barcodes (the dream)

[christian thompson]

As to Doug's statement:

Only the "barcoding" part, and I, for one, don't consider that
unfortunate at all. It's cheaper and more time-efficient to use
regular old human-readable serial numbers and type the data in by
hand.

One need only to consider history:

Years ago, perhaps before you were around, all stores had cashiers who
punched "old human-readable" price tags into a cash register.  It was slow
and very error prone.  Then the industry got together and came up with bar
codes.

If what you say were true, then that would have never happened.

Personal experience with INBIO material, I can also say, having processed
more than 20,000 specimens, I could never had captured as much data as fast
as I can now with barcodes and databased specimen label information. I can
easily process more than a 1,000 specimens a  day (only real factor is that
one's thumb and fore finger get real sore in pulling out and pass the pinned
specimens under the scanner, etc.).

The other caveat, which must always be stated, is there is a difference
between PROspective barcoding and specimen label information capture and
RETROspective work. This was articulated again back in 1994 in Insect
Collection News (still available on WWW at www.diptera.org\barcodes.htm).

Machine-readable unique identifiers along with databasing the collection
information should be mandatory for all newly acquired specimens. Since 1994
many new data formats for these machine-readable identifiers have come
along, but as most modern scanners are image-based, the software can still
process old formats like code49. So, those institutions, like INBIO, who had
the foresight to barcode and database their specimen label information, now
have far more valueable biological information than the old guard of USNM,
BMNH, MCZ, etc. and it is all WWW accessible!

Sorry, Doug. But you were right about the dream. So long as we as a
community continue to act as individuals doing it our own way, we won't be
able build useful biological information databases like our colleagues in
the other Sciences are doing now.

For others, there are a pair of positive statements in this week's issue of
Nature. The news feature is "All living things, online" and there is a
letter in correspondence on "Taxonomy, at the click of a mouse: Informatics
and taxonomy are working together to achieve more than either could alone."


Oh, well ... It isn't Friday yet?

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 [NB: no terminal "n"]
visit our Diptera site at www.diptera.org


>>> Doug Yanega <dyanega at POP.UCR.EDU> 07/24 12:55 PM >>>
Chris Thompson wrote:

>Sorry, Andrew, but once some of us had hoped that we could have
>standardized acronyms, like the Botanical community has. The purpose was
be
>able to create unique barcodes, where the code would have the alpha
acronym
>for the collection, followed by a sequential number. Such as
INBIO000222445
>or USNM ENT 00012345. Then intelligent programs could parse the barcode,
>then go onto the Internet and retrive the specimen label data from the
>appropriate database.
>
>Unfortunately, that was a dream.

Only the "barcoding" part, and I, for one, don't consider that
unfortunate at all. It's cheaper and more time-efficient to use
regular old human-readable serial numbers and type the data in by
hand. Here at UCR we have both systems available for use (and our
labels all use our official "UCRC ENT" Coden). In side-by-side
trials, it takes at least 2-3 times longer to do any given data entry
or retrieval tasks if one uses the barcode reader as opposed to
simply eyeballing the serial number and entering the digits manually.
Why? Because one has to open the drawer, pull out the unit tray, pull
the specimen out of the tray, wave the scanner at the label, wait for
the beep, then put the specimen back. If just eyeballing, you don't
even have to open the drawer. Not only does it take substantially
longer, you run the risk of damaging the specimen in the process of
removing it, scanning it, and putting it back. The equipment for
barcode reading is an expensive investment, requires maintenance,
requires that anyone doing data entry has to be sitting where the
equipment is located (meaning only one person at a time can use it -
if we had more helpers, it would be even more efficient NOT to use
the scanner), and having the labels printed ain't cheap (compared to
serial number labels, which anyone can print on any old printer).

The one marginal advantage to barcode scanning is a lower error rate,
but - frankly - in my experience, I've found that it's harder to find
a student laborer who can remove and replace insects in a tray
without breaking off legs and antennae than it is to find one that
can read and type six-digit numbers without error. So, we have a
barcode reader but we don't use it except for our slide collection,
because time efficiency is our highest priority. I'd rather have
someone enter 1000 specimens per day (and get one number wrong per
1000) than have someone enter only 400 specimens per day flawlessly.
After all, *either way* there is a data-checking step, during which
mistakes are corrected, so the end result of the former approach is
comparable, at less than half the cost per record.

The idea of data sharing for specimen records is still alive, and I
believe it will come to pass. It just won't (or shouldn't) involve
barcodes. None of us is so flush with funding and labor that we can
afford to waste it that way.

My two cents,
--

Doug Yanega        Dept. of Entomology         Entomology Research Museum
Univ. of California - Riverside, Riverside, CA 92521
phone: (909) 787-4315 (standard disclaimer: opinions are mine, not UCR's)
            http://entmuseum9.ucr.edu/staff/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82