Model.
4. N266 says, "There shall be no such constraints that are not part of the model specification." Again, this statement could be misconstrued to mean that the Level 1 model defines what Topic Maps are, and there is no need to refer to the Core Model.
5. N266 says, "The data model shall be written in such a way that third parties can write specifications defining the process of building instances of the model from data sources other than the two standardized topic map syntaxes." There is nothing wrong with this, but it will be unnecessary to state it here if the Core Model meets its requirement to provide the necessary rigorous conventions for defining Applications, and if the Level 1 Model conforms to the Core Model.
6. N266 says, "The data model shall define the nomenclature for the individual components of topic map structures." This is too broad. Some of this nomenclature must be defined in the Core Model, and some of it in the Level 1 Model.
7. N266 says, "The data model shall capture the key constructs of topic maps as first class entities..." At least some of "the key constructs of topic maps" will be defined in the Core Model. It is not clear what N266 means by "first class entities", but it is clear that one of the key purposes of making a distinction between the Level 0 (Core) and Level 1 Models is to establish that some "entities" are more fundamental than others. Will we use the term "zeroeth class entities" to describe the key constructs of the Level 0 Model?

There are several statements in N266 that appear to be intended to impose constraints and requirements on the standard as a whole, and on the standardization process, rather than on the Level 1 Model. These statements are outside the scope of the purpose of N266.

1. N266 says, "The foundational model shall be specified as part of the ISO 13250 standard." As far as we know, there is no dispute about this, but this statement doesn't belong here because it proposes a constraint on other things than the Level 1 Model itself.
2. N266 says, "The foundational model shall be 100% compatible with every aspect of the implicit data model of XTM 1.0 that is actually defined in the XTM 1.0 specification. This includes annex F." If adopted, this requirement would constrain the process of standardization very severely, leaving the Working Group in the position of being unable to fix problems that remain in the XTM 1.0 Specification, even in the interests of achieving the goal of establishing a Level 1 Model that will represent the consensus. Regardless of the context within which it has been or may ever be made, this purpose of this statement is at odds with the idea that normal ISO procedure will govern the further development of the Topic Maps standard. This requirement would have the effect of preventing due deliberation regarding any of the many technical issues that were not adequately discussed in the process that led to the publication of the XTM 1.0 Specification.
3. N266 says, "The foundational model shall not contradict any constraints on the model laid down by XTM 1.0, including those of annex F." The same objection applies.
4. N266 says, "The foundational model shall be 100% compatible with the interpretation of the XTM 1.0 syntax as defined in XTM 1.0, including annex F." The same objection applies.
5. N266 says, "The foundational model shall be able to represent all logically significant aspects of ISO 13250 topic map documents." This is too vague; it is not clear what is meant by "logically significant".
6. N266 says, "The foundational model shall not contradict any constraints on the model laid down by ISO 13250, except in so far as they are contradicted by XTM 1.0. In cases of discrepancy XTM 1.0 shall take precedence." There are at least three problems with this extraordinary requirement. (1) Discrepancies are a matter of the interpretation of each of the standards. The interpretations of both standards must be fully agreed upon before the implications of this requirement can be understood. (2) If adopted, this requirement would privilege as-yet-unspecified interpretations of both standards, and then demand that one of them must be preferred over the other. Such a policy would prevent one or both of the interpretations themselves from being adjusted in such a way as to achieve harmony. (3) It would be especially inappropriate, in an ISO standards-making process, to require the prejudgment of all technical issues in favor of the resolutions reflected in a publication that emanated from a non-ISO process, preferring them even over the resolutions reflected in a publication that resulted from the regular ISO process.