<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Our entomology collection has had GUID labels for roughly 15 years
now, and we are one of a few institutions that routinely use both
barcode GUIDs (thermal transfer 2D) and non-barcode GUIDs (laserjet
paper labels). As such, I can speak with some confidence about the
following points of contrast between these two systems:<br>
<br>
(1) the thermal-transfer 2D barcode labels are <b>vastly</b> more
expensive and time-consuming to produce than paper labels with a
unique human-readable number, and with a much greater initial cost
of equipment (our Sato printer, the same model used at the AMNH,
cost over 5000 dollars, rolls of plastic cost several hundred, and
it prints seven labels a minute; the HP Laserjet cost 200 dollars,
archival paper is trivially cheap, and it can print around 8,000
labels a minute)<br>
(2) when used for legacy material, they take up 150% of the space of
a normal insect label if they are positioned so as to make the 2D
barcode scannable from above - that means the same number of
specimens takes up 150% more unit trays, drawers, and cabinets, so
the entire cost of physical storage is multiplied by 150%<br>
(3) the scanner will never make a typo unless the label itself is
glitched (there is indeed a very small but finite rate of glitches
during printing, such that occasional barcodes will not scan
properly, mostly 1 bits turning into 0s, and confounding the scan)<br>
(4) they are less prone to abrasion when dry than paper labels, but
far more troublesome when kept in ethanol, to the point where I
would simply never use them for liquid storage (except if taped on
the OUTSIDE)<br>
(5) they are very slightly more difficult to cut, and to put pins
through, or re-pin, but they are better suited for being attached to
microscope slides<br>
<br>
Certain types of routine workflow go faster with barcodes, others go
slower, and others are the same; specifically (1) inventorying an
outgoing loan is much faster whenever the specimens already have
barcodes on them (but not if barcodes are being added to the
specimens as they are being packed, which is more typical) (2)
anything that could normally be done by reading labels on specimens
in situ in the collection takes longer (because the scanning devices
aren't wireless, requiring you to bring the specimens out of the
storage area to the scanner location) (3) they are no better during
legacy label data entry, which is the most common routine activity
(if anything, it actually takes very slightly longer to stop typing,
pick up the scanner, wave it over the barcode, wait for the beep,
put the scanner down, and resume typing, as opposed to simply typing
in a 6-digit number manually)<br>
<br>
The bottom line for me is this: for our purposes, the contrast
between barcodes versus human-readable labels generally favors the <b>latter</b>,
especially regarding <b>points 1 and 2 above</b>. We operate on a
very limited budget, and within a finite storage space, so the few
circumstances where barcodes have the upper hand do not really
justify the expense. Accordingly, out of some 3 million specimens
total, we have around 425,000 specimens with paper GUID labels, and
about 100,000 with barcodes, most of the latter being specimens
loaned to us from other institutions, or on microscope slides.<br>
<br>
To me, personally, for the tasks I do as collection manager, the
human-readable labels are just fine, and I almost NEVER find myself
wishing "Boy, it's too bad this batch of specimens doesn't have
barcodes!". Until and unless one uses both systems side-by-side, the
contrast will not be obvious; it is easy to conceive how barcodes
are superior in theory, but in practice their superiority is highly
context-dependent, and comes with a significant cost. If you find
that student helpers who are typing in numbers are making more than
one error per 10,000 records, then it's easier to find better
student helpers.<br>
<br>
As an aside, I would also caution against over-reliance on GUIDs as
tracking tools, at least in collections like ours; our primary
tracking tool is taxonomy, so any specimen databased to taxon will
be easily found in the collection simply by knowing what it is. If a
curator needs to shift unit trays (or individual specimens) among
drawers or cabinets (due to expansion, consolidation, or changes in
classification), then they can simply do so <b>without incurring
any changes in the database</b> (other than changes in taxon ID,
when needed); collections whose database tracks the physical
location of specimens incur a significant workload any time
specimens are moved around this way (the more fine-scaled the
location is recorded, the more work involved). I'd be surprised if
there are many natural history collections that do NOT organize
material by taxon, so find it hard to imagine when one would ever
need a system that fine-scaled.<br>
<br>
Sincerely,<br>
<pre class="moz-signature" cols="72">--
Doug Yanega Dept. of Entomology Entomology Research Museum
Univ. of California, Riverside, CA 92521-0314 skype: dyanega
phone: (951) 827-4315 (disclaimer: opinions are mine, not UCR's)
<a class="moz-txt-link-freetext" href="http://cache.ucr.edu/~heraty/yanega.html">http://cache.ucr.edu/~heraty/yanega.html</a>
"There are some enterprises in which a careful disorderliness
is the true method" - Herman Melville, Moby Dick, Chap. 82</pre>
</body>
</html>