data sharing

[Steve Ginzbarg]

Dear Dave,

I am very glad that the Z39.50 Biology Implementers Group (ZBIG) is
working on developing a Z39.50 profile, dubbed the Darwin Core. It
will provide a common language for data exchange which promises to
make it possible to send a single query over the internet to multiple
biological collection databases each using their own database
structures (flat file, relational, object oriented), data fields and
formats and to have the data returned in real time as if a single
database was being queried. It will also allow participating databases
shared access to taxonomic authority files. I am glad that the
shared data exchange language is being developed and will be
maintained by the biological sciences community. I don't see the
development of the Darwin Core as a substitute for individual
collection websites or sites like those being developed at TAMU which
will provide public outreach to those not connected to the ZBIG network
within the internet.

Visitors to the ZBIG website at
http://chipotle.nhm.ukans.edu/zbig/
under The Darwin Core, The Collection Attributes, and The Taxon
Attributes are encouraged to suggest changes or to help flesh out
definitions by sending you an email. It would be helpful to see the
profile attributes in something more approaching real time (last
modified May 1, 1998) so that the comments I submit will be more
relevant to the current draft. Here is my comment on the "very
preliminary list of attributes":

Suggested changes to the Use Attributes for Biological Collections.

Add an attribute Type Name
Description: The name for which the specimen was designated as a type.
Rational: Attribute 42 Type Status tells us if a specimen is a type
but doesn't tell us for what scientific name it serves as type.
Attribute 16 Scientific Name is the most recent determination of the
specimen. A type specimen may or may not be a type of its accepted
name. It is not always possible to determine from the accepted name
what the Type Name of a type is. If the accepted name is a
nomenclatural synonym of the Type Name, it would be possible to trace
the synonomy back to the basionym which will in this case be the Type
Name.

Ex.: Type Name: Aster sonomensis E. Greene
     Accepted Name: Aster chilensis Nees var. sonomensis (E. Greene) Jepson
     Basionym of Accepted Name: Aster sonomensis E. Greene

If on the other hand the accepted name is a taxonomic synonym of the
Type Name it will be based on a different type and its basionym will
not be the Type Name.

Ex.: Type Name: Aster sonomensis E. Greene
     Accepted Name: Symphyotrichum lentum (E. Greene) Nesom
     Basionym of Accepted Name: Aster lentus E. Greene

Suppose that the holotype of Aster sonomensis E. Greene has been
destroyed and I want to examine the available isotypes in order to
select a lectotype. I would not be able to find out in what
institutions and collections they are housed using a Darwin Core
Profile query based on Scientific Name and Type Status. If I knew
attribute 25 Collector Name and the collector's collection number (for
which no attribute has been assigned in the preliminary list) of the
type I could obtain this information for some names. (This query would
also be useful as it would return not only type specimens which have
been recognized as such in the collections where they are housed as
indicated by attribute 42 Type Status but also unrecognized types.)
However many older descriptions do not give the collection number of
the type. To locate all the recognized types for a taxonomic name a
Type Name attribute will be needed.



-Steve Ginzbarg

Steve Ginzbarg, Assistant Curator
Herbarium (UNA)
Department of Biological Sciences
Box 870345
The University of Alabama
Tuscaloosa, AL 35487-0345

(205) 348-1829, FAX: (205) 348-6460
sginzbar at biology.as.ua.edu