Registration (was: TurboTaxonomy?)

[Richard Pyle]

> Thanks to Richard Pyle  for providing the links to the calacademy
> ichthyology website! (Again I'm deeply impressed by the wonderful pioneer
> work presented by that team!)

Me too!

> I'd like to comment on two points in Richard Pyle's and Brian Tindall's
> recent responses in this thread:
> 1) Unavailable Names
> On 25.08.04, Richard Pyle wrote:
> >However, I think it would be a mistake to include only the
> Code-"Available"names (as Listed according to Art. 79), and
> ignore all "Unavailable" names that
> have historically been used at one time or another as though they were
> available (with appropriate fields detailing their unavailability).<
>
> I doubt that the total of such names could ever be "registered"
> since only a small portion of them (e.g., incorrect original spellings,
names
> totally or partly suppressed by a ruling, nomina nuda, nomina oblita)
could
> be caught in the process of scanning the taxonomic literature.
> But how to include all such
> things like inadvertent misspellings in literature (not only in taxonomic
> literature!), keystroke errors in databases like ITIS, vernacular
> names, - and not
> forgetting misapplied names or even "pro parte" misapplied names?

...and don't forget "Museum names" -- names that appear only on Museum
specimen labels.

There's a not-so-perfect line between what I would think of as an
"unavailable name", as distinct from a misspelling of an available name.
The latter are pretty easy to deal with in any index that endeavors to track
which name was used in which reference (i.e., simply record how it was
spelled in each indexed usage).  But then there are a fair number of
code-designated "unavailable" names that have, in the past, been used as
though they were available names (indeed, some still are used).  Bill
Eschmeyer's team seems to do a reasonable job (in my opinion) of including
names that are unavailable by code rules, but that still are valuable to
keep track of when sorting out historical taxonomic works, and what authors
meant.  Of the 10.5K genera names in the Catalog of Fishes, only about 600
(<6%) are flagged as not available -- hardly a burdensome load of excess
baggage to keep track of.  Among the 56K species-group names, just over 2.6K
(<5%) are flagged as "not available".  Eliminating this small set of
unavailable names (nomina nuda, infrasubspecifics, homonyms, names proposed
for hybrids, suppressed names, etc.) would rob future generations of the
work and research that were required to identify these names as unavailable.

Besides, computers are very very good at keeping track of very LARGE sets of
information.  Even if we indexed every text string ever used by anyone in
print to represent any organism (including vernacular names in all
languages), I bet we'd fall far short of the size and scale of index that,
say, Google uses to locate 80 appearances of "Centropyge boylei" among >4.2
billion web pages in only a quarter of a second.

> So why not start with Lists of existing available names as mandated in
> Article 59  ICZN-4, supplemented by a register mechanism for
> updates. In my mind, as
> a second step,  the bulk of unavailable names could be better addressed in
> other context, e.g. when detailed catalogues of  literature
> citations of names
> are prepared (yet another difficult task that would be strongly
> supported by
> the existence of Lists of Available Names)

We seem to be in full agreement on this.

> 2) Taxonomic judgement
> On 26.08.04, Brian Tindall wrote:
> >Trying to work out names, dates, authors, publucation and type for all
> published taxa is one task - the other is also flagging the
> status of a name (I
> don't think you can avoid this).<
>
> Since the Code deals with nomenclature, Article 79 of ICZN-4 does
> not cover
> the flagging of the taxonomic status of names and I suspect that we cannot
> expect such things to happen in a Code-mandated registration
> system. Was not, so
> far, the mixing of nomenclature and taxonomy one of the main reasons for
> dispute and reluctance to the Registration idea?

In my view, the only Code-mandated part (if any) would be to register the
establishment of the name (protologue, or original description) to the
extent that it is demonstrably compliant with the Articles in the Code.

However, that is not to say that a Registration system could not serve as
the kernel for a larger system that kept track of name usages -- that is, an
index of how names have been used in various References (published &
unpublished) through history.  Such an index would serve as the primary
source for who treated which names as valid, and who treated which names as
junior synonyms of other names.  Another index layer could be applied on top
of this Name-Reference index by what I would call "Meta-Authorities" (ITIS,
SP2K, CoF, etc.).  Rather than each such Meta-Authority directly proclaiming
which names are valid (in the Zoological sense) and which are synonyms, they
could each pick one of the pre-existing Name-Reference instances that "got
it right" in their respective views.  Yet another layer is the concept
mapping stuff -- something along the lines of what the SEEK taxon group is
trying to develop
(http://seek.ecoinformatics.org/Wiki.jsp?page=SEEKTaxonCommunity).

At the core we find the names.  Then there is a layer for how those names
have historically been used.  Then there is a layer of opinions about which
of those name-usages represent the "current usage" in the eyes of some
authority.  Then there is a layer for how each name-usage maps to an actual
scope of organisms living, recently dead, and yet-to-be-born.  Many things
to index.  In my view, only the first, core layer should be intertwined with
the Codes in the form of "Registration" -- but the registration system
should be designed such that it can easily serve as the kernel for the
higher layers of bio-nomenclatural indexing.

> On the other side, I fully agree it would be highly desirable to have such
> information included by reference to important publications.
> However, don't you
> think that producing such work is another task - not for a Code-mandated
> R-thing but for the individual taxonomist or team of taxonomists?

To re-cap the essentials of my diatribe above, I think the "R" word should
only apply to the names -- but the system should anticipate and accommodate
the needs of other efforts to index things outside the scope of Code
mandate. (I know your comments are directed at Brian, but I can't help
myself...)

Aloha,
Rich

Richard L. Pyle, PhD
Associate Zoologist & 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