Report prepared for the Experts Meeting Towards the Implementation of a Global Invasive Species
Information Network (GISIN), 6-8 April, 2004. Baltimore, Maryland, USA.
Page 38
8/30/2004
of result from the activity of a computer
somewhere on the Internet. But when
referenced with capitalized name it is a Web
Service whose query is wrapped up in a
software envelope that uses a standard
language or Simple Object Access Protocol
(SOAP), and which conforms to certain
other standards about how the service
behaves.
Web Service protocols also offer a method
of registering services in a registry, which is
thus a sort of database of databases. The
registry itself functions as a Web Service.
A web service is any facility on the
Web that can answer a question or
provide some sort of result from the
activity of a computer somewhere on
the Internet.
A Web Services Scenario
Consider the scenario in which a researcher
asks: “Find the potential plant invaders of
the State of Karnataka.” The steps to be
taken in addressing this include: 1)
identifying where Karnataka is (it’s a State
in India); and 2) establishing potential
criteria for ‘invasion’ (e.g. what are the
invasive species in adjacent states?).
Representing Step 2 to a computer is quite
difficult. The establishment of criteria for
invasion is a subject of a lot of research in
the domain of the Semantic Web and
ontology in relation to invasive species. Jim
Quinn (University of California / Davis) has
recently received funding to work with Jim
Hendler (University of Maryland), one of the
inventors of the Semantic Web, to address
the question of what it might mean to ask a
database for criteria for invasion.
Once the criteria for invasion have been
identified, one must determine what data
meet those criteria. So to complete Step 3
one might ask: “Find all of the GISIN
databases whose geographic coverage
intersects States that are adjacent to
Karnataka.” The data obtained in Steps 1
and 3 may not be represented by GISIN
databases, but rather usually comes from
another service such as a Digital Gazetteer
(DG).
When databases have been identified, they
must be queried and compared. In the case
of taxonomic names, in the event that two or
more databases treat names differently, the
names must be resolved into taxonomic
concepts because there is no guarantee
that because two databases refer to the
same scientific name they are referring to
the same species. The Global Biodiversity
Information Facility (GBIF) and a few large
projects in the United States have recently
begun to address this particular issue.
Web Services Architecture
The Web Services Architecture diagram
below represents the situation where an
A scenario and some subscenarios
•
Find the potential plant invaders of the state of
Karnataka
–
Find out where Karnataka is
–
Establish potential criteria for invasion (e.g.
“species is in an adjacent state”)
–
Find all GISIN databases whose geographic
coverage intersects states adjacent to
Karnataka
–
Query all
–
Integrate data (e.g. resolve taxonomic names
into taxonomic concepts and compare)
–
Present answer
What
•
What is a web service?
–
Any facility on the web that can answer a
query or provide the result of a computer
activity.
•
What is a Web Service?
–
A web service in which the query is wrapped
in an envelope using a standard envelope
language (SOAP)
–
A method of registering Web Services
–
A method of querying registries for registered
Web Services. This method is itself a Web
Service!
 |
 |
 |