|  
         
      Draft 
        Requirements for OASIS Topic Maps Published Subjects TC 
        This 
        version : November 5, 2001 
         
          
         
        Status of this document : 
        The content of this page is only informative and should not be used or 
        referred to otherwise. 
        Please send any comments to Bernard 
        Vatant 
      1. Statement 
        of Purpose 
      Whereas 
        most other semantic specifications and languages are resource-centered, 
        Topic Maps are subject-centered. Since every Topic in a Topic Map should 
        be a formal representation of an unique and non-ambiguous subject (abstract 
        or concrete), consistency and efficient interchange of Topic Maps must 
        rely on efficient, non-ambiguous and trustable process for definition 
        and usage of subjects' identity. 
        Neither ISO 13250 nor XTM 1.0 provide Topic Maps authors, managers and 
        users with any concrete hint of what that process should be, excepted 
        the general recommendation to refer to authoritative Published Subjects. 
        The purpose of OASIS Topic Maps Published Subjects Technical Committee 
        is to promote efficient use of Published Subjects by specifying recommendations, 
        requirements and best practices, for their definition, management and 
        application.  
        It should also provide efficient use cases based on those recommendations. 
         
      2. 
        Recommendation for Published Subjects Definition 
      The 
        TC will recommend a standard definition (grammar and syntax) for Published 
        Subjects, allowing both human users and topic maps engines, or any other 
        kind of intelligent agents, to identify and use in a non-ambiguous way 
        the elements capturing subject identity and context of publication and 
        use. 
        This should 
        include at least a formal declaration of Published Subject, name(s), description(s), 
        publication context (publisher, creator, update ...), intended scope of 
        use and/or community of users.  
      As 
        far as possible, the recommendation should be consistent with and/or use 
        related standards such as RDF description and Dublin Core elements, and 
        generally state-of-the-art metadata standards.  
         
        Some open issues: 
      
        - The recommended 
          syntax should certainly at least be specified in XML ...
 
          Should it be written in XTM or use XTM element, 
          or use a specific DTD or schema? 
        -  Given that Published 
          Subjects will most of the time set in repositories ...
 
          Should the recommendation extend to the structure 
          of repositories, relationships between Subjects such as classifications 
          and/or other kind of associations? 
        - Given that Published 
          Subjects are to be permanently available on-line ...
 
          Should the recommendation define a standard namespace 
          structure? 
           
       
      3. 
        Recommendation for Published Subjects Management Authority 
      The 
        TC will recommend a general methodology for Published Subjects Management. 
        That should include at least: 
      
        -  Requirements for 
          stability of URIs
 
        - Permanent visibility 
          and availability of publishing authority
 
        - Transparency of 
          IPR status and security requirements
 
        - Liaison and consistency 
          between repositories in the same scope, and with state-of-the-art standards 
          
 
        - Requirements on 
          process of validation and update
 
        - Versioning, including 
          backward compatibility with ancient versions
 
         
       
      4. 
        Recommendation for Published Subjects Applications and Users 
         
         
       "Published 
        Subjects Applications and Users" include the non-exhaustive following 
        list. 
        Note 
        that even if TC recommendations focus mainly on Topic Maps applications, 
        they are not restrictive to those.  
        They assume that many other agents and applications will take advantage 
        of, and use efficiently, Published Subjects. 
      
        - Semantic Web Agents 
          (Topic Maps engines and parsers, intelligent agents, spiders, search 
          engines ...)
 
          Those agents will use Published Subjects to control subject identity, 
          merge, bind, or process Topic Maps in any relevant way, on the basis 
          of the identity they provide. 
        - Topic Maps authors 
          (either human or any kind of software) 
 
          They will use Published Subjects to identify Topics when creating, updating 
          and modifying Topic Maps. 
        -  Authors of Ontologies, 
          Thesaurus, and generally speaking all Classification and Indexing tools.
 
       
      The TC will recommend 
        best practices for those applications and users, including: 
      
        - Consistent technical 
          use of Published Subjects inside Topic Maps and other applications
 
        - Consistency of 
          Published Subjects usage with the scope intended by Publishers
 
        - Respect of IPR 
          as intended and declared by Publishers
 
        - Registration of 
          users by the Publishers, whenever it is relevant and useful for the 
          community of users
 
        - And more generally, 
          any relevant term of contract between Publishers and Users
 
       
      5. Use Cases - 
        Registries Management under TC authority 
      The TC will "eat 
        its own dog food", by building use cases of its own recommendations, 
        that could include: 
      
        - Registries of Generic 
          Published Subjects, e.g. languages, countries, generic relationships 
          and associations widely used in classifications and ontologies.
 
        - Registries managed 
          by Subcommittees devoted to specific industry fields, e.g. XML industry, 
          Agriculture and Food.
 
       
     |