[Taxacom] Chameleons, GBIF, and the Red List

Richard Pyle deepreef at bishopmuseum.org
Sun Aug 24 16:47:14 CDT 2014

I'm not sure I understand what you mean when you say "GBIF is trying to
treat all areas as the same" in this context.  It's not clear to me that you
fully understand what GBIF actually does.

The world is full of thousands of databases containing organism occurrence
data.  These databases are scattered across hundreds (thousands?) of
institutions.  Many of these institutions are investing effort to digitize
the information about these organism occurrences, and expose those data on
the internet.

Without GBIF (and/or iDigBio), it's EXTREMELY time consuming for a
researcher to hunt down these thousands of individual databases, convert the
records into a single standard format, and pull all the information together
in one place.  The role of GBIF is to serve as an aggregator of these
organism occurrence records, so researchers like you and I can search
content across thousands of individual databases scattered all over the web
with a single, fast query.  Moreover, GBIF developers have created some very
helpful tools to make creating such queries useful.

The only way that GBIF is "trying to treat all areas as the same" is in its
support and implementation of the DarwinCore standard (via the DarwinCore
Archive format).  But this is entirely about the structure of data as
presented and packaged in electronic form.  It has nothing to do with what
this thread seems to be about -- which is the quality of information content
(particularly in terms of taxonomic "accuracy").

As Donat has already explained, we're not talking about "GBIF data", we're
talking about data managed by hundreds (thousands?) of institutions in
thousands of databases around the planet.  GBIF is doing us the
(tremendously valuable) service of aggregating those data in on place, to
make it incredibly easy for us to locate it.

People in this thread are complaining about the "quality" of that
information; yet GBIF (inexplicably) is the apparent target of the
criticism.  Don't blame the mirror if you don't like what you see in the
morning when you gaze into it.

Perhaps some would prefer that GBIF only show "clean" data (which, of
course, would require a definition of "clean", and a mechanism from
identifying such "clean" records from "unclean" records).  Personally, I
want to see ALL the data, and then I'll make my own decisions about what is
clean and what is not. 

> -----Original Message-----
> From: Stephen Thorpe [mailto:stephen_thorpe at yahoo.co.nz]
> Sent: Sunday, August 24, 2014 11:14 AM
> To: 'TAXACOM'; deepreef at bishopmuseum.org
> Subject: RE: [Taxacom] Chameleons, GBIF, and the Red List
> But therein may be the problem! What works well for one (taxonomic) area
> doesn't work well for others. GBIF is trying to treat all areas the same.
> that most species are tiny arthropods, GBIF may well end up a big mess
> just a few useful small things thrown in.
> Stephen
> --------------------------------------------
> On Mon, 25/8/14, Richard Pyle <deepreef at bishopmuseum.org> wrote:
>  Subject: RE: [Taxacom] Chameleons, GBIF, and the Red List
>  To: "'Stephen Thorpe'" <stephen_thorpe at yahoo.co.nz>, "'TAXACOM'"
> <taxacom at mailman.nhm.ku.edu>
>  Received: Monday, 25 August, 2014, 4:18 AM
>  > I suspect your
>  hypothetical "tough call" is actually the usual  case, and  that a  >
great  many
> published mentions of "Aus bus" may refer to  two or more  > species,
> without there  being an easy (or even any) way to tell which  species.
>  I
>  suppose that may be true in some areas, but not in ours.  For us, the
> calls are the edge case  (<1%).  But even for groups where tough calls
> more common, you do realize that the fault  is not with the Museums, or
> with the data  aggregators or even (gasp) the bureaucrats, right? The
fault  is
> with the nature of taxonomy and the  taxonomists themselves.
>  Disambiguating the
>  "tough calls" is only "tough" because of  insufficient  documented
> information (by the
>  taxonomists) -- not because of bad data.
>  Clever data structures and software services  can do some pretty magical
> things, but one  cannot extract blood from a stone.
>  Aloha,
>  Rich

More information about the Taxacom mailing list