Open Data Access and Web 2.0 have a very strange relationship
that continues to blur the lines of demarcation between where Web
2.0 ends and where Web.Next (i.e Web 3.0, Semantic/Data Web, Web of
Databases etc.) starts. But before I proceed, let me attempt to
define Web 2.0 one more time:
A phase in the evolution web
usage patterns that emphasizes Web Services based interaction
between “Web Users” and “Points of Web Presence” over traditional
“Web Users” and “Web Sites” based interaction. Basically, a
transition from visual site interaction to presence based
interaction.
BTW - Dare Obasanjo also commented about Web usage patterns in
his post titled:
The Two Webs. Where he concluded that we had a dichotomy along
the lines of: HTTP-for-APIs (2.0) and HTTP-for-Browsers (1.0).
Which Jon Udell
evolved into: HTTP-Services-Web and HTTP-Intereactive-Web during
our recent podcast
conversation.
With definitions in place, I will resume my quest to unveil the
aforementioned Web 2.0 Data Access Conundrum:
- Emphasis on XML's prowess in the realms of Data and Protocol
Modeling alongside Data Representation. Especially as SOAP or REST
styles of Web Services and various XML formats (RSS
0.92/1.0/1.1/2.0, Atom, OPML, OCS etc.) collectively define the Web
2.0 infrastructure landscape
- Where a modicum of Data Access appreciation and comprehension
does exist it is inherently compromised by business models that
mandate some form of “Walled Gardens” and “Data Silos”
- Mash-ups are a response to said “Walled Gardens” and “Data
Silos” . Mash-ups by definition imply combining things that were
not built for recombination.
As you can see from the above, Open Data access isn't genuinely
compatible with Web 2.0.
We can also look at the same issue by way of the popular M-V-C
(Model View Controller) pattern. Web 2.0 is all about the “V” and
“C” with a modicum of “M” at best (data access, open data access,
and flexible open data access are completely separate things). The
“C” items represent application logic exposed by SOAP or REST style
web services etc. I'll return to this later in this post.
What about Social Networking you must be thinking? Isn't this a
Web 2.0 manifestation? Not at all (IMHO). The Web was developed /
invented by Tim Berners-Lee to leverage the “Network Effects”
potential of the Internet for connecting People and Data.
Social Networking on the other hand, is simply one of several ways
by which construct network connections. I am sure we all accept the
fact that connections are built for many other reasons beyond
social interaction. That said, we also know that through social
interactions we actually develop some of our most valuable
relationships (we are social creatures after-all).
The Web 2.0 Open Data Access impedance reality is ultimately
going to be the greatest piece of tutorial and usecase material for
the Semantic Web. I take this position because it is human nature
to seek Freedom (in unadulterated form) which implies the
following:
- Access Data from a myriad of data sources (irrespective of
structural differences at the database level)
- Mesh (not Mash) data in new and interesting ways
- Share the meshed data with as many relevant people as possible
for social, professional, political, religious, and other
reasons
- Construct valuable networks based on data oriented
connections
Web 2.0 by definition and use case scenarios is inherently
incompatible with the above due to the lack of Flexible and Open
Data Access.
If we take the definition of Web 2.0 (above) and rework it with
an appreciation Flexible and Open Data Access you would arrive at
something like this:
A phase in the evolution of the
web that emphasizes interaction between “Web Users” and “Web Data”
facilitated by Web Services based APIs and an Open & Flexible
Data Access Model “.
In more succinct form:
A pervasive network of people
connected by data or data connected by people.
Returning to M-V-C and looking at the definition
above, you now have a complete of ”M“ which is enigmatic in Web 2.0
and the essence of the Semantic Web (Data and Context).
To make all of this possible a palatable Data Model is required.
The model of choice is the Graph based RDF Data Model - not to be
mistaken for the RDF/XML serialization which is just that, a data
serialization that conforms to the aforementioned RDF data
model.
The Enterprise Challenge
Web 2.0 cannot and will not make valuable inroads into the the
enterprise because enterprises live and die by their ability to
exploit data. Weblogs, Wikis, Shared Bookmarking Systems, and other
Web 2.0 distributed collaborative applications profiles are only
valuable if the data is available to the enterprise for meshing
(not mashing).
A good example of how enterprises will exploit data by
leveraging networks of people and data (social networks in this
case) is shown in this nice presentation by Accenture's Institute
for High Performance Business titled: Visualizing
Organizational Change.
Web 2.0 commentators (for the most part) continue to ponder the
use of Web 2.0 within the enterprise while forgetting the
congruency between enterprise agility and exploitation of people
& data networks (The very issue emphasized in this original
Web vision
document by Tim Berners-Lee). Even worse, they remain
challenged or spooked by the Semantic Web vision because they do
not understand that Web 2.0 is fundamentally a Semantic Web
precursor due to Open Data Access challenges. Web 2.0 is one of the
greatest demonstrations of why we need the Semantic Web at the
current time.
Finally, juxtapose the items below and you may even get a
clearer view of what I am an attempting to convey about the virtues
of Open Data Access and the inflective role it plays as we move
beyond Web 2.0:
Information
Management Proposal - Tim Berners-Lee
Visualizing
Organizational Change -
Accenture Institute of High Performance Business