In message <[email protected]>, George Oates <[email protected]> writes >Hi, > >Richard Light wrote: >> I see that you're going for the "heading in the URL" approach: >> >> http://openlibrary.org/subjects/texas_in_fiction >> >> rather than minting identifiers for subjects. Is this the long-term >> plan? > >If you mean, _will there be numeric identifiers for subjects?_ the >answer is no. >The bit after the /subjects/ can act as an identifier, can't it?
Yes it can: this is, for example, how dbpedia identifiers are typically formed. The key thing is that you should publish the identifiers which you support (see below), and give some additional information about them so that everyone is clear about their intended meaning and scope. >> Will there be Linked Data support, so that this subject URL is >> dereferenceable with some supporting RDF/XML, for example? > >What does dereferenceable mean? Sorry: Linked Data jargon. It simply means that if you request this URL you get something back. The general LD approach is that you use content negotiation to decide what that something might be. If the HTTP request has an Accept header which says "I want RDF/XML, please" then that's what the server should return. If the HTTP request makes no particular demands, it's OK to return standard HTML (i.e. a human-readable set of info about the concept). Richard -- Richard Light _______________________________________________ Ol-discuss mailing list [email protected] http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss To unsubscribe from this mailing list, send email to [email protected]
