aa at astro.ex.ac.uk
Wed Sep 12 09:45:29 PDT 2007
Rob Seaman wrote:
> Frederic V. Hessman wrote:
>> ...If our original idea of "simple" translations (comparisons of
>> the "equivalences" between vocabularies) has been utterly
>> superceeded by modern semantic software technology (don't worry
>> about the translations and let someone's fancy software do it for
>> you automatically), then we can leave out my
>> <voc:isEquivalentTo>'s entirely and use vanilla SKOS.
> I'd like further info on the "fancy software". There is no
> assumption with VOEvent that the packets are even being parsed by a
> general-purpose XML engine. A packet needs to be able to stand
> alone while simultaneously being able to benefit from modern
> semantic techniques. We certainly can't wait on uncontrolled web
> services responding as part of a workflow.
I've been trying really hard to stay out of this one, but I'd like to
back up Rob's point here. We can't put a call to a web service, to do
some sort of ontological lookup, inside the loop where a piece of
software is trying to decide what a specific bit of data in a VOEvent
packet means. Not in a real time system, it just is going to take too
much time during a decision where seconds could be crucial.
>> If someone can suggest a basic RDF pattern which expresses what
>> you'd like to see and what we wanted to express - a minimum of
>> ontological info - then I'd be happy to produce the complete SKOS
>> (draf) vocabularies.
> I think this is a worthy exercise whether or not this new
> prototype, in turn, ever leads to an official work product of the
> group. Should be a great topic for the VOEvent BoF and/or IVOA
Yes, we're going to have to talk about this, or something like it,
during the VOEvent session.
More information about the semantics