« This Week's Semantic Web | Main | Jason Calacanis takes his stab at Web 3.0 »
3 October 2007
Curmudgeon or realist ?
Posted by Paul Miller at October 3, 2007 11:18 PM
Jennifer Zaino writes today in Jupiter's new Semantic Web channel, revisiting some of Alex Iskold's thoughts on the Semantic Web.
Alex, you'll remember, was the author of two recent pieces for Read/Write Web, both of which I commented upon here at the time.
The title to Jennifer's piece describes Alex as a 'semantic curmudgeon', presumably because she reports him to believe that;
“It [the Semantic Web] lacks memory and is not iterative in nature.
Its ultimate goal is to deliver perfect answers, which are unattainable.
It is technologically impractical to achieve.”
Curmudgeonly, or a pragmatically realistic response to years of hype and failed promise?
The noble vision of the Semantic Web is just that; a noble - and long term - vision. The years of seeking perfect answers to perfectly formed questions - a practice of which too many in the Semantic Web community are guilty - have not helped to move us nearly as far forward as we should have come. The over-reliance upon complex and impractically all-encompassing ontologies have bogged us down, and invited ridicule. The close alignment to machine intelligence and inference has drawn many of the brightest to extremely long term plays, diverting them away from the real and immediate advances that could be offered on the ground, now.
However (and I believe that Alex would agree, despite his argument's reliance upon the 'top down' and 'bottom up' distinction with which I disagreed when he wrote it), semantic technologies such as those that the Semantic Web will require are immensely powerful - today. RDF and its kin offer a compelling and flexible means to deal with large bodies of heterogeneous distributed data - today. Approached pragmatically and with intelligence, these technologies and approaches move us beyond a mere Web of Data, past John Battelle's 'Database of Intentions' to a (distributed and evolving) Web of Intentions capable of harnessing relationships between items, people, contexts, and goals. And they do it without globe-spanning ontologies, without endless attempts to count the angels on the head of this particular pin, without widespread mandating of XML editors with which the world's secretaries, teachers, and administrators will lovingly craft the RDF that describes their letters, their learning objects, and their spreadsheets.
The Semantic Web has much to offer us, today, long before it comes to fruition.
[And you can tell that this particular site is new; it consistently reports in both Safari and Firefox that Jennifer's article consists of two pages; both of which appear from my clicking to be identical, bar their url. Browser error, teething trouble, or the missing part of the story that would have changed my interpretation of what it was trying to say?
To add insult to injury, an attempt to tell Jupiter privately that this was the case resulted in a 'Bad Referrer' error from their FormMail application, which apparently doesn't know where semanticweb.com is... Hence this public description of the problem, in case they're watching...]
Technorati Tags: Semantic Web, Talis, Web 3.0
Trackback Pings
TrackBack URL for this entry:
http://blogs.talis.com/mt/mt-tb.r280.cgi/1017
