[SCL] equivalence of SCL, GXL, RDF, RDF/XML, and XML?

[Rich Morin]

The design of SCL seems to assume a "Turing equivalence" for graph
encoding methods.  That is, a minimum level of expressiveness that
allows encoding of attributes and relationships, supports traversal,
etc.  The idea also seems to be implicit in the existence of other
graph interchange formats such as GXL, RDF, RDF/XML, XML, and YAML.

In particular, it should be possible to define an RDF (and therefore
RDF/XML) encoding for any SCL structure.  The TAP (tap.stanford.edu)
folks seem to depend on this, basing their entire KR system on an
"edge list" of RDF statements.

Or, using another approach, it is trivial to implement an edge list
as an RDBMS table.  The tricky part seems to be finding better ways
than raw SQL to retrieve the information.  Several groups are working
on this problem, but I don't see a "category killer" emerging yet.

So, this might be a useful place for SCL to exert some influence.  A
definition for encoding SCL as RDF could be very useful.  If it went
a bit further and encompassed the use of RDBMSs to store RDF-encoded
SCL, so much the better.  Is anything like this on the table?

-r
-- 
email: rdm at cfcl.com; phone: +1 650-873-7841
http://www.cfcl.com        - Canta Forda Computer Laboratory
http://www.cfcl.com/Meta   - The FreeBSD Browser, Meta Project, etc.