These days I increasingly qualify myself and my Semantic Web
advocacy as falling under the realm Linked Data. Thus, I tend to
use the following introduction: I am Kingsley Idehen, of the Tribe Linked Data.
The aforementioned qualification is increasingly necessary for
the following reasons:
- The Semantic Web vision is broad and comprised of many
layers
- A new era of confusion is taking shape just as we thought we
had quelled the prior AI dominated realm of confusion
- None of the Semantic Web vision layers are comprehensible in
practical ways without a basic foundation
- Open Data Access is the foundation of the Semantic Web (in
prior post I used the term: Semantic Web Layer 1)
- URIs units of Open Data Access in Semantic Web parlance i.e..
each datum on the Web must have an ID (minted by the host Data
Space).
The terms GGG, Linked Data, Data Web, Web of Data, and
Web 3.0 (when I use this term) all imply URI driven Open Data
Access for the Web Database (maybe call this ODBC for the Web) -- ability to point to
records across data spaces without any adverse effect to the remote
data spaces. It's really important to note that none of the
aforementioned terms have nothing to do with the "Linguistic
Meaning of blurb". Building a smarter document exposed via a URL
without exposing descriptive data links doesn't provide open access
to information data sources.
As human beings we are all endowed with reasoning capability.
But we can't reason without access to data. Dearth of openly
accessible structured data is the source of many ills in cyberspace
and across society in general. Today we still have Subjectivity
reigning over Objectivity due to the prohibitive costs of open data
access.
We can't cost-effectively pursue objectivity without
cost-effective infrastructure for creating alternative views of the
data behind information sources (e.g. Web Pages). More Objectivity
and less Subjectivity is what the next Web Frontier is about. At
OpenLink we simply use the moniker: Analysis for All! Everyone
becomes a data analyst in some form, and even better, the analysis
are easily accessible to anyone connected to the Web. Of course,
you will be able to share special analysis with your private
network of friends and family, or if you so choose, not at all
:-)
Recap, it's important to note that Linked Data is the foundation
layer of the Semantic Web vision. It's not only facilitates open
data access, it also enables data integration (Meshing as opposed
to Mashing) across disparate data schemas
As demonstrated by DBpedia and the Linked Data Solar system emerging around it,
if you URI everything, then everything is Cool.
Linked Data and Information Silos are mutually exclusive
concepts. Thus, you cannot produce a web accessible Information
Silo and then refer to it as "Semantic Web" technology. Of course,
it might be very Semantic, but it's fundamentally devoid of
critical "Semantic Web" essence (DNA).
My acid test for any Semantic Web solution is simply this (using
a Web User Agent or Client):
- go to the profile page of the service
- ask for an RDF representation of my profile (by this I mean
"get me the raw data in structured form")
- attempt to traverse the structured data graph (RDF) that the
service provides via live de-referncable URIs.
Here is the Acid test against my Data Space:
-
My Profile Page (HTML representation
dispatched via an instance of OpenLink Data Spaces)
- Click on the "Linked Data Tab" (HTML representation endowed
with Data Links the link to information resources containing other
structured descriptions of things).