|Title:||Japan National Body Comments on ISO/IEC JTC/SC34 N0393 Topic Maps Model|
|Project editor:||Steven R. Newcomb, Michel Biezunski, Martin Bryan|
|Action:||For information and review|
|Distribution:||National Bodies and Liaisons of SC34|
|Refer to:||ISO/IEC JTC/SC34 N0393 Topic Maps Model|
The Japan National Body has done a preliminary review of N0393, with emphasis placed on what role this document plays in the New Work Item for ISO 13250, voted and approved in the N0388 Ballot.
This seems to be a proposal to supersede the model formally called the Reference Model; however, this is not clearly stated, so we request confirmation as to what purpose the contents expressed herein will serve. Is it to supersede the Reference Model, or is this a new model, independent of what is described in the N0358 New Work Item Proposal for Restatement of Topic Maps?
The authors state that there was a previously published version, that is, N0344. As far as we can ascertain, the authors are contributing a new draft to supersede a previous draft, but since there is a new name to the model it is not clear if the authors intend it to become one of the parts of the multipart standard as expressed in the Proposal for the Restatement of ISO 13250 Topic Maps or if it is a subsidiary proposal for a new topic map paradigm to be developed separately.
If this is a draft to supersede a previous one, the Japan National Body strongly advises from now on, not to submit the draft with a different title or name change since it causes undue confusion. Sufficient discussion and agreement with the committee should be required before a change to avoid ambiguities.
This document begins with redefining concepts previous defined in ISO 13250, so it is not clear what purpose this document serves. The committee needs to discuss whether there is a need to define key concepts in more than one part of the standard.
If this, in fact, is to be considered as one of the parts, there should not be more than one word to describe the same concept, and agreement must be reached to decide on the best word for each of the concepts. All parts of the standard must work together and this includes the definitions.
The requirements expressed in 0 Introduction of N0393 seem to go beyond the purpose and justification in the following points described in N0358 Proposal for the Restatement of ISO 13250 and in the business requirement described below:
The business requirement included in the PROJECT ACCEPTANCE CRITERIA is stated as follows:
A Business Requirement
A.1 Market Requirement Essential _X_ Desirable ___ Supportive ___
Without a formal data model (currently termed the Standard Application Model) it is not possible to guarantee interoperability between implementations or define a standard mechanisms for expressing queries and constraints on topic maps. (Note: A Topic Map Query Language and a Topic Map Constraint Language are already approved work items with the ISO/IEC numbers 18048 and 19756 respectively. Work on these cannot be completed without the formal data model.)
It is clear to us that there is a need for a formal data model which is now N0396, the Standard Application Model. This is what the Japan National Body has voted on. It is not clear to us what role the Topic Maps Model plays in this restatement.
In N0278, the Reference Model is defined as "the new name for the level 0 model (PMTM4), which emphasizes that it is intended to be used as a reference point for knowledge representations rather than as an implementation guide." Is the definition and purpose of the Topic Maps Model the same as that of the Reference Model? This requires clarification. The Topic Maps model does not seem to follow this definition; in addition, it seems outside of the scope of the restatement of ISO 13250 and would be best considered as a new work item in its own right with its own requirements.
If the Reference Model (here called the Topic Maps Model) is to become one of the parts of the multipart standard, the glossary terms need to be in agreement with terms expressed in the other parts. The same words in one part should not be defined differently in another part. For example, there is currently a discrepancy between the term "application" in the "Topic Maps Model" and in the "Standard Application Model." Many terms here are different from what is in ISO 13250 now, so we guess that they are proposals for changes to or addition of terms.
If the authors intended to provide new glossary terms that will add to or supersede other terms in ISO 13250 then this should be stated explicitly. In other words, we request an explicit listing of new concepts and extensions that goes beyond what is described in ISO 13250 and the rational for including them.
These topics need updating in the multipart standard, however these seem to require further work, if they are to be included in some part of the multipart standard. Section 7 is a description of deserialization in general. It is difficult to know exactly what the context is without a description of the syntax or what type of deserialization is being performed. This leads us to the conclusion that it may have a better use in another part of the standard. It is not clear what explicit or implicit deserialization is and its context. Section 8 is needed but should be part of the conformance clause of the standard and not part of this model. For section 9, the meaning of merging is a central concept of all topic maps. Please clarify that this is a proposal for a new description of merging than that expressed in ISO 13250.