INQLE 0.1.7

[David]

This version of inqle adds these features:
* wizards actually work
* new sampling algorithm starts with user-selected or randomly-
selected URI subject class. Then assembles attributes as arcs from
each instance of this class. selects random 500 records.
* Experimental attributes are assembled from arcs. So inqle can
traverse multiple steps along the RDF graph, from each subject
instance (row) to each associated value (cell).
* added nifty support for qnames = abbreviated URIs
* lookup of subject and predicates from Central INQLE Server works!
I experimented w/ MySQL and PostgreSQL databases, but found that Jena
SDB & by extension INQLE perform MUCH better with embedded H2 database
than the external ones. I.e. it di not perform at all using
PostgreSQL or MySQL, and it is adequate so far. So when load
increases significnatly, we might have to find a clusterable solution.

INQLE 0.2 is nigh! For this release (by 12/31/2008) plan to reengineer
& perfect the data importer wizard, the the point that I am proud to
show it in a screencast & in demos.