uBio & Free taxonomic information

[Richard Pyle]

My reluctance to jump into this fray is exceeded only by my distain for
miscommunication leading to highly unproductive and skewed dialog on this
list.  It may well be that I am the only one here who is missing the point;
but in case I'm not, I'd like to take a shot at clearing up the perceptions
of others who, from my perspective, seem to be the ones who are missing the
point.

I'll start with the soapbox rant:

If the future of taxonomy hinges on the value of lists of text-string names,
then our profession is surely doomed.  Thankfully, this is not the case. The
REAL value of taxonomy has to do with organisms, not the names that we apply
to them.  The science of taxonomy involves understanding the scope of
biodiversity.  The names we apply to taxa are simply a tool to allow us to
communicate with each other, and at some gross level, apply some
quantitative assessments about biodiversity. It's not the names themselves
that have value, it is the interpretation of the biological MEANING of those
names that taxonomists should be paid highly for.

As anyone who understands the difference between a "taxon name" and a "taxon
concept" knows, our profession has long been guilty of using the former as
the sole representation the latter.  I use the word "guilty", because this
practice has put us in a situation where extracting information from
historical literature into a biologically meaningful context is an
exceedingly arduous task. I won't delve into the intricacies of this issue
here, except to underscore that the value of taxonomic work is mostly about
interpreting taxon concepts; not creating indexes of names.

I wholeheartedly agree with those who believe that a large part of the
reason taxonomy as a profession is now in such dire straights is due to the
generally uncooperative and "proprietary" sociology of the taxonomists
themselves.  We spend so much time in-fighting with one another, that it's
no wonder we haven't been as successful as other sciences (physics,
astronomy, etc.) in landing the really huge money.  Imagine what we could do
with the scale of funding that has gone into manned space exploration, or a
super collider, or an astronomical observatory. Is it because the questions
such scientific endeavors seek to answer are somehow more important to the
average taxpayer than understanding the scope of biodiversity? To whatever
extent the idea of "Biophilia" is real, I would wager not.

Stepping down off the soapbox....

In the context of the above, I would like to share my own perspective of the
uBio effort. Through extensive communications with Dave Remsen, I feel that
I have a pretty good understanding of what they are all about.  Here's my
take:

As I understand it, uBio is about mapping names to relevant published
information, and also mapping names to other names; NOT (as some seem to
think) about mapping names to taxonomic concepts.  It is intended as a
thesaurus in the literal sense -- linking text strings to other text
strings -- not distinguishing valid taxonomic concepts from invalid
taxonomic concepts.

There is an OCEAN of published literature out there.  Locked within this
literature is extensive information relevant to biological organisms.  The
link between the information and the organism exists only in the form of a
text string (a biological name).  The library community, on their own
accord, has begun to digitize reams and reams of published literature,
creating not only images of printed pages, but searchable text conversions
as well.  At the moment, we (both as taxonomists, and as human beings) do
not have easy access to this information, except by reading it, line by
line, looking for the pearls of information among the sea of irrelevant
text.  As such, a huge part of the advantage of converting it to a digital
form (automated indexing) is not realized. The uBio effort is trying to
build tools that DO take advantage of this feature of digitizing -- such
that the information currently locked up within the vast published
literature can be quickly and easily located.

In order for a computer to recognize the difference between a text string
that is a biological name and any other text string, the computer needs
access to a list of text strings of known biological names. Given such a
list, the computer can scan a large text document for instances of
biological names as they appear within the document (at MUCH greater speeds
than a human could), and thereby automatically build an index of occurrences
of such names in the document.  This, by itself, could be a tremendously
useful feature for both layfolk and taxonomists, not to mention an essential
step in building an online Encyclopedia of Life.  But that's not where
uBio's task ends.

Among the vast list of biological names, there are certain equivalencies.
For example, someone who wanted to locate information linked to the
text-string "ant" might also want to locate information linked to the
text-string "Formicidae". Similarly, someone wanting to locate information
linked to the text-string "Pan troglodytes" would likely also want to
discover information linked to the text-string "Simia troglodytes". Thus, a
thesaurus is needed to establish some form of equivalencies among names.

It does indeed require the talent and skill of a seasoned taxonomist to
assert that "Species A is a junior synonym of Species B" (i.e., making an
assertion about a taxonomic circumscription).  It does NOT, however, require
such talent and skill to simply record the fact that "AuthorX (YearY)
asserted that Species A is a junior synonym of Species B". In compiling a
list of the latter, one can build a useful thesaurus link between "Species
A" and "Species B", such that a search for information linked to one will
also retrieve information linked to the other.

Would such a thesaurus and index be perfect?  OF COURSE NOT!!  There are
problems of homonymy (note Dave Remsen's earlier query to the list about
this topic), and problems of hopelessly convoluted synonymies that only a
truly seasoned and talented taxonomists could ever sort out.  But that
doesn't mean that the existence and subsequent refinement of such tools
would be of much greater help to both taxonomists and layfolk, than they
would hurt.

Now, some questions for the skeptics:

1) How many of you would find value in the ability to search for any
biological name, and instantly get a listing of publications that made some
reference to that name, as well as direct links to the relevant published
content?

2) Would you rather have to manually search for every potentially synonymous
name one at a time, or would you prefer that a thesaurus of PREVIOUSLY
PUBLISHED synonymies be included such that you get links to published
information related to both the name you searched for as well as other names
that have been treated as junior or senior synonyms?

3) Are any of you actively building tools that will automatically establish
links between biological names and published information (emphasis on
"automatically")?  If so, what is the taxonomic breadth of your efforts? Are
you coordinating with others doing similar projects so that your efforts can
be combined?  Will the tools that you are developing be useful for other
taxonomic groups, and will they be made freely available?

4) Can you successfully articulate an argument that the problems of homonymy
and hopelessly confused synonymies outweigh the potential benefits such an
index would have, both to taxonomists and to layfolk?

The point is, uBio is NOT trying to replace or otherwise devalue the efforts
of taxonomists, in any way, shape, or form.  Rather, it is trying both to
make taxonomic information more easily accessible to non-taxonomists
(something that would probably lead to an increased perception of the value
of our field among average taxpayers), and also remove a lot of the tedium
of taxonomic progress within our field by allowing instant access to large
quantities of information, which would otherwise have required many hours of
manual reading.

As for funding, the way I see it, the money that uBio has brought into this
is NOT money that would likely otherwise have been available for us
taxonomists to do what we think is important (I base this on more than
speculation).  In other words, this is much less about stealing biodollars
than it is about increasing the total number of biodollars.  To say that the
uBio effort had some sort of moral obligation to include direct support for
taxonomists is no different from saying that every ecological grant has a
moral obligation to include direct support for taxonomists.  Maybe they
do....but that basically turns us into parasites -- existing only to serve
the needs of others.  I think we need to get past this, and represent
taxonomy as a field that stands on its own merits for understanding the
world around us -- just like physics and astronomy.

I realize that my perspective is not at all consistent with the popular view
on this list, and I am bracing myself for the anticipated response.  But my
bottom line is this:  whereas others see the uBio effort as some sort of
insult (assault?) on the field of taxonomy, I see it as a welcomed input
from an outside but related field that will save me a LOT of work, and allow
me to spend more of my time doing what my talent and skills as a taxonomist
(stop snickering!) are best suited for -- understanding the nature and scope
of biodiversity.  I also see the uBio effort as a necessary step towards
building Wilson's concept of an online "Encyclopedia of Life" -- a step that
doesn't seem to be actively pursued from within the taxonomic community (at
least not on such a broad taxonomic scale).  Best of all, I see it as
funneling money that would likely otherwise NOT have been spent on
biological endeavors into a project that can help taxonomists and the field
of taxonomy at several levels.

That is all.

Aloha,
Rich

Richard L. Pyle
Natural Sciences Database Coordinator, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org
http://www.bishopmuseum.org/bishop/HBS/pylerichard.html