Dear John: (01)
I would argue that well-designed conceptual schemas for the Web of
Linked Data (Semantic Web), combining (02)
a) the art and science of conceptual modeling from Chen's ER approach to
Guarino and Welty's OntoClean and
b) standardized formalisms for representing and sharing the schemas
(e.g. RDFS, OWL DL, OCML, or F-Logic) (03)
are actually meeting your requirements 1. and 3. a) and don't prevent
anybody from obeying 2. and 3. b) and c). (04)
For example, you can represent GoodRelations-related data as triples,
but you don't have to.
Should there be more efficient ways of storing respective data at the
physical level, that is fine.
Though there is a dominance of the triples in the common perception of
Web of Linked Data data structures,
that is IMO an implementation detail that can be changed without
destroying the whole. (05)
What's more important for me is the use of a global, extendable naming
scheme (i.d. URIs) for referencing / linking entities
(both conceptual categories and data instances), combined with a
transportation mechanism (i.e. http) that supports the
platform-neutral and simple retrieval of meta-data associated with
entities (e.g. via 303 redirects). (06)
The only point where I think a well-designed RDFS or OWL DL schema will
partly violate your requirements is that due to
the lack of higher arity relationship types, you have to introduce
additional conceptual entities just for representing the "binding clip".
And even that point is arguable - if I remember a discussion with Robert
Meersman correctly (CC, Robert - correct me if I am wrong), he once
argued that being limited to binary relationship types is actually even
cleaner from the modeling point of view, because you do not
require all parts of the relationship instance to be present prior to
modeling any part thereof. (07)
Best wishes and have a good trip! (08)
John F. Sowa wrote:
> Phil and Kingsley,
> I am not arguing for or against any particular technology, and
> my points are independent of any implementation or benchmark
> comparisons. I cited benchmarks only to counter claims that
> the logical interface should be biased toward one particular
> physical layout. My primary point about benchmarks is that they
> change with the weather, and they should never be used to bias
> the logical views of the data.
> The primary points I am making were well understood by people
> like Don Knuth in the 1960s and by the conceptual schema work
> of the 1970s:
> 1. The logical interface should be independent of any
> physical storage layout or access method.
> 2. Any optimizations tailored for a particular technology or
> algorithm are certain to be obsolete when the next new
> technology or algorithm is invented.
> 3. Therefore, the ANSI/SPARC three-schema approach of 1978 is
> still the best advice about how to design standards for
> data access:
> a) The conceptual schema (logic + ontology) should be
> completely neutral about storage mechanisms, benchmarks,
> algorithms, etc.
> b) The physical schema that specifies how the data is
> stored and organized is the level where the latest and
> greatest technology of the day should be considered.
> c) The application schema defines the APIs to programming
> languages and other software systems.
> This wisdom has never been superseded, and it is independent
> of any and all benchmark battles.
> However, I also realize the importance of smooth transitions
> that enable new technology to interoperate with legacy systems
> for indefinite periods of time. Therefore, I believe that any
> new standards must be upward compatible and interoperable with
> current standards. If the new standards are truly neutral,
> they should support older standards that may have some kinds
> of built-in bias toward one physical layout or another.
> I have been traveling for the past two weeks, so my email
> time has been limited. I'm catching a plane tomorrow morning,
> and I don't know when my next email opportunity will occur.
> But I'll be home on Aug 2, and I can catch up with whatever
> discussions have transpired.
> John Sowa
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
e-business & web science research group
universitaet der bundeswehr muenchen (011)
www: http://www.unibw.de/ebusiness/ (group)
twitter: mfhepp (012)
Check out the GoodRelations vocabulary for E-Commerce on the Web of Data!
Talk at the Semantic Technology Conference 2009:
"Semantic Web-based E-Commerce: The GoodRelations Ontology"
Tool for registering your business:
Overview article on Semantic Universe:
Project page and resources for developers:
Tutorial at ESWC 2009: The Web of Data for E-Commerce in One Day: A Hands-on
Introduction to the GoodRelations Ontology, RDFa, and Yahoo! SearchMonkey (019)
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (01)