[SEEK-Taxon] Re: GUID

[Kennedy, Jessie]

Hi James,

> 
> thanks for explanations.  So far, my understanding is that
> SEEK Taxon is looking for an indirect pointer system usable
> globally and uniquely (am I right?  I hope so...).

yes
> 
> Suppose two databases.  Each of them has a record for a single
> potential taxon, but these recoreds are incomplete.  Say,
> database A has (rank, name, author, null) record while
> database B has (null, name, author, citation) record.  It is obvious
> for an expart of the small taxon that these records should point to
> the same taxon, but not for others.  Does SEEK Taxon team looking for
> an automatic mechanism to map these two entries to the same GUID
> without human expart's help?

we certainly would like both of these potential taxa to have the same GUID,
however the mechanism by which they get the GUID has not been defined as
yet. This is part of our ongoing discussions as you will have seen firstly
looking into GUIDs per se and is something that will be discussed at the
Edinburgh meeting. In an ideal world both of the databases in your example
would start to use GUIDs and the issue would then be at allocation time of
the GUIDs how do we ensure that we do not allocate a new GUID for a taxon
which has already been issued a GUID but is not obvious from the contents of
the records. This then brings us to centralised or distributed issuing of
GUIDs each of which have their pros and cons as you may have seen from
Dave's writings. So I'm sure we will be exploring this issue. In the interim
we are planning on allocating GUIDs at the point of transfer from a provider
database to the SEEK database and we will be developing a mechanism to
semi-automate the matching of entries. So we won't likely be doing it
without an expert at first however this may change as we learn. 
There are other issues which will have to be decided before this anyway,
like "in existing taxonomic databases, are the taxa represented there an
interpretation of the taxa from that DBs providers point of view or are they
simply the representation of some other taxa described elsewhere". If the
former the taxon would get a new GUID if the latter it would use a GUID
(possibly)already allocated.

I hope you're going to make it to Edinburgh to be part of these
discussions......

Jessie
> 
> Cheers,
> James
> 
> --
> Dr. Nozomi "James" Ytow
> Institute of Biological Sciences / Gene research center
> University of Tsukuba
> Tsukuba, Ibaraki 305-8572
> Japan
> _______________________________________________
> seek-taxon mailing list
> seek-taxon at ecoinformatics.org
> http://www.ecoinformatics.org/mailman/listinfo/seek-taxon
>