Maintaining datbases

[Rob Colwell]

Doug Yanega wrote, regarding my discussion of specimen det updating
using barcodes:

>I've noted in the past that the protocol in step 1 [including a barcode
>on each specimen during the preparation process] is not the same as
>the protocol for [snip] insect specimens in a museum. A pinned ALAS
>insect specimen apparently has a barcode label there on top where one
>can read it or scan it easily. A museum specimen will not, unless one
>has gone to the trouble of removing the 2-5 labels already there,
>putting the bar code label on top, and then replacing the original
>labels (running risks of both losing or damaging some of the original
>labels as well as damaging the specimen). Has anyone on this
>newsgroup actually taken already-identified pinned museum
>specimens and added barcodes to them? [Snip] An upside-down code
>label is a problem even if NOT bar-coded, for this same reason, and
>also because if one then gets a request for, say, "specimen 00003478"
>there is no good way to make a quick visual search.

Project ALAS and everyone else I know that uses barcodes for pinned
material puts them upside down on the bottom of the label stack, for
new or old specimens. (In fact, we have many pre-ALAS,  older specimens
in the ALAS working collection at La Selva, to which barcodes were
added as they were accessioned.) Yes, putting the label on the bottom of
the stack means you have to handle each specimen to scan them into a
database.

But consider these facts: for newly prepared material, you have to handle
each specimen in the storage process anyway; in the protocol I outlined
in my last message, the records are created before adding the barcode to
the specimen, but alternatively, you could just pass it under the scanner
in the process of placing it in the unit tray for storage. When a specialist
visits your collection and sorts specimens into groups for identification
or re-identification, they must be handled anyway; just scan them as
they are taken out of storage for study (to pull up the records) and again
as they are replaced in the collection (to update the records). To make a
loan, you have to handle each specimen anyway; just scan them as they
go into the shipping tray/box, or back into storage for a return.

In short, all these key tasks involve removing specimens, one by one,
from storage and handling them anyway. If someone wants to see
"specimen 00003478," let's say based on something in the database
record, there are two ways a database application can help. First, as
mentioned by Norm Johnson, the Specimen record can include information
on where to find the specimen. (The Biota database management app has
a field called "Storage," and other re-namable fields could be used if
more than one is needed.)

Second, consider the following preparation protocol for new material: (1)
line up a series of N newly-pinned specimens from a single collection-
event. (2) Create Specimen records for the N specimens, linked to a
record for that Collection-Event, using the first and last barcode in the
sequence (if your application can do this, as Biota does), or one-by-one.
(3) Print out Locality labels directly from the database, with the unique
barcode number *on each locality label.* (Biota does this as an option,
for pin labels as well as slide, vial, and herbarium labels.) (4) Place the
locality label in the standard, top position, then the corresponding
barcode upside down beneath it. Now you can scan a unit tray visually,
and find a specimen by specimen code (number).

Three caveats: (1) This is no help for old specimens, which of course I
agree should never have old labels removed. (2) Unless matters have
changed recently, there is no consensus that the laser printed pin labels
are sufficiently durable for the ages. (In fact, ALAS does not use them, in
spite of Biota's capability to produce them; we do print and apply slide
labels with specimen numbers using precisely the above protocol,
however.) But this will surely change. (3) You have to take the barcode
off and put it back on again to add a determination (or other) additional
label). Since the barcodes in use are tough polyester, they survive this
well physically, but the chance of mixing them up among specimens does
exist.

A note on barcodes and manually entered specimen codes (numbers):
Barcode readers are simply an alternative, parallel input device to the
keyboard. Specimen database management systems (Biota included) also
accept *manual* keyboard entry for any field that can accept a barcode;
barcodes are in no sense required.

======================================================
Robert K. Colwell
Department of Ecology and Evolutionary Biology
University of Connecticut, U-42
Storrs, CT 06269-3042
=====================
E-mail colwell at uconnvm.uconn.edu
***PLEASE NOTE THE NEW AREA CODE for northern. CT: 860***
Voice (860) 486-4395
Fax (860) 486-3790