Sjir, John B, David P, and Kingsley, (01)
I'll start with the following point by Sjir, which gets to the heart
of many of my complaints: (02)
SN
> I do believe that the concept of Conceptual Schema of ISO TR9007
> has a much better chance of helping in the decomplexification process. (03)
I agree. I was disappointed that the ANSI-SPARC work on the conceptual
schema ended with a technical report in 1978 instead of a standard. (04)
In my 1984 book, I tried to relate the semantic issues in databases,
knowledge bases, and natural languages. I participated in IFIP WG 2.6
during the late 1980s, and I liked the ISO work, which ended in a TR
in 1987. I also participated in a later ISO project on the conceptual
schema, which ended in another technical report in 1999. (05)
For the Semantic Web, I liked the vision by Tim Berners-Lee. The
Semantic Web Logic Language (SWeLL), which he proposed in 2000 looked
like the kind of foundation we needed. In 2003, Guha and Hayes
developed LBase (Logic Base), which could serve as a unified semantics
for *all* the SW logics. (See http://www.w3.org/TR/lbase/ ) (06)
But certain people managed to convince the W3C voters, most of whom
had little or no understanding of logic, that LBase was too expressive.
(Its expressive power, by the way, is what Tim B-L stated as the
*requirements* for SWeLL in his proposal of 2000.) LBase is just a TR. (07)
I still believe we can and must have a unified semantic foundation that
supports the requirements for a conceptual schema, a truly semantic SW,
and the legacy systems that run the world economy. (08)
JFS
>> for something as simple as RDF, the complexity is not in the language,
>> but in the choices about what to describe and what aspects to represent. (09)
JB
> I agree. The best ontological approach would be to develop a paradigm
> that permits linked data but which also addresses other valid ontological
> needs. Legacy systems are significantly important commercial systems and
> if they are not in the mix for ontologies, then that need is going unmet. (010)
Yes. Any system that cannot interoperate with the trillions of dollars
invested in legacy software has no chance of becoming mainstream. (011)
DP
> Comparing RDF and a Mac or iPhone user interface is just silly. (012)
David E was comparing RDF to JCL. I made the point that any usable
system must have multiple interfaces at different levels. They include
very detailed notations for the system programmers at the deepest levels
to interfaces like the Mac or iPhone at the top. (013)
DP
> A more reasonable comparison is the RDF stack and the SQL stack... (014)
I certainly do *not* want to defend SQL. I used to call it the worst
notation for logic ever invented. There were much better notations
than SQL in the 1970s, and the goal of the conceptual schema work
was to support an open-ended variety of different notations with
the same fundamental semantics. (015)
DP
> our experience of 4 days training being plenty to get people being
> productive with the languages and our TopBraid tool. (016)
That's great! I am always supportive of any practical languages and
tools that help people become more productive. (017)
But one point I emphasize is that there is no such thing as a one-size-
fits-all notation that is ideal for all purposes. I would not do
algebra in English, nor would I talk on the phone in predicate calculus. (018)
But we can have a unified semantics that can support a very wide range
of different notations for different people for different purposes. (019)
KI
> As much as I admired Steve Jobs he never really understood the value in
> true openness. His elegant interfaces ultimately lead to Apple technology
>silos. (020)
I agree. In the mid 1980s, Bill Gates wanted to license the Mac user
interface. If Steve had agreed, he could have made a mint licensing
the Mac interface. But that would still not make it open. (021)
KI
> This is the whole thing in a nutshell, the ability to move the power
> of basic natural language sentences in the realm of computing... (022)
That gets into a large number of complex issues. But it's almost
midnight, so I'll stop. (023)
John (024)
_________________________________________________________________
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 (025)
|