Computers and the species-description backlog.

William L at
Fri May 23 08:50:06 CDT 1997


Stephen P. Rae wrote:
>
> At 01:07 PM 5/21/97 -0700, William L Pratt wrote:
> >More thoughts regarding Dave Chesmore's suggestions.
> >There are areas where computer facilities and the web could be of real...

Steve wrote:

> This may be a difficult project to implement as the list manager and
> referees would be drawn from an already fully committed pool of talent.
> However, if a pilot could be launched, perhaps the merits would be obvious
> to more players!!!!

Sorry to be so long getting back, I was in the field yesterday.

One of the big problems in taxonomy right now is a backlog of
descriptions, which are difficult to publish. There's no problem getting
a single description published in regional journals, and a revision can
find a home with no real trouble, but just try to publish a paper
entitled "Two hundred new species of X". That is harkening back to the
bad old days of the 19th century when people weren't doing science, just
publishing description after description. It is, too.  And for the same
reason: new species are turning up faster than they can be described.
There is real interest in breaking the bottleneck, though there may be
problems with the rules of zoological nomenclature. I believe that the
botanical rules would permit it.

Will wrote:

> >With regard to manuals, why not have a server where want lists could be
> >posted. A group of specialists could undertake a cooperative project to...

Steve wrote:

    > There are many software application programs used to construct
> identification aids (from word processors to DELTA, etc.).  However, such a
> product must take advantage of the PC platform in order to become truly
> widley used and supported in the computer world.  If anything, we need to
> draw in mainstream PC users and systems managers in order to improve the
> quality of the professional software used by taxonomists.  And, thus, the id
> aids created for lab and field use would become more widely used and cheaper
> to create and distribute.
>
> With this in mind, I have a Windows NT 4.0 server running SQL server.  I am
> also interested in hosting applications that help researchers and students
> identify biological resources.  My primary interests are in native vascular
> plants and mosses, however, I could be enticed to experiment in other taxa
> in order to demonstrate that the platform is worth exploring.
>
> In other words, if we can draw together a cooperative group of professionals
> to support the information management requirement, the scientific content,
> and the user interface, then I could be convinced to host the developmental
> materials as a pilot program.  So, how would you want to proceed?  Please
> let me know.
>
> As someone said recently, "Here's a nice fat target!"  I am not asking to be
> shot down, I am calling bluffs and offering a high end technological
> platform to address the taxonomic needs as we professionals see them.
>
> Stephen P. Rae
> www.musci.com

I'm not a closet compuphobe, Steve.  My present machine is a
floorstanding tower, 166 mhtz, 486 cpu, with a 2 gig hard drive. It
started life as a 286 desktop with a 20 meg hard drive, but it has grown
piece by piece until the only original pieces remaining are the floppy
drives. Before too long it will be a Pentium machine. And I write my own
database applications.

I wouldn't use the computer to run identification software, however, for
practical reasons. The screen is hard to read at an angle while working
at the dissecting microscope, and most computer-based keys are presented
as single couplets. Answer one, and the next appears.  But that's not
how taxonomists work.  After a very little experience with a fauna, you
start to "cheat", examining a few characters, then turning not to
couplet one, but to couplet 65, where you know that specimen will go to.
Even when the construction of a key permits it, scrolling is awkward
compared to flipping the pages of the key. And you can't xerox a working
copy of the computer that's no loss if you spill a jar of alcohol on
it.  Also, in the regions where the problem is most severe, computer
facilities are often limited. Out at the field lab, electricity may be
limited.  But if you can download a file to whatever system you have
access to on campus and print it out, then you can take it along to the
field, no problem.  Sure there are laptops, but many Latin American
workers are doing well if they can lay hands on a 286, and often have to
settle for a terminal attached to the (single) university mini. Usually
the available facilities can handle line-drawing illustrations. Thus my
insistance that the final output must be printed-out hardcopy.

There is no question that the key-writing software can be a help in
producing keys. To be a success, however, any program would also have to
be able to take in old-fashioned handwritten keys typed out on a
wordprocessor, because many of the people who have the expertise to
produce them prefer to work that way. Also, most of the software assumes
a fresh start, and much of what's needed would be cobbled together from
existing keys. For an example, there is no key to Generic level for
tropical North American spiders, but there is one for the US and Canada,
and there's one for Panama. A group of Araneologists might go over the
two and combine them, adding in any genera missed. The result would be
an awful mess, but it would work, and it would be manna from heaven for
an Araneologist working in southern Mexico or northern Central America.

My ideal of an ideal key would illustrate each key choice, even if the
characters had been illustrated earlier in they key, to facilitate its
use by non-taxonomists. A couple of examples are : Michener et al, The
bee genera of Northg and Central America (Hymenoptera: Apoidea)
Smithsonian Press, Washington, 1994, and Goulet & Huber, Hymenoptera of
the world: an identification guide, Agriculture Canada, 1993. A good
on-line example is Scott Stockwell's beautiful key to American scorpion
genera http://wrbu.si.edu/www/stockwell/keys/nakey/nakey.html  Such keys
could be handled by any system that could support line art, and would
allow an inerested Biodiversity researcher to make his/her own
identifications.

I'm running on, but one thing just popped into mind: one major problem
for many people will the matter of credit towards tenure. In some
enlightned departments, publications on-line are discounted at about a
two to one exchange rate. Less genrous ones tend to ignore them. And the
preparation of 200 species descriptions weighs less than revising a
genus of ten species. But then, there are a lot of people out there who
have tenure already, or are in non-tenure track poositions.

I think the interest _is_ out there.

Comments, anyone?

Will



--
Dr. William L. Pratt, Curator of Invertebrates
Marjorie Barrick Museum of Natural History
University of Nevada, Las Vegas Box 454009
Las Vegas, NV 89154-4009
(702)895-1403, fax (702)895-3094 e-mail prattw at nevada.edu




More information about the Taxacom mailing list