uBio & Free taxonomic information

[Doug Yanega]

Dave asked:

>Do you disagree, in principle, to our trying to address the problem
>from our library perspective?  Do you recognize that we have a need and
>stake in the global taxonomic initiatives?

In principle, as you well know, taxonomy is more than a library
exercise; therefore, the product you envision is not something that
would help the taxonomists as much as it would be a convenient
interface by which non-taxonomists could extract taxonomic
information without ever consulting a taxonomist. Meaning, of course,
that this information would come to them without all of the necessary
caveats about lack of consensus about various taxonomic opinions -
and if you did give them a complete listing including every variant
classification through history including a given taxon, they wouldn't
necessarily be happy with it: after all, the main gripe of
non-taxonomists is the flux in classifications, when all they want is
a single correct answer. That's why lists like ITIS which *give* a
single "authoritative" name for each taxon are their favored
interface. As for why existing global taxonomic initiatives have
failed, that's a complex issue, but I don't really see how uBio has
any better odds of success - mostly because one of the primary
reasons these initiatives fail is because they need to have an
enormous workforce of taxonomists at their foundation, and they
don't. The best electronic taxonomic resources we have now are those
that were designed and implemented by taxonomists working in their
specialized areas of interest; the smaller the taxon, the better the
resource, as a rule of thumb. To me, that indicates the way to get
what we need is to build from the bottom up, rather than
"trickle-down".

>Do you feel we should spend less on developing the technical
>infrastructure which does not exist to meet our needs and spend more on
>simply purchasing names for the data model?

Who says the technical infrastructure doesn't exist already? I know
of at least one major research team that has already demonstrated
their ability to extract information from all existing online
resources simultaneously via a single query - in essence, fusing all
existing lists into one big list, from the client's perspective.
Unless I'm mistaken, that makes your approach look nearly obsolete.
In essence, your plan sounds like building a giant library with
everything in it, when a simpler and INCREDIBLY more efficient
solution is to link all existing libraries.

>Should we pay more
>taxonomists directly?  About 20% of the grant (100K) is being used to
>support taxonomists or the collection of name information.

Where does one look to find the application forms for these grants?
Or is it not intended to be a competitive award process?

To my mind, this is all backwards. What needs funding is the
development of taxonomic expertise - meaning hundreds or thousands of
monographic, species-level revisionary works - and *linking* the
results of all these revisions is secondary. 80% of the budget should
be taxonomy, 20% data management, not the other way around. Compiling
a bunch of taxon names that have not been examined by a competent
revisor in the last 100 years is not really performing the kind of
service that the community needs.

>What is your opinion of the current suite of offerings of 'lists of
>lists' such as ITIS or the SP2000 annual checklist?

My gripe is the interface. In my capacity as a museum curator, I do
actually need a definitive checklist, because every unit tray can
only have a single name on it, and every genus has to be in a single
family in a single location in the collection - but it also means I
need to be able to get *output* in a checklist format, rather than
trying to compile it manually by doing one name search at a time. So
far I've personally compiled some 19,000 insect generic names and
over 100,000 species names into the hierarchic authority file I use
for my museum, and not one of them has come via ITIS, Sp2000, the
BDWD, or other resources that lack the right sort of interface for
bulk data extraction. If all you're doing is building yet another
giant name-retrieval engine, uBio won't do me any good for my
specialized needs, either.

Sincerely,

--

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://cache.ucr.edu/~heraty/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82