Hi All --
The working draft section http://www.w3.org/TR/rdf-dawg-uc/#d4.1
suggests that there should be a human-friendly syntax for a future RDF
One can drill down a bit on this, and ask -- "what kind of human?" and
"fixed set of keywords, as in SQL, or open vocabulary?"
The usual "human friendly" assumption is that this means "programmer
friendly", although most humans are not programmers. So, there's an
opportunity here to see if a future query language could be made
friendly for non-programmers. We could call this Human-friendly, with
a capital H.
The usual assumption is also to limit a query language definition to
have a fixed set of keywords. This means that there is little in the
way of application semantics that can be expressed -- all of that is
relegated to an application layer, as in say, Java programs over SQL.
Is there an opportunity at this stage of the working draft to take an
approach that is friendly for the majority of Humans, and that can
capture more application domain semantics than usual?
A possible approach is described by some examples in
As you may see in the examples, a considerable amount of domain
knowledge gets written down in executable form in the process of
defining the queries. Moreover, the vocabulary is open. This allows, amongst other
things, for Human friendly explanations of how query results are
arrived at. (Without this, even in simple examples, deducation chains
over RDF quickly become hard for programmers to follow, let alone
There is a possible administrative level stumbling block though. The
suggested approach uses rules , which
at first sight are out of scope for dawg. (This note is cc-ed to
>From the technical point of view, do folks think the above is a good
approach? If so, is there a way around the possible administrative
level stumbling block?
Internet Business Logic -- online at www.reengineeringllc.com
Reengineering LLC, PO Box 1412, Bristol, CT 06011-1412, USA
Phone 860 583 9677 Mobile 860 830 2085 Fax 860 314 1029