[MOBY-l] lengthy missive on MOBY status after ISMB/I3C

[Phillip Lord]

>>>>> "Mark" == mwilkinson  <mwilkinson at gene.pbi.nrc.ca> writes:

  Mark> Phillip Lord wrote: I met with Carole Gobel at the I3C meeting
  Mark> and we had some Mark> deep
  >> chats about myGrid & MOBY.  I presume this means you went to the
  >> bar and got drunk?

  Mark> no, but we flirted like mad!  

No comment.



  Mark> I don't remember that... very well...
  >> If your ontology, for instance, does not distinguish between a
  >> Blast service which takes a sequence, and one which takes an ID,
  >> you could give the wrong stuff to the wrong service.

  Mark> I don't think this can happen if you are using MOBY corectly.
  Mark> Only services that *can* use your input will be presented to
  Mark> you if you pose the query based on your in-hand object type.
  >> Or you could give an ID to the wrong database, such as giving an
  >> SWISS-PROT id to a TAIR only search facility.

  Mark> again, since service providers register both the object types
  Mark> *and* the namespaces that they deal with, this should never
  Mark> happen either... touch wood!


Okay, I can understand this. 

What if you do not have an object in hand though? One of the notions
within mygrid is to define workflows, which describe what you want to
do with the data. If you want to define a workflow, you should be able
to use metadata about the services, independent of actually having a
running instance of that service. You might, for instance, want the
system to decide when the workflow is enacted, which is the best of
several mirrors to use. You might also want to be able to define a
workflow without data that it is operating over, if, for instance, you
want to apply the same workflow multiple times. 

I think that using metadata, you should be able to get some
guarantee's about the workflow, before you run it. 



  >> I think this is a slightly separate issue from where the data
  >> stored. To me it makes sense to have all the metadata about a
  >> service in one place, rather than have some of it in one place,
  >> and some in another.

  Mark> insofar as it is all part of a single registration, it is all
  Mark> in one place- a couple of joins on the database tells you
  Mark> everything you need to know about the service.
  >> With your system, you would have to ask both the class directory
  >> to get information about BLAST services, and then the instance
  >> repository to get information about the instances, and which ones
  >> could cope with TAIR ID's, which with sequences, and so on.

  Mark> the query is: locateServiceByInput(Object, [Namespace,
  Mark> Servicetype, Provider, maximum_return...])  This query simply
  Mark> returns all services that can use your in-hand object.  You
  Mark> can further filter this by namespace, service-type, etc., but
  Mark> it is a single query on the database.

Okay. 

I think I need to understand moby, and moby central, far better than I
do at the moment. Hopefully this will happen over the next few weeks!

Cheers

Phil