[NHCOLL-L:677] Re: Specimen databases

[Kathie Hodge]

We're using Biota now at CUP.  We work mainly with fungal specimens. 
The software is working well for us, although we've had to invent 
some hacks to get around a few problems.  It has a nice, relational 
structure, which saves typing in the end, and promotes uniformity of 
the data. It does a good job of keeping track of determination 
histories, and can do loan management adequately (but see below).  We 
haven't attempted to put it up on the web, but our colleagues have 
put up a Biota-based fungal inventory-in-progress:
http://ccfb.cornell.edu

We use Filemaker to get around some deficiencies.  This has 
occasionally led us to question why we didn't use Filemaker for the 
whole thing, but when we look at the relational data structure in 
Biota we realize how many complex things we take for granted now, and 
how difficult it would be for us non-programmers to successfully 
implement these things in Filemaker, and then keep up on them through 
software and system upgrades and changing personnel.  So we export 
Biota data to Filemaker to generate labels of several different types 
(for special subcollections, photographs, etc.--which need different 
data on the label), and also use Filemaker for shipping notices. 
Filemaker has much prettier output capabilities, and any modestly 
computer-aware person can learn to customize it.

Some gripes:
Biota currently doesn't track loan histories--when a specimen is out 
you can tell where it is, but once it's back Biota forgets who has 
had it in the past and keeps no thorough historical records of loan 
activity (useful when justifying one's existence).  We haven't 
developed a satisfactory hack for this.

In the data structure, we've developed our own conventions.  "Host" 
ends up as "substrate" for us (sometimes a plant name, sometimes 
"dung").  The synonymy system uses jargon from zoological 
nomenclature ("junior synonym") that we don't see much in mycology, 
and we have had to be careful not to let Biota change all taxon names 
to the "correct" name until someone has actually looked at the 
specimen.  We have also had trouble sometimes trying to export fields 
that are indirectly related, and again we have worked up Filemaker 
hacks for these instances.

Biota's author, Rob Colwell, has been extremely helpful and 
responsive to our questions.  I know he has big plans for the next 
version, so perhaps some of our hacks can be discarded in the near 
future.

regards,
Dr. Kathie T. Hodge
Director, Cornell Plant Pathology Herbarium
Assistant Professor of Mycology
334 Plant Science Bldg.
Cornell University
Ithaca, NY 14853

email: CUP-herbarium at cornell.edu
web: http://ppathw3.cals.cornell.edu/CUPpages/CUP.html
phone: 607-255-1546