I am posting this for Bryan as his message seems to have bounced. I leave our comments in as they are informative re identifiers.

Roger


Begin forwarded message:

From: "P. Bryan Heidorn" <pheidorn@illinois.edu>
Date: 27 April 2009 14:36:31 BST
To: Roger Hyam <roger@hyam.net>
Subject: Re: Auto-discard notification

No, I did not know it bounced. Can you forward for me while I figure it out? Univ of Illinois changed its domain which might be the problem... case in point for the need for stable identifiers that are not university domain names!

Thanks,
Bryan
-- 
--------------------------------------------------------------------
   P. Bryan Heidorn    
  Graduate School of Library and Information Science
  University of Illinois at Urbana-Champaign    

On Apr 27, 2009, at 4:36 AM, Roger Hyam wrote:

Hi Bryan,

Did you know this one bounced? Did it get through eventually?

Roger


Begin forwarded message:

Date: 25 April 2009 15:04:03 BST
Subject: Auto-discard notification

The attached message has been automatically discarded.
From: "P. Bryan Heidorn" <pheidorn@illinois.edu>
Date: 25 April 2009 15:03:33 BST
To: Technical Architecture Group mailing list <tdwg-tag@lists.tdwg.org>
Subject: Re: [tdwg-tag] darwin core terms inside tdwg ontology


Some taxon concept/name is always part of a Darwin Core record and in the loose constraints of DwC that can just be a name string but (typing back to the dreaded LSID/GUID argument) it could be a term that is "defined" within the broader taxon concept world. This can of course be _optional_ and when used might just be a string plus a reference to the other standard where it is defined. So you can just say "Quercus alba" or say Quercus alba plus a taxonomic identifier that would lead to the particular concept of Quercus alba you mean in this case with relationships to synonyms, historical usage etc, but none of those relationships are in the DwC record.

A user of a collection of such records could then decide if the string "Quercus alba" is a sufficient qualifier for the task at hand or if a particular circumscription resolved by the taxonomic reference is required. For example, if you are really interested in all oaks you could build a parser that reads the records and decides that the word "Quercus" means oak or you could choose to use the taxonomic LSID/QUID to have the computer go to a taxonomic resolution service that would be able to answer if or if not this is a Quercus. 

The same type of mechanism could be used for locality. 

Any of this requires just an optional side element on fields identifying two things. It need not be a globally accepted standard if there are two components. One component that is an identifier and another identifying the resolution context. That context could be the "Backwoods Arboretum Shop and Taxonomic Authority Database (BASTARD)"  that defines everything by its historical woodworking use or the context can be a  TDWG approved taxonomic resolution service respected by systematists (but not necessarily wood workers or conservationists).

-------------------------------------------------------------------- 
P. Bryan Heidorn, Program Director
Division of Biological Infrastructure Rm. 615 N
National Science Foundation
4201 Wilson Blvd.
Arlington VA 22230
(703) 292-8470

On Apr 25, 2009, at 2:52 AM, Kevin Richards wrote:

I see the ontology as a model of ALL (hopefully, eventually all) the data in our domain of biodiversity informatics. 
I would love to see it as a standard (at the least it might give it a bit more clout).
I agree that the ontology is useful to tie other TDWG schemas together, using it as a core/master model.  I would be happy to see it used for ALL tasks within TDWG, but I understand the usefulnes of the more specific schemas/standards - horses for courses.
 
If I understand Stan here, I agree with him about the dubious use of DwC for representing Taxon Concepts/Names.  As far as I know, it was really intended as a transfer standard for observation records??  It contains very limited taxon information!  It really is not a overly difficult job to use a more suitable schema/ontology.  I think the popularity of Darwin Core is due to its simplicity - and I wonder if what Roger is proposing will help with this - ie an XML implementation of the ontology as well as an RDF version.  This will allow people to create very simple XML documents with reasonably simple/flat data, eg an xml document of TaxonName entities, with perhaps 6 or 7 or so key fields - even simpler than DwC.   :-)
 
Kevin
 
From: tdwg-tag-bounces@lists.tdwg.org [tdwg-tag-bounces@lists.tdwg.org] On Behalf Of Blum, Stan [sblum@calacademy.org]
Sent: Saturday, 25 April 2009 6:12 a.m.
To: Technical Architecture Group mailing list; exec@tdwg.org
Subject: Re: [tdwg-tag] darwin core terms inside tdwg ontology

From: tdwg-tag-bounces@lists.tdwg.org on behalf of John R. WIECZOREK
Sent: Fri 2009-04-24 8:58 AM
To: Roger Hyam
Cc: Technical Architecture Group mailing list
Subject: Re: [tdwg-tag] darwin core terms inside tdwg ontology

Anything I should do on the DwC side in anticipation of harmony?

http://rs.tdwg.org/dwc/terms/index.htm#theterms

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

John,
 
At some point, all or (most) of the DarwinCore terms need to be added to the TDWG ontology.
 
But having said that, I also need to say that I'm uncomfortable with:
 
1) The current state of the TDWG ontology (primarily the naming conventions; lets just use terms names), and our understanding of the role it plays in TDWG and how it will be managed (entry of terms, integration of terms into the conceptual [is-a / has-a] relationships to other terms); and
 
2) the fact that the new DarwinCore straddles or overlaps the roles of an ontology and an application schema.
 
I understood the past TAG roadmaps to indicate that we were adopting an approach in which the TDWG Ontology would be a repository for data concepts that are present in (or implied by) TDWG standards; and that real data transmission would be accomplished with application schemas.  The ontology itself would not be a standard, but would be a tool that helps integrate standards.  I thought our standards would be created to function as application schemas or components of application schemas (as in the DwC and its extensions).  I am now pretty confused.  I'd like to hear the rationale for combining taxonomic name/concept with organism occurrence.  I haven't gone over all the existing docs, so apologies if I've missed that, but I think it's confusing that a (new) DarwinCore record could be either a taxonomic name or an organism occurrence, or maybe something else.  Maybe I'm too attached to object orientation and just don't GET the semantic web, but it feels to me like we are stepping into squishy ground.
 
Also, I the the DCMI maintenance procedures are also more appropriately applied to the ontology than a TDWG standard.  The existing process for ratifying TDWG standards and the procedure in the DwC seem to be pretty explicitly in conflict; one can change the other cannot (without becoming another thing).
 
Is anyone else having these same trepidations?  I don't think I've been as much of a Rip Van Winkle as Jim Croft, but I clearly missed some important shifts. 
 
-Stan
 
 
 

Please consider the environment before printing this email
Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails.
The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
_______________________________________________
tdwg-tag mailing list