[seek-dev] Re: [kepler-dev] seek/kepler conference call notes

Ferdinando Villa ferdinando.villa at uvm.edu
Thu Aug 26 08:39:51 PDT 2004


On Thu, 2004-08-26 at 10:53, Edward A Lee wrote:
> This perspective could be strongly reinforced if "Look Inside"
> were overridden to reveal the WSDL, EML, or R definition, rather
> than the Java code for the actor that implements the web service
> interface, etc.

I'm still not sure I have all points straight, but what would make the
most sense to me is an intermediate language (along the lines suggested
by Bertram) that EML/R/WSDL/other alternative specs are compiled into,
and actors being created from it. At that point (particularly if the
intermediate, semantically rich representation is human readable) "Look
Inside" could show the intermediate representation. Support for more
standards capable of defining an actor's semantics could be added
plugin-wise to either a registration service (e.g. for retrieved EML) or
a filter incorporated in client kepler; the latter would ultimately only
have to deal with the intermediate rep. Those who look inside would
always see the same thing, which could probably be represented
graphically in ways that make sense to a user.

All this, of course, subjected to me having got the point at all :)

ferdinando
> 
> See for example the PythonActor in MoreLibraries/Python.
> 
> In this case, I would agree with Matt that the semantic annotation information
> should be in the WSDL/EML/R file, since that is the actor definition.
> Mechanically, however, the underlying actor would still process these
> semantic annotations by creating an instance of the SemanticAnnotation
> attribute.
> 
> One advantage of this approach is that the SemanticAnnotation attribute
> could handle creating P2P indexes, which means then that these indexes
> would be equally able to refer to Java actors, models in MoML, WSDL
> URLs, EML specifications, or R files.  I.e., the SemanticAnnotation
> attribute would mediate the multiple languages...
> 
> Does this make sense?
> 
> Edward
> 
> 
> 
> ------------
> Edward A. Lee, Professor
> 518 Cory Hall, UC Berkeley, Berkeley, CA 94720
> phone: 510-642-0455, fax: 510-642-2739
> eal at eecs.Berkeley.EDU, http://ptolemy.eecs.berkeley.edu/~eal
-- 
Ferdinando Villa, Ph.D., Associate Research Professor, Ecoinformatics
Gund Institute for Ecological Economics and Dept of Botany, Univ. of Vermont
http://ecoinformatics.uvm.edu




More information about the Seek-dev mailing list