Linked Data discussions require better communication

To be honest, this entire thread has reminded me of the lengthy threads on the ontolog listserv that finally caused me to unsubscribe. I could not characterize those threads as discussions because so many of the participants were actually talking past each other and were making assertions based on their particular perspective. And many of those threads involved discussants whose perspectives lived on entirely different levels of the subject matter: Ontology as a philosophy, ontology as a first order logic language and ontology as a way to share conceptual models a la Gruber[1]. These are not entirely disjoint domains, but one has to be very careful to ensure the discourse takes place across those levels and between the domains otherwise the purpose and focus of the discussion is lost.

Seems the same was happening here. I gather that Kingsley was attempting to ensure that we don’t forget that the roots of RDF and triples goes way back to early work on E-R diagrams. Fine. And it seems others were frustrated because they didn’t want to lose the hard-won set of W3C specifications and standards that would enable Linked Data to be more than a theoretical exercise. Also good. But it wasn’t clear to me for a while what Kingsley’s intent was in his posts – some context would have been very helpful to me. It was only when I remembered that Virtuoso takes data from a very wide variety of sources that it occurred to me that Kingsley’s perspective involves looking for triples anywhere and everywhere regardless of the source format & syntax. I could be wrong so I’m checking my assumptions up front here.

Perhaps if this kind of thread starts up again:

  1.  Restate/reflect ideas that in other posts that are troubling/puzzling and ask for confirmation or clarification.
  2.  Restate the actual subject and focus of the discussion; the subject line just doesn’t always cut it.
  3.  Do more explication with the awareness that we might be talking about two (or more!) related but separate ideas/concepts. Or we could be using the same terms but with slightly different definitions.
  4.  Define the terms inline rather than just linking out. One’s interpretation of an external standard or specification could be different from someone else’s, so I think it would be good to own it.

I learn so much from most of the discussions that do take place since I am still learning how the semantic web works – I get it on a conceptual level but I’m really interested in how to ground the conceptual model in a useful, usable form. I look forward to many other interesting threads.

Paul Courtney

[1] Gruber, Thomas R.<http://en.wikipedia.org/wiki/Tom_Gruber> (June 1993). "A translation approach to portable ontology specifications"<http://tomgruber.org/writing/ontolingua-kaj-1993.pdf> (PDF). Knowledge Acquisition<http://en.wikipedia.org/wiki/Knowledge_Acquisition> 5 (2): 199–220.

:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:~:

Paul K. Courtney, MS
Applications Specialist/Biomedical Informaticist
Information Systems
Dana-Farber Cancer Institute
T: 617.582.7389
C: 603.727.8171
F: 617.632.4030

On 6/20/13 7:15 AM, "Kingsley Idehen" <kidehen@openlinksw.com<mailto:kidehen@openlinksw.com>> alleged:

On 6/19/13 10:47 PM, Young,Jeff (OR) wrote:
My impression is that Kingsley is arguing that triples is triples. Concrete syntax is irrelevant, even if those triples are barely recognizable by naive agents. If that's what he's saying, I would agree. Converting barely recognizable triples into a standard form is a trivial process.

Yes, that's my point. It's why I say that RDF didn't invent the Triple.

I've posted a document denoted with the URI/URL
<http://kingsley.idehen.net/DAV/home/kidehen/Public/DropBox/Public/Linked%20Data%20Resources/linked-data-rdf-test2.ttl>
in defense of my claim :-)

Kingsley

Jeff
________________________________________
From: David Booth
Sent: Wednesday, June 19, 2013 10:20:49 PM
To: Young,Jeff (OR)
Cc: Luca Matteis; Kingsley Idehen; Linked Data community
Subject: Re: Proof: Linked Data does not require RDF

Hi Jeff,

I guess I could have said *concrete*-syntax-independent to be more
precise -- to distinguish it from the *abstract* syntax (or model) --
but "serialization-independent" works too.  Or "format-independent".

David

On 06/19/2013 09:55 PM, Young,Jeff (OR) wrote:
David,

I think you've confused syntax-independence with
serialization-independence. RDF is syntax-dependent. The syntax is
triples. OTOH, triple syntax can be serialized in a wide variety of
ways.

Jeff

-----Original Message----- From: David Booth
[mailto:david@dbooth.org] Sent: Wednesday, June 19, 2013 9:42 PM
To: Luca Matteis Cc: Kingsley Idehen; Linked Data community
Subject: Re: Proof: Linked Data does not require RDF


Can you please then setup a pool asking "Does creating and
publishing Linked Data require knowledge of RDF?"
I would be willing to make such a poll if it seemed that people
wanted it, but I don't think it is necessary.  There are *many*
document formats that can carry RDF, and it seems self-evident that
someone who publishes an RDF-interpretable format like JSON-LD or
(GRDDL-enabled) XML may not understand RDF **at all**.  This is one
of the great benefits of RDF being syntax independent.  The JSON-LD
group understood this very well and did a great job crafting the
JSON-LD spec to ensure that web developers would *not* have to
understand RDF in order to happily publish their JSON-LD.

If the data is *interpretable* as RDF, then who cares whether the
publisher understood RDF?  It seems irrelevant to me.

David












--

Regards,

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









The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.