Re: Semantic Web & oil & gas

Since this is now a Community Group, this kind of communication, which
contains technical opinions and can be viewed as part of the "work" of the
group, should be sent to the public email group -- which I have cc-ed.

On Mon, Nov 12, 2012 at 4:24 PM, Rurik Thomas Greenall <
rurik.greenall@gmail.com> wrote:

> Hi folks,
>
> I have been silent a while for the simple reason that I have left the
> group, however, since I seem to be on the mailing list anyway, I thought
> I'd share a few thoughts with you about why I struggle with "semantics" in
> the oil & gas scene (as I exit it stage left — my time at Statoil will end
> at the end of the week).
>
> I have limited experience with oil & gas, so excuse my lack of knowledge
> regarding this domain.
>
> In the time I have worked with Semantic Web in oil & gas, I have noted
> that there is a strong tendency to view the kind of activity done in
> "semantic data modelling" <
> http://en.wikipedia.org/wiki/Semantic_data_model> as a pertinent Semantic
> Web activity. Thus, we talk of ISO/ISA/MIMOSA/OpenA&M standards that have
> no normative representation in RDF.
>
> From the point-of-view of a Semantic Web programmer like myself (or indeed
> any of those that I work closely with), "semantic data modelling" of this
> kind is not a Semantic Web activity. We're more concerned with the core
> view of the Semantic Web described in <http://www.w3.org/2001/sw/>. I
> quote: "The Semantic Web provides a common framework that allows data to be
> shared and reused across application, enterprise, and community boundaries.
> It is a collaborative effort led by W3C with participation from a large
> number of researchers and industrial partners. It is based on the Resource
> Description Framework (RDF)."
>
> The last sentence is essential, especially considering the current view of
> what RDF is (look at any of the documentation provided by the W3C), which
> is in essence a view based on <
> http://www.w3.org/DesignIssues/LinkedData.html>, which entails HTTP-URIs.
>
> Given that the influence of TBL's Linked Data document is so pervasive in
> the RDF documentation, it's clear to me (and much of the rest of the
> community) that Linked data is the Semantic Web. Attempts at providing an
> ontological view (cf. ISO15926) are based on a misapprehension*.
>
> Given this, the aim of any group broaching Semantic Web activity should
> involve identification of relevant existing components and potentially the
> production of new ones that fit into this framework.
>
> Note now the first sentence in the quote: "The Semantic Web provides a
> common framework that allows data to be shared and reused across
> application, enterprise, and community boundaries". While HTTP facilitates
> this, licensing is a key issue; here it needs to be understood that it is
> not an option to fall back onto non-normative versions of standards that
> require substantial sums for access to the normative documentation (cf.
> ISO15926). We need truly open standards produced by this community.
>
> My friendly challenge to the group then is to move towards the core of the
> Semantic Web and away from the niche we're working in.
>
> *See Gruber, T. R. (1995). Toward principles for the design of ontologies
> used for knowledge sharing. International journal of human computer
> studies, 43(5), 907–928, which explains "An ontology is an explicit
> specification of a conceptualization. The term is borrowed from philosophy,
> where an Ontology is a systematic account of Existence." Note, that an AI
> ontology does not equate to a philosophical ontology, but is rather a
> declarative declarative representation of a set of concepts from a domain.
>
> Regards,
>
> Ru.
>