[MOBY-dev] registering INB services in Canada

[Phillip Lord]

>>>>> "Mark" == Mark Wilkinson <markw at illuminae.com> writes:

  Mark> The immortal words of Phillip Lord are ringing in my head
  Mark> right now... "An ontology is not an ontology unless it is
  Mark> SHARED!" :-)


I'm sure that I stole it from somewhere. 



  Mark> This is a topic that has been discussed (perhaps not on-list?)
  Mark> for many years - going all the way back to the first group
  Mark> (PlaNet) who set up their own registry.  If you "fork" the
  Mark> ontology, then everything breaks, unfortunately.  I don't
  Mark> think we have ever found a workable solution to this
  Mark> situation.  Next-generation MOBY, with RDF/OWL and reasoning
  Mark> and such, may be able to deal with this problem, but MOBY-S is
  Mark> pretty much stuck with one, centralized ontology (which, in
  Mark> our defence, is how the vast majority of ontologies work at
  Mark> this point in Web history).

  Mark> I'm just catching up with my emails after being away for 10
  Mark> days.  I don't see anyone else responding to this so far.  I
  Mark> don't have any suggestions to help you at the moment, but I
  Mark> can raise it at my next lab meeting and perhaps an idea will
  Mark> come up...??  I have a feeling that there isn't a "magic"
  Mark> solution.  MOBY works *because* we all agree on the ontology.
  Mark> If you don't agree on the ontology, then you aren't
  Mark> interoperable... it's pretty much the core principle of the
  Mark> project...


If I may be so bold, I think Natalia has a point, but not necessarily
the right solution. 

INB could easily register their services by knocking out a mapping
between the two ontologies and then they could translate. But this
would be a pain, and error prone. So the question has to be, why have
two ontologies arisen. 

My suspicion is that there is not enough discussion and change within
the existing ontology allowing it to extend to their purposes. After
all, I've read lots of stuff on this list about code enhancement (and
not read far more), but relatively little about ontology extensions. 

Perhaps the INB ontology can be used to integrate into the Canadian
ontology, to the value of both? Of course, I'm only really a lurker on
this mailing list, so my apologies if I am way of the mark. 

BTW, reasoning and RDF/OWL is never going to deal with this problem;
it might make the problem less severe, but it's still going to be an
issue. 

Phil