 |
 |
On Sat, May 9, 2009 at 00:32, Jonathan Rochkind <[log in to unmask]> wrote: > I don't understand from your description how Topic Maps solve the > "identifying multiple versions of a standard" problem. It's the mechanism of having multiple identifiers for Topics, so, in pseudo ; Topic "MARC21" psi "info:ofi/fmt:xml:xsd:MARC21" psi "http://loc.org/stuff/marc21" property #mime-type "whatever for the binary" Topic "MARC 1.1" is_a "MARC" psi "info:srw/schema/1/marcxml-v1.1" psi "http://loc.org/stuff/marcxml-v1.1" property #mime-type "whatever 1.1" Topic "MARC 1.2" is_a "MARC" psi "info:srw/schema/1/marcxml-v1.2" psi "http://bingo.com/psi/marcxml" property #mime-type "whatever 1.2" Or, if if "MARC 1.2" is backwards compatible with 1.1 ; Topic "MARC 1.2" is_a "MARC 1.1" psi "info:srw/schema/1/marcxml-v1.2" Or, if I make my own unofficial version ; Topic "MARC 2.0" is_a "MARC 1.2" psi "http://alex.com/psi/marc-2.0" This is enough to hobble together what is and isn't compatible in types of formats, so if your application is Topic Maps aware, this should be trivial (including what format to ignore or react to). The point is that you don't need *one* identifier for things; Topics are proxies for knowledge, and part of the notion of "knowledge" is what identifies that knowledge. Multiple PSIs help us leverage both rigid and fuzzy systems. As to the identifiers themselves (as in, the formatting), is that important? Anyway, I'm suspecting I don't see what the problem seems to be. To create "the best identifier" for things seems a bit of a strange notion to me, but is this based on that there is only (or rather, that you're trying to create) one identifier for any one thing? Alex -- --------------------------------------------------------------------------- Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps ------------------------------------------ http://shelter.nu/blog/ --------