RE: FWD [Work in Progress on Semantics for Web Services (Advance Notice)]

Hi, Josh,

Thanks very much for your comments. I used OGC's WMS in ArcIMS project and
understand OGC's standards in general. Unfortuantely people just regard such
kind of application as RESTful Web services which is definitely separated
from WSDL Web services. I had such ideas before but now I think if we
transform WSDL Web services into document-based exchange style as proposed
in my paper, we can then associate both together and eventually we can build
semanitc Web services just via HTTP/POST other than WSDL/SOAP. I am not sure
how W3C will consider such a suggestion although I already submitted my
paper to certain staff. However I think at least you can understand those
questions and problems I mentioned in those emails to this group when I
develop GIS functional Web services to process geospatial data and for
geospatial analysis (I used ArcGIS Server functions but the vector data set
is defined myself since I think GML is redundant in many cases in such
testing projects)

Regards,

Xuan

-----Original Message-----
From: Josh@oklieb
To: David Martin
Cc: Shi, Xuan; 'Carine Bournez '; 'Battle, Steven Andrew ';
'public-sws-ig@w3.org '
Sent: 9/1/05 3:46 PM
Subject: Re: FWD [Work in Progress on Semantics for Web Services (Advance
Notice)]

Greetings,

It has certainly been the experience of OGC work in Geospatial Web
services that neither WSDL nor OWL-S (nor OGC Capabiliities) is yet a
full and actionable description of services by itself. WSDL is all about
syntax and says nothing explicitly about either semantics or content
coupling. OWL-S still (at least from my perspective) lacks the ability
to describe content coupling. Yet I do appreciate the approach of OWL-S
not to throw away the useful contributions of WSDL to Web services
syntax description in the process (so to speak) of going beyond it.

This does, however, leave me with a sneaking suspicion that no service
description approach is perfect.  An important adjunct to any such
scheme will be an explicit description of its knowledge preconditions –
in what knowledge community will the descriptions be useful and
understandable. So far we've been better at serving "ourselves" than at
serving "others" whether in the GIS community or elsewhere.

Josh Lieberman


Joshua Lieberman, Ph.D.

Principal, Traverse Technologies Inc.

mailto:jlieberman@traversetechnologies.com

tel +1 (617) 395-7766

fax +1 (775) 514-6621

Received on Friday, 2 September 2005 02:59:14 UTC