Open GoogleCodeExporter opened 8 years ago
UBC is also unsure as to how to tackle unnamed hybrids.
Currently, if we have a name such as "Potamogeton alpinus x lucens", we enter
'Potamogeton' in GENUS, and 'alpinus x lucens' in SPECIES. The authorities we
are unsure how to enter... we currently just put them both in SPECIES AUTHOR,
separate by a '/'. Any recommendations?
Also something to try to tackle: more complicated unnamed hybrids. For
example, "Centaurea diffusa x stoebe subsp. micranthos". For now, similar to
above, we enter 'diffusa x stoebe subsp. micranthos' all in SPECIES, with all 3
authorities separated by '/' in SPECIES AUTHOR.
These are of course not ideal, but what we have done in the meantime until a
recommendation could be suggested. Thank you!
Original comment by amber.sa...@gmail.com
on 2 Aug 2011 at 6:20
Would definitely be useful to tackle in the guidelines. I have asked Dusty
McDonald, programmer on Arctos at the University of Alaska Museum of the North,
to comment on how this is all handled in Arctos, in case it sheds light on the
issue. Here is his response:
"There are two distinct things going on here.
1) Named hybrids are taxonomy
(http://ibot.sav.sk/icbn/frameset/0069AppendixINoHa001.htm) and are therefore
handled under our rules for forming taxon names (implemented as a DB trigger):
http://code.google.com/p/arctos/source/browse/arctos/DDL/triggers/trg_taxon_comp
ound_names.sql
Cleverly-named variable "stoopidX" is the multiplication sign - long story.
ICBN genus allows both "Propercasename" and "{stoopidX}Propercasename", species
allows "somename" and "{stoopidX}somename", etc.
Those taxon names are then available to be used in identifications. (Even
complex ones - but we unfortunately don't seem to have any named-hybrid hybrids
to showcase.)
http://arctos.database.museum/name/%D7Agroelymus%20hultenii
http://arctos.database.museum/name/Carex%20%D7lidii
etc.
2) Non-named hybrids are not taxonomy, but rather identification. Table
identification_taxonomy allows more than one taxon to be used in
Identifications, and Identification.Identification_Formula hooks those taxon
terms together under Identification.scientific_name. I don't think we have
anything more complicated than "A x B" (see, e.g.,
http://arctos.database.museum/guid/UAM:Herb:47064,
http://arctos.database.museum/guid/UAM:Herb:136369, and
http://g-arctos.appspot.com/arctosdoc/identification.html#id_formula) so far,
but implementing say "(((Ax(BxA))xA)xC)xD)" would be a matter of inserting a
value in a code table and modifying one form to use >2 variables - that's true
at least until we run out of upper-case ASCII characters.
Identification.Scientific_Name is just a string and could be represented in any
number of ways, but the components of that string point to
names/concepts/whatever you want to call e.g.,
http://arctos.database.museum/name/Eriophorum%20scheuchzeri%20subsp.%20arcticum
(ER diagram available at http://code.google.com/p/arctos/downloads/list)
All that said, and probably what you're looking for here, we'd not expect
anything using DWC to support that level of precision. We'd just send whatever
the accepted Identification.Scientific_Name string is in our DWC data, and
expect a link back to the specimen detail page so people can (hopefully!) make
sense of what we're doing.
MaNIS, Ornis, and GBIF all fail at providing a link back to specimens, and GBIF
just ignores everything after a second space in the scientific name
(http://data.gbif.org/occurrences/201421766/). We're still a long ways from
being able to share our precise data in any sort of useful manner.
-D"
Original comment by gtuco.bt...@gmail.com
on 3 Aug 2011 at 7:17
Original issue reported on code.google.com by
peter.de...@gmail.com
on 9 Jun 2011 at 4:35