Crystallography Online: the website of the International Union of Crystallography

Joe Krahn raises an interesting issue.  The array_structure category
defines the organization of data, not the data itself.  imgCIF
allows you to define many frames of data, all using the same
array_structure and to disambiguate them by providing distinct
binary id's.  The fact that binary ids are numbers makes a lot
of sense when working with frames, but for more general
applications, it might be better to have a more general
identifier. Joe Krahn is proposing the creation of a more
general identifier that would combine into one id the
array_id and binary_id.

Would people find this useful?  If it is useful, what
are the implications for existing software?  Is there
a way to preserve the current functionality, but
also to add the capability of a single identifies?

   -- Herbert

At 8:38 PM -0500 2/26/07, Joe Krahn wrote:
>I find it strange that the 'array_id' is not a complete array reference.
>For example, in the new MAP category, a MAP array reference requires
>both the array_id and the binary_id:
>
>   _map_segment.array_id
>   _map_segment.binary_id
>
>It is also confusing that array_id is really an _array_structure.id. Is
>a given _array_structure.id supposed to be unique for a given array? If
>so, why doesn't it include a binary_id reference?
>
>If _array_structure.id is not intended to be unique, so that multiple
>binary arrays with the same format can reference the same
>_array_structure.id, then doesn't it make sense for 'array_id' to be a
>reference to a name given to the _array_data category?
>
>If neither of the two are meant to be unique, why not define a simple
>_array category:
>
>    _array.id
>    _array.array_structure_id
>    _array.binary_id
>
>This would make the most sens if the 'array_id' used everywhere points
>to the above _array.id name.
>
>Is there a reason not to have a name for each unique array? Is it a good
>idea, but we are now stuck with a rather messed up organization for
>historical purposes?
>
>Joe Krahn
>_______________________________________________
>imgcif-l mailing list
>imgcif-l at iucr.org
>http://scripts.iucr.org/mailman/listinfo/imgcif-l


-- 
=====================================================
  Herbert J. Bernstein, Professor of Computer Science
    Dowling College, Kramer Science Center, KSC 121
         Idle Hour Blvd, Oakdale, NY, 11769

               Office:  +1-631-244-3035
            Lab (KSC 020): +1-631-244-3451
                  yaya at dowling.edu
=====================================================