Specimen databases

[christian thompson]

Given the comments by Matt Yoder and others I feel I should re-iterate my
bottomline.

There is a significant difference between handling data for your own
research and that for a large project, like ALAS, or organization.

For your own research I strongly feel you probably are better off using an
commerical database application like MS Acess, FileMakerPro, etc. to build
your own, because you will get to know your data better, you will build a
tool that isn't large, complex, etc., and fits your needs precisely, etc.
That is what we try to teach the students.  And we also teach them community
standards. For specimen data, look at the ASC model, just take what you need
of it, etc. And in some cases, BIOTA might be best for you.

On the large scale, I do recommend solutions like BIOTA.  But using these
for research has more problems than those Michael Chamberland originally
noted. I am working at the Smithsonian and started back in those days where
SELGEM, a COBOL custom product, was going to do everything for us. In the 30
years since, there has been a constant parade of new museum database
programs to do everything for us.  Hundreds of thousands of dollars have
gone into them, and NSF is still throwing more money at the problem. For us
who are expected to enumerate the World Biota, at best these database
projects only work on what is known and at worst they have wasted precious
funding that could have been spent on exploration and description of new
species.

In summary, I feel the focus should return the grass roots. I will capture
the critical data associated with existing specimens in a standard digital
format when I revise groups within which the new species are being
described, etc. But spending dollars to inventory the National Entomological
Collections at a specimen level today is the old "garbage in, garbage out."
Only a specialist who is revising a taxon can really confirm the
identification in most groups (Insecta, etc.) and also it is only the
specialist who can interpret the old historical data labels associated with
many of the specimens in our current collections.

The only real positive programs I have seen are those like ALAS (La Selva
Arthropoda Survey) and INBIO in Costa Rica, where they are processing NEW
material which is barcoded and the associated data can be easily export to
those researchers who want it, etc. For my group of flies, INBIO has
accumulated thousands of well-coded specimens, precise localities, dates,
etc., where as the existing collections have only a few hundred flower flies
from Costa Rica. Our knowledge of what these flies do, when and where they
do it, etc., will come from the ALAS and INBIO projects, not the
Smithsonian.  Not that the Smithsonian collection isn't important, its
importance is the vouchers that link the previously published information
with the new. Unfortunately, for many groups this is very little and mainly
consists of nomenclatural types.



F. Christian Thompson
Systematic Entomology Lab., ARS, USDA
Smithsonian Institution
Washington, D. C. 20560-0169
(202) 382-1800 voice
(202) 786-9422 FAX
cthompso at sel.barc.usda.gov
visit our Diptera site at www.diptera.org