by Taowei Wang

Most of us sat in the SWUI workshop today.  There were a lot of posters/demo and discussions.  Tim demoed the Timulator.. err.. Tabulator, a generic browser of RDF data.  It incorporates SPARQL query on a multiple-select then query interfaces.  The results of queries can be displayed on a calendar or a map (if they are time-tagged, or geo-tagged).  The nifty AJAX makes it sleak, though the interface itself is geeky.

A lot of the discussions and demos are concerned with representing the instance data in a ‘useful’ way.  And since usefulness is in the eyes of beholder, the UI people say “Hey, we can let people choose/create/extend the right UI to represent the data”.  Semantic Web is good that you can take in different ontologies/schemas and use/reuse these vocab to extend your UIs (or forms, fields, etc).  However, to naively believe that customizable UIs in this form (pun intended) can solve the problem is simply, well, naive.

What Semantic Web technologies bring in is allowing the instance data to have a formal backing (ontologies/schemas), and a unified framework for access (RDF), using standardized technologies (SPARQL, etc.).  But thess all have analogies from the traditional database systems.  The data analysts are interested in the data, they don’t care about the formalization or where/how it’s stored.  The analysis of the data does not depend on whether it’s stored in a database, spreadsheet, or triplestore.  The analysis tools don’t care about that, too.

So, the real question is, what are the fundamental differences between Sem Web technologies in representing data and traditional relational databases that requires a new paradigm to analyze data.  Unless we know what that difference is, and how to exploit that difference, the pushing for a new tool or new UI to interact with the data will not be tailored for the Semantic Web data, and cannot be justifable as an innovation.

Unfortunately, the presentations largely went on without addressing these critical issues that can make or break the data access/visualization/analysis research efforts.  Just my humble opinions, not the official stance from the MINDSWAP group
Taowei

This entry was posted on Monday, November 6th, 2006 at 7:40 pm and is filed under Personal Ramblings, Trips, Semantic Web Services. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.