On 4/8/12 2:27 PM, John F Sowa wrote:
> David and Kingsley,
>
> DE
>> Linking all this stuff together isn't new. How long has ODBC been
>> around? At least 20 years?
>>
>> What does Linked Data add to the challenge of "I wonder what's in this
>> column/field/cell?"
> I would like to add that the ANSI/SPARC conceptual schema addressed
> these issues in 1978. It was designed to support *all* the kinds of
> databases available in the 1970s -- that included hierarchical (IMS),
> network (CODASYL DBTG), and relational (IBM, Oracle, Ingres).
>
> In 1987, ISO produced a TR that extended the ANSI report to include
> all versions of databases in those days *and* the work on knowledge
> based systems in AI. The annual VLDB conferences were very active
> for years in supporting R& D on these topics. They had implemented
> commercial systems that demonstrated solutions to these problems.
>
> KI
>> We've developed and deployed ODBC Drivers for all major RDBMS
>> engines for 20 years. One problem that ODBC never addressed was
>> the ability to extend data access beyond RDBMS engines. It also
>> had a degree of platform specificity, so Windows always had the
>> richest collection of ODBC compliant applications. I've written
>> some posts [1][2][3][4] about the data connectivity and platform
>> specificity shortcomings of ODBC, JDBC, ADO.NET, and OLE-DB.
> I followed up those links and checked out the FAQ on openlinksw:
>
> http://virtuoso.openlinksw.com/virt_faq/
>
> That is good technology, and the Semantic Web should have *started*
> with that approach. They could have done the world a major service:
> (1) integrate all those data sources with a uniform conceptual schema,
> and (2) adopt an *implementation-independent* logic to define the
> semantics of the data and the operations.
>
> This is *not* a criticism with 20-20 hindsight. This is what the
> DB experts had been analyzing, designing, and implementing long
> before Tim B-L proposed the SW in 1994.
>
> The major reason why the ANSI/SPARC conceptual schema had not been
> adopted as a standard is that the DB vendors did not want to support
> an implementation-independent standard -- for the simple reason that
> they wanted to lock customers into their implementations.
>
> But the SW gang had the hubris to claim "We're so big and so new
> that we can ignore everything that went before." They did. And
> they repeated every mistake of the past plus a lot of new ones.
>
> John
>
> _________________________________________________________________
> 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
>
>
John, (01)
Yep! (02)
We always perceived the Semantic Web project as a source of standardized
technology that enabled virtualization of heterogeneous data sources, in
a manner that exposed virtues of both the intensional and extensional
dimensions of DBMS technology. Unfortunately, we also assumed that
everyone already understood the virtues of building upon existing
efforts such as ODBC, and to some degree I actually strongly fault
myself for *not* sticking to that world view when evangelizing virtues
of Semantic Web related technologies and standards, modulo RDF specificity. (03)
The desire of RDBMS vendors to lock-in customers ultimately rammed ODBC
down a predictable technology cul-de-sac. Thus, today we have better
context and technology for making the ANSI/SPARC conceptual schema idea
of yore a contemporary reality. None of this is RDF specific or
dependent, as you know :-) (04)
-- (05)
Regards, (06)
Kingsley Idehen
Founder& CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen (07)
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
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 (01)
|