Report prepared for the Experts Meeting Towards the Implementation of a Global Invasive Species
Information Network (GISIN), 6-8 April, 2004. Baltimore, Maryland, USA.
I - ii
8/30/2004
Consensus on mutually-useful controlled vocabularies (also referred to as “thesauri”, or
“name authorities”) will be essential to developing interoperable information systems.
Timely access to data is especially critical in countering newly appeared invasive
species. Therefore, a successful network includes incentives for sharing data. These
include increased professional recognition for developers of databases, metadata
strategies that help assure that providers of data are properly credited, tools to make
preparation of standardized data and metadata easier and more automatic. Electronic
publishing offers opportunities to provide peer review and recognition of intellectual
achievements beyond the traditional journal papers, and should be encouraged as a
means to recognize and validate the efforts required to share invasive species
information with the wider community.
A Network of Nodes
The simplest configuration of a hemispheric invasive species information network hub is as
a nexus of connected “nodes”, each contributing comparable information derived from its
domain of geographic, taxonomic, and organizational expertise. Interlocking projects
developing under GISP, NABIN, IABIN, NBII, IUCN-ISSG, and many other partners have
begun to establish a network of regional and national invasive species nodes in the
Americas, and a few elsewhere, that provide one model for a broader network. Each node
agrees to catalog a minimum set of simple but widely applicable data types, and to express
them on a website in a format (XML-based) readily accessible to the other nodes. Standards
for the minimum set are still evolving, but include content on experts, metadata on data
holdings, catalogs of invasive species management projects, lists of organizations
concerned with invasive species and their species of concern, and actual locations and
dates where particular invasive species have been documented to occur (sightings and
specimens). Other kinds of resources, for example bibliographies and fact sheets, have
been added at some nodes.
Recommendations for principles and minimum capabilities for establishing nodes are given
in Appendix I (in prep). However, and effective and scalable network for invasive species
practitioners cannot be exclusive. Any machine with an always-online web server should be
able to act as a node, and tools and standards should permit a reasonably accomplished
system operators and database manager to operate a bare-bones node with limited
additional training. The linking technology should be platform-independent whenever
feasible, and the information posted should be discoverable through standard search
engines as well as specialized portals. Prototypes have outlined the information and training
needs, but the workshop participants recommend that GISIN (probably with partners, such
as GISP and GBIF) sponsor the development of full specifications.
Interoperability – XML, RDF, and the Semantic Web
Most of the data resources currently maintained on-line by workshop participants and the
invasive community at large were constructed using commercial software (notably relational
database management systems and geographic information systems). Nevertheless, the
workshop recommends that open-source and public standards, such as those under
development by the World-Wide Web consortium, be promoted for posting, harvesting, and
searching public invasive species data. While there are other attractive approaches, the
workshop participants believe that “semantic web” technologies (specifically XML, RDF,
SOAP, and related technologies) should and will be widely adopted as a framework for
 |
 |
 |