[seek-taxon] SEEK taxonomic model

[Robert K. Peet]

Hi Jessie et al.

> We do not believe this will be the final schema and that as we start trying
> to do things will find that we need to add other attributes (especially to
> the SEEK database schema but possibly less to the transfer schema).
>
> There are a few things we discussed that we will follow up later.
>
> 1. It is possible that publication date and author should be the same as
> concept date and author.
> In principal this is true but we're not sure whether or not for example we
> might see an ITIS concept as an ITIS concept or maybe the inclusion in their
> hierarchy of someone else's concept which is referred to in a publication by
> the original author rather than seeing ITIS version xxx as the publication -
> this will be tested.

I still vote for these being equal.  Inclusion in ITIS constitutes a
concept, which ITIS might well synonomize with a different concept with a
different date and author.

> 2. Many taxonomic systems specify the status of a concept name). Currently I
> have excluded this the reason being that if someone publishes a
> classification then as far as they are concerned it is accepted (by them!)
> If it is not accepted then it won't appear. We are unsure whether or not we
> need to record the fact that someone (e.g. ITIS) thinks that a particular
> concept does not exist. - this will be evaluated for importance in the SEEK
> context.

For purposes of marking up data, sending queries, and returning data from
queries, the status stuff does not seem to matter.  However, I do think we
will want to support multiple party perspectives in our central database
for taxonomic mediation.  If we build such a database, we will need to
support the idea of status.  Mapping of relationships among taxa is not
fact, but party-specific interpretation and we should allow certain
parties to submit their interpretations to the mediation database.  Once
we start exchaning these sorts of data, status becomes critical, as do
parties (as distinct from authors).

> 3. Synonymy relationships are those defined by the original author of the
> concept. When a third party wants to say that 2 taxa are synonymous we are
> undecided as to whether or not this requires the creation of a new concept.
> This will be explored further.

In a strict potential-taxon world, there would be a new concept generated,
so I suppose we need to support it.  However, this is very much a case of
the kinds of concepts that would be lost in the condensation to super
concepts referred to in 4 below.

> 4. Bob has the notion of super taxon concepts to help stabilise taxon
> concepts this will need to be explored further but we think that if it turns
> out to be useful we will be able to extend what we've got to fit.

I agree.

> 5. the publication entity hasn't been fully defined yet, Bob is going to
> send me what he's got for VegBank and we'll look at adding this - apparently
> it's an extension of what's in EML which we could've put in but it's a bit
> much for what we need.

I have asked Mike Lee to send the XML for the VegBank publication model,
which is a modest extension of the EML model, to Jessie.

Best,
Bob



 ====================================================================

 Robert K. Peet, Professor              Phone:  919-962-6942
 Department of Biology, CB#3280         Fax:    919-962-6930
 University of North Carolina           Cell:   919-368-4971
 Chapel Hill, NC  27599-3280  USA       Email:  peet at unc.edu

             http://www.bio.unc.edu/faculty/peet/

 ====================================================================