Names for BioDiv Informatics

Richard Pyle deepreef at BISHOPMUSEUM.ORG
Tue Feb 1 00:32:44 CST 2005


Hi Wolfgang,

What you describe is, in my mind, the current "hot topic" in taxonomic
informatics (actually, a lot of folks have been thinking about it for a long
time, but only now are there real signs of progress).

I also see three different kinds of "valid" names, but in a slightly differ
way.

I think your number 1 is pretty universal.  I use the term "code-compliant"
to represent these names.  Note, however: I do NOT believe that
non-code-compliant names can be summarily ignored by taxonomic informatics.
There are many "scientific names" that appear in literature, that do not
fulfill all requirements of the relevant code; but which need to be
indexed/tracked so as not to lose access to information associated with that
non-code-compliant name.  This is a big and ugly topic of discussions,
perhaps not best delved into here.

As for your number 2 -- this is what most advanced taxonomic informatics
folks are focusing on right now (e.g., the TCS group of the SEEK project,
and many others).  That is, build an index not just of code-compliant names;
but of names anchored to a usage of the name (Aus bus Smith SEC Smith, vs.
Aus bus Smith SEC Jones). In most manifestations, this takes the form of a
taxonomic "Concept"; but the point is, there is wide recognition amongst us
database/taxonomic indexing nerds that that the units of cross-referencing
should be "Name SEC Usage"; rather than simply "Name".

Where we differ slightly is with regard to your number 3.  I don't think
we'll ever get enough agreement among taxonomists to universally accept a
set of "standard names" (i.e., one agreed-upon classification &
nomenclatural landscape).  But there might be certain organizations (e.g.,
ITIS, Species 2000, BIOSIS, IPNI, Catalog of Fishes, etc.) that consider the
complete spectrum of names, and make some assertions about which should be
thought of as "valid" (sensu Zoology) or "accepted" (sensu Botany), and
which should be regarded as subjective synonyms.  It's not yet clear whether
this group is fundamentally different from the previous one.  The main
difference is really in how broad a scope of names are covered.

In any case, the first step (in my mind) is to focus on building the index
of Level-1 (Code-compliant) names, and move on from there. This task would
be MONUMENTALLY easier if the caretakers of the major nomenclatural Codes
would adopt some sort of universally-accepted name registration system,
perhaps in collaboration with GBIF (watch this space). This will serve as
the core foundation on which efforts such as TCS can grow and blossom.

> -----Original Message-----
> From: Taxacom Discussion List [mailto:TAXACOM at LISTSERV.NHM.KU.EDU]On
> Behalf Of Faunaplan at AOL.COM
> Sent: Tuesday, February 01, 2005 12:10 AM
> To: TAXACOM at LISTSERV.NHM.KU.EDU
> Subject: Names for BioDiv Informatics
>
>
> Dear All
> so far, published information on taxa is distributed in thousands
> of papers &
> websites and unpublished data are mainly deposited in collections
> ("libraries
> of life"). Still there is no infrastructure for easy access to such vital
> information, but GBIF and its partners have started to build the roads...
>
> What I feel is still basically missing are checklists of valid names that
> could serve as keyword directories for the fast-growing
> biodiversity info system.
>
> Ron Gatrell has recently used the term "taxonomically correct
> names" when he
> announced the Lepidoptera checklists, and many taxonomists have already
> created wonderful on-line checklists of valid names.
> But, will there ever be checklists of universally accepted/ valid names?
> In my understanding, valid names are all Code-compliant names that are
> accepted by individual taxonomists according to their preferred
> concepts. E.g.,
> Bembidion (Nothocys) jeannelicum Toledano, 2002 and Nothocys
> nitidus Jeannel, 1962
> are both valid/ accepted names for the same neotropic ground beetle, and I
> cannot say one is "correct" and the other one is not; it depends
> on whether you
> prefer Bembidion as a wide or narrow genus...
> Yet there is no doubt that universally accepted names lists for
> the purposes
> of biodiversity informatics would facilitate a lot of things. So
> why not use a
> specific term in order to avoid confusion, - e.g. "standard names",
> "keynames", or whatever you prefer.
> Such names would have to be carefully checked for
> Code-compliance, and they
> should represent current systematic concepts as far as possible
> (preferrably
> staying on a moderate "conservative" side, especially in context
> with premature
> splitting of traditional genera).
> In other words, we would have 3 categories of names that should not be
> confused:
> 1.) Available/ validly published names, i.e., all names that are Code
> compliant incl. synonyms, etc.
> 2.) Valid/ accepted names, i.e., all names that are accepted as valid by
> individual taxonomists/ representing alternative classifications.
> 3.) "Standard names" (or whatever term you prefer), i.e., all
> names that are
> recommended for biodiversity informatics purposes; names that would
> facilitate globally compatible "species banks", on-line species
> distribution maps,
> google searches, etc.
> Such lists (with yearly updates) could be organized by GBIF
> partners, - and,
> actually, they are already offered for several groups of organisms (e.g.,
> Species2000 checklists).
>
> Well, these are just some musings that came to my mind while finishing the
> second edition of the carabid beetle checklist.
> I do feel enthusiastic about the chance to contribute to the CoL
> (Catalogue
> of Life) but the names I can provide should not be misunderstood as
> disapproving alternative taxonomic concepts...
>
> Best wishes,
> Wolfgang
> ------------------------------------------------------------------
> ------------
> --------------------------------
>
> Wolfgang Lorenz
> Buero/ Verlag fuer Faunistik und Umweltplanung
> D-82327 Tutzing
> Germany
>
> www.carabidfauna.net




More information about the Taxacom mailing list