Collaborative articulation of how abstraction and language is employed in the computational manifestation of numbers -- including analysis of the role of syntax, semantics, and meaning in the specification and use of software interfaces.

Recent Items
 
Searching for the Semantic Web
 
Metadata for the desktop
 
A Java int is not an Integer; that's the point!
 
Agile Scope Creep and How to Detect It
 
All that's needed isn't written down; how much more might be enough?
 
Abstract or concrete? Maybe both.
 
A mesh of agreements, I am, I am
 
Service interface experiences
 
Compiling Num.java 0.1x
 
Deja Double Vu

Archives
2004-05-30
2004-06-06
2004-06-13
2004-06-27
2004-08-22
2004-08-29
2004-10-03
2004-12-05
2005-01-23
2005-06-19
2005-08-14
2005-10-09
2006-01-01
2006-02-05

Searching for the Semantic Web

ACM News Service: Presenting Semantically Enabled Knowledge Technologies. I don't spend much time thinking rapturously about the advent of the semantic web. Rather, I find the hype about it to be at best misleading, and I do agree that there is a great deal to learn about the prospect of "semantic annotation" and the conditions under which that is actually meaningful and useful.

In that respect, the SEKT project has some key ingredients of good science by following this trajectory:

1. basic research in ontology and metadata management, including new techniques for annotation
2. automated extraction of metadata from natural language text
3. discovery and delivery of only relevant information to users
4. confirmation of tools in that spectrum in large-scale case studies in three language communities represented by Britain, Spain, and Germany.

The 2004-10-20 IST Results feature reports that the full range of activities is included in reports to be made to the November International Semantic Web 2004 conference in Hiroshima. The project was begun on January 1, 2004, and will continue for 36 months. I suppose that does imply a certain expectation of rapid success. It will be valuable to watch.

From 2004-10-21: Another old draft article. I haven't checked to see how well the project did, or whether they fall into syntax/semantic/ontological confusions as many of these efforts seem to do. Again, the treatment of What Computers Know and the progression from that point may be helpful here.

posted by orcmid at 2/08/2006 03:20:00 PM

Metadata for the desktop

2004: Metadata for the desktop.  This Edd Dumbill article provides worked examples and demonstration of the value of metadata (via RDF) on the desktop.  It is rather interesting that it is automatically presumed that RDF expresses semantics, with little demonstration of how that occurs, except for some hand-waving around the use of <dc:creator>.

From 2004-07-04: This was in the backlog of drafts. There are more-recent topics concerning where the semantics are to be found in Dublin Core and other entity tag sets in XML and especially RDF. This is going to be part of the development around What Computers Know as the progression is expanded.

posted by orcmid at 2/08/2006 03:15:00 PM