[Taxacom] GBIF data

[Thomas Moritz]

An approach that we proposed testing several years ago at AMNH in a
(successful) NSF ITR proposal was to identify the "materials examined"
element in digitized taxon treatments and then to extract, parse and
expand the very brief   "telegraphic" descriptions of specimens examined
-- by both interpretation (based ideally on automated Web services like
digital gazetteers or biological names servers) and inference (from the
larger paper in which the taxon treatment is imbedded) -- into  a more
complete Darwin Core type specimen record. 

The case for this approach is based on the notion that published lists
of specimens constitute a special class that have been closely examined
by the author -- and also have (at least putatively?) been subject to
peer scanning... (if not close review).

Given the as yet unlikely prospect of successful mass scanning of insect
collections  -- though industrial strength enginnerring approaches are
being discussed -- this literatute-based approach seemed an alternative
worth pursuing...
(This will be particularly true if initiatives like the Biodiversity
Heritage Library Project succeed in accomplishing the  mass digitization
of the extensive legacy of natural history publications...)

One of the project Websites offers details: 
http://wiki.cs.umb.edu/twiki/bin/view/Ants/MeetingsPastAndPresent


Tom Moritz

Tom Moritz
Associate Director and Chief, Knowledge Management
Getty Research Institute
1200 Getty Center Drive Suite 1100
Los Angeles, California 90049
United States of America
1 310 440 6363 [voice]
1 310 440 7781 [fax]
<tmoritz at getty.edu>
GMT -8
>>> "Mary Barkworth" <Mary at biology.usu.edu>  >>>
One can make data available using the DarwinCore model - it draws from a
flat table. We simply export data from our database to a table on the
server to which we give GBIF access. It really is pretty simple (but,
being lazy, I did get help from a Computer Science student). I do not
see that as the major impediment to sharing data. Probably the greatest
is creating the database in the first place (which includes choosing a
system to use), with the second being reluctance to share for free
something that costs a lot to create and may be viewed as someone's
private research data, only after that do I see the mechanics of making
the data available being a problem. 

Mary

_______________________________________________
Taxacom mailing list
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom