Increasingly, I am encountering commentary from the ReadWriteWeb
data space that highlights critical problems solved by a Linked Data Web. Unfortunately, most of the
time, there is a disconnect between the problem and the solution.
By this I mean: technology in the Semantic Web realm isn't seen as
the solution.
A while back, I wrote a post titled:Why we need Linked Data. The aim of the
post was to bring attention to the implications of exponential growth of User Generated
Content (typically, semi-structured and unstructured data) on
the Web. The growth in question is occurring within a fixed data
& information processing timeframe (i.e. there will always be
24hrs in a day), which sets the stage for Information Overload as
expressed in a recent post from ReadWriteWeb titled: Visualizing Social Media Fatigue.
The emerging "Web of Linked Data" augments the current "Web of
Linked Documents", by providing a structured data corpus
partitioned by containers I prefer to call: Data Spaces. These
spaces enable Linked Data aware solutions to deliver immense value
such as, complex data graph traversal, starting from document
beachheads, that expose relevant data within a faction of the time
it would take to achieve the same thing using traditional document
web methods such as full text search patterns, scraping, and
mashing etc.
Remember, our DNA based data & information system far
exceeds that of any inorganic system when it comes to reasoning,
but it remains immensely incapable of accurately and efficiently
processing huge volumes of data & information -- irrespective
of data model.
The Idea behind the Semantic Web has always been about an
evolution of the Web into a structured data collective comprised of
interlinked Data items and Data Containers (Data Spaces). Of course
we can argue forever about the Semantics of the solution
(ironically), but we can't shirk away from the impending challenges
that "Information Overload" is about to unleash on our limited
processing time and capabilities.
For those looking for a so called "killer application" for the
Semantic Web, I would urge you to align this quest with the "Killer
Problem" of our times, because when you do so you will that all
routes lead to: Linked Data that leverages existing Web
Architecture.
Once you understand the problem, you will hopefully understand
that we all need some kind of "Data Junction Box" that provides a
"Data Access Focal Point" for all of the data we splatter across
the net as we sign up for the next greatest and latest Web X.X
hosted service, or as we work on a daily basis with a variety of
tools within enterprise Intranets.
BTW - these "Data Junction Boxes" will also need to be
unobtrusively bound to our individual Identities.