[Taxacom] Ebbe Nielsen Challenge submission

Shorthouse, David davidpshorthouse at gmail.com
Fri Sep 7 07:38:46 CDT 2018


Dan, Rod -

This is interesting. I'm assuming we'd want a way to differentiate
what was ID'd/collected by a machine vs. what was accomplished by a
human. They have very different signals. Are the Darwin Core terms
identifiedBy and recordedBy appropriate? The definition of
identifiedBy is as follows: "A list (concatenated and separated) of
names of people, groups, or organizations who assigned the Taxon to
the subject." No machines there. That said, if you poke around ORCID a
bit, it's clear that there are a heap of spammy accounts that are not
people at all, eg https://orcid.org/0000-0002-5950-6036 (sigh).
Clearly, someone made these with the expectation that machines at
Google will bump up their page rank. In theory, a mass collecting
machine and a mass identification machine could have an ORCID or other
identifier such that if we had the desire to, we could differentiate
human vs. machine instead of attempting to do this via strings of text
as we do now (eg is that D. Janzen in identifiedBy the same person as
this other D. Janzen?!). But, that puts the onus on some third party
to ensure GUID-assigning organizations like ORCID (meant for humans)
are never purged if these accounts do not represent the identifies of
humans.

This has to be taken seriously in an "Agent Actions" extension to Darwin Core.

David
On Fri, Sep 7, 2018 at 4:32 AM Roderic Page <Roderic.Page at glasgow.ac.uk> wrote:
>
> Dan, David will no doubt have something to say about this, but I think there are frameworks to handle this, such as nanopubs http://nanopub.org/wordpress/ . Put simply, one way to handle machine ids is to have a DOI for, say, a barcode dataset, and use that DOI as the GUID for the identifier. Then we can attribute all these identifications to the machine/process that did the ids. If specimen gets new identification in subsequent analysis, Darwin Core can already handle that (it’s possible to have multiple ids).
>
> Regards,
>
> Rod
>
> Get Outlook for iOS
>
>
>
> On Fri, Sep 7, 2018 at 4:53 AM +0200, "Janzen, Daniel H" <djanzen at sas.upenn.edu> wrote:
>
>> what is the plan for when the ID is done by a DNA barcode, a machine, and then equally, later on upgraded or otherwise modified by machine (and collected in the beginning by machine, such as a Malaise trap and never seen by a taxonomist other than a machine).  Or when the ID is done by unknown persons sorting the specimens into unit trays in a museum (whenever).
>>
>> ??
>>
>> Dan Janzen
>>
>>
>> > On Sep 6, 2018, at 8:00 PM, Shorthouse, David  wrote:
>> >
>> > Thanks! Am pleased it's sparked interest. It means we're on the right track
>> > with our thoughts about an extension to Darwin Core.
>> >
>> > -D
>> >
>> > On Thu, Sep 6, 2018 at 3:54 PM Eduardo Dalcin  wrote:
>> >
>> >> Well done! Amazing tool.
>> >>
>> >> Eduardo
>> >> --------------------------------
>> >> [image: photo]
>> >> *Eduardo Dalcin*
>> >> Inst. Pesq. Jardim Botânico do Rio de Janeiro
>> >> +55 21 3204 2116 <+55+21+3204+2116> | +55 21 98393 3344
>> >> <+55+21+98393+3344> | edalcin at jbrj.org | Skype: eduardo.dalcin
>> >>
>> >>  | WhatsApp: 55 21 983933344
>> >> Rua Pacheco Leão 915 - Sl 203 - Rio de Janeiro - RJ - Brazil
>> >>
>> >>
>> >> Portal de Dados: http://dados.jbrj.gov.br
>> >> Blog: http://eduardo.dalc.in
>> >> Agende uma reunião: https://doodle.com/edalcin
>> >> "To organize Brazil's biodiversity information and make it universally
>> >> accessible and useful."
>> >>
>> >>
>> >> Em qui, 6 de set de 2018 às 11:26, Shorthouse, David <
>> >> davidpshorthouse at gmail.com> escreveu:
>> >>
>> >>> All -
>> >>>
>> >>> I threw my hat in the ring for GBIF's Ebbe Nielsen Challenge and
>> >>> welcome you to take a gander at https://bloodhound.shorthouse.net/. If
>> >>> you also have an ORCID account, try your hand at declaring the
>> >>> specimens you identified or collected - assuming they're shown to you!
>> >>> (*ducks and cowers*) And, if you're up for it, make your profile
>> >>> public.
>> >>>
>> >>> I'm using this as fodder to help develop a 1:many extension to Darwin
>> >>> Core with the joint RDA/TDWG Interest Group who are interested in how
>> >>> to best attribute occurrence data to people. The extension has yet to
>> >>> be named but will attempt to express from the source, a better
>> >>> mechanism to share granular information about the actions people take
>> >>> when collecting, identifying, and curating occurrence data & will
>> >>> include identifiers for people. The logical identifier in today's
>> >>> academic environment is ORCID - thus we can have a collective think
>> >>> about academic credit systems for people who work at and contribute to
>> >>> natural history collections -, but we'd expect the extension to be
>> >>> supportive of other such identifiers. It's early days but here's its
>> >>> workspace: https://github.com/tdwg/attribution. Others involved in
>> >>> this possible extension to Darwin Core are cc'd here.
>> >>>
>> >>> Cheers,
>> >>>
>> >>> David P. Shorthouse
>> >>> _______________________________________________
>> >>> Taxacom Mailing List
>> >>> Send Taxacom mailing list submissions to: Taxacom at mailman.nhm.ku.edu
>> >>>
>> >>> http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> >>> The Taxacom Archive back to 1992 may be searched at:
>> >>> http://taxacom.markmail.org
>> >>> To subscribe or unsubscribe via the Web, visit:
>> >>> http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> >>> You can reach the person managing the list at:
>> >>> taxacom-owner at mailman.nhm.ku.edu
>> >>>
>> >>> Nurturing Nuance while Assaulting Ambiguity for 31 Some Years, 1987-2018.
>> >>>
>> >>
>> > _______________________________________________
>> > Taxacom Mailing List
>> > Send Taxacom mailing list submissions to: Taxacom at mailman.nhm.ku.edu
>> >
>> > http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> > The Taxacom Archive back to 1992 may be searched at: http://taxacom.markmail.org
>> > To subscribe or unsubscribe via the Web, visit: http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> > You can reach the person managing the list at: taxacom-owner at mailman.nhm.ku.edu
>> >
>> > Nurturing Nuance while Assaulting Ambiguity for 31 Some Years, 1987-2018.
>>
>> _______________________________________________
>> Taxacom Mailing List
>> Send Taxacom mailing list submissions to: Taxacom at mailman.nhm.ku.edu
>>
>> http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> The Taxacom Archive back to 1992 may be searched at: http://taxacom.markmail.org
>> To subscribe or unsubscribe via the Web, visit: http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
>> You can reach the person managing the list at: taxacom-owner at mailman.nhm.ku.edu
>>
>> Nurturing Nuance while Assaulting Ambiguity for 31 Some Years, 1987-2018.


More information about the Taxacom mailing list