Some Issues Related to DCMI Extension Namespaces

2005-05-13

-- "Documentation clutter":  If terms are going to "expire"
   (i.e. cease to be actively maintained) at some point, we 
   would want to have some way to present a view of the
   "fresh" terms.

-- Maintenance: The SKOS vocabulary is maintained directly
   in RDF, which is used to generate Web pages.  We should 
   perhaps look into this as a method for maintaining EXT-NS
   terms, or perhaps even share tools with the SKOS community.

-- Scope Statement: There is an old action on Andrew to draft
   a Scope Statement: what are the boundaries and criteria for
   inclusion of terms in DCMI (i.e. what 'resource discovery'
   and 'cross-domain' means in practice), to be included
   in UB Mission and Principles, UB Process, or elsewhere.
   Is this still relevant?

-- Status of "Conforming": In March 2004 in Bath, we reaffirmed
   that the Usage Board can assign the the status of
   "conforming" to an Application Profile based on a
   significantly more thorough review focused on elements
   and element refinements at the point of review.  The AP
   designated as "conforming" (i.e., a snapshot of the AP
   document at the time reviewed) would be archived on the
   DCMI Website.  Changes to the AP should result in a new
   AP and resubmission to the UB (i.e., for new "time stamp").

-- To batch or not to bach.  The DCMI Extension Namespaces
   proposal is deliberately in the plural - "namespaces".
   The idea is that batches of terms will be subject to 
   maintenance by different communities or organizations.
   Putting all terms into one big would make it less 
   convenient to divide up maintenance responsibility.

-- Naming (URIs) - base URI.  The DCMI Extension Namespaces
   need a base URI; the current preference is to
   use something based on http: //dublincore.org/,
   such as http://dublincore.org/extension/ or http:
   //dublincore.org/ext/.  However, the argument has been made
   that, for consistency, http: //purl.org/ should be used.

-- Naming (URIs) - sub-namespace URI.  Given that
   the DCMI Extension Namespaces will be subdivided into 
   "sub-namespaces" with separately maintained batches of 
   terms, then the question is how to name those batches.
   In theory, URIs are opaque.  In practice, however, people
   do try to interpret the strings used in a URI.  We should
   therefore be careful to choose names that do not convey
   misleading notions.  Some possibilities:
      
   a) To use strings such as "dccdwg", as in:
      http://dublincore.org/ext/dccdwg/.

   b) To assign a "time-stamped" base URI to a working 
      group, e.g., http://dublincore.org/extra/2005/09/
      for Collection Description WG.  Advantage: avois
      strings which reflect the origins of a batch
      of terms in a particular working group or activity.  

   c) To number the extension namespaces sequentially:
      http://dublincore.org/ext/01/ - for collection description
      http://dublincore.org/ext/02/ - for libraries

-- Use of the term "Namespace".  DCMI needs to be clear as
   to whether a DCMI Namespace is "a collection of URIs"
   (i.e., the generally accepted notion that a "anemspace"
   is a "set of names") or "a collection of terms identified
   by URIs" -- and at any rate not an XML namespace.  DCMI --
   but not necessarily the Usage Board -- would need to use
   the term "namespace" consistently in the Abstract Model
   and the Namespace Policy.  For example:

        DCMI namespace

        A collection of 'term names' (i.e. 'term URIs')
        where each term is assigned a URI that starts with
        the same 'base URI'.  The 'base URI' is known as the
        'DCMI namespace URI'.  (Note that a 'DCMI namespace'
        is not the same as an 'XML namespace').

        Note that the grouping of 'term names' into a 'DCMI
        namespace' is orthogonal to the grouping of 'terms'
        into a 'DCMI vocabulary'.  'Term names' are grouped
        into 'DCMI namespaces' in order to ease the assignment
        of URIs to 'terms' and to streamline their use in
        particular encoding syntaxes.  'Terms' are grouped into
        'DCMI vocabularies' in order to meet a functional need.