Hi all, since convincing approaches for interacting with very large (web scale) graph-based dataset are still lacking, I would like to start a discussion on this topic here.
It seems to me that currently single-resource/instance centric data browsers like Tabulator, Zitgist or Disco still live in the mindset of the current document-centric web, where users primarily look at one resource (person, project, company, etc.) at a time. Hence, they compete for usefulness with the document centric web itself, a battle that's hard to win. If I want to publish data about my friends, my projects, my publications, etc. then I publish it on my homepage and provide links to their web pages. Why use RDF for that? In my opinion Semantic Web technologies help solving problems of data-aggregation, and that's an area which is closely related to today's business intelligence tools and multi-dimensional databases. That is resource-list centric. But current resource-list centric approaches like faceted browsers (Longwell/Exhibit) lack a browsing capability and hence don't scale for very large datasets from a UI perspective. Therefore, I'd like to discuss the idea of browseable facets, which enables browse lists of resources (facets). I made a small demo available here: http://www4.wiwiss.fu-berlin.de:8800/uberblic/pivot/ please note it's just a prototype and quite buggy. Select a type from the tagcloud on the left-hand side to start with and the domain of facets you want to look at from the tagcloud on the right hand side. Then hit the "explore" link to pivot the view around a facet. I'm looking forward to hearing your opinion. Cheers, Georgi -- Georgi Kobilarov www.georgikobilarov.com _______________________________________________ General mailing list [email protected] http://simile.mit.edu/mailman/listinfo/general
