danny.ayers at gmail.com
Tue Jul 12 10:42:33 PDT 2005
An idea I'd appreciate a little sanity checking on -
* Microformats  are a way of making data explicit in human-readable XHTML
* Fitnesse  is a software testing tool, based on describing test
data on a Wiki
* DOAP  is Description of a Project, an RDF vocabulary for machine
readable descriptions of software projects
So why not construct a little vocabulary to extend what's already in
DOAP to describe tests? The vocabulary could be expressed as a
microformat (so it could appear on Wikis a la Fitnesse) but would also
be available for arbitrary machine processing via RDF.
I've already played around with mapping between DOAP expressed as
RDF/XML and as XHTML (hDOAP, see ). It works quite nicely, the
XHTML being human-presentable, with a GRDDL-disambiguated  XSLT to
transform it into directly machine-processable RDF/XML.
The RDF in  provides a common data model, meaning that data from
one source is easy to integrate with that from another, such as
personal profiles (FOAF), event information (iCal/RDF), or blog posts
(RSS), or just about anything else. The triplestore offers a very
agile approach to data storage and query (via SPARQL ), you don't
have to worry about building schemas in advance.
I got to wondering about this as although I'm a big fan of agile
technologies, I've still managed to let the thing I'm working  on
grow fairly sizeable without having any proper tests in place. Because
I've got an RDF store in the background, it would make sense for me to
use this for managing tests. On the one hand it'll be easy enough to
set up Wiki-like content management, on the other it's relatively
straightforward to build declarative, rule-based processing on RDF
(there's RDF/OWL inference and rule engines available).
I'm on the thin ice of my knowledge in many places here, so I thought
I'd run this by Folks That Know About These Things before going any
More information about the microformats-discuss