[tcs-lc] Name-Name Relationships vs.Concept-Conceptrelationships

Richard Pyle deepreef at bishopmuseum.org
Wed Sep 21 12:40:27 PDT 2005


> Yes but....if we have the generic relationship approach to TaxonNames
> and I added a concept type of relationship to the enumeration of
> relationships allowed I would be worried that people would start marking
> up names where they should be using concepts... and we'd never improve
> our data...

...which is one of the reasons why I struggled (unsuccessfully) to convince
everyone that "TaxonName" data should be embedded within "Nominal"
TaxonConcept instances (rather than become a top-level "object").  The only
real reason I can think of for why this would not be the best solution
overall, is that there may be some limitations on enforcement within XML.
But given that the schema is riddled with examples of where we "just have to
trust the data provider to conform to the standards of practice", this seems
like a small

> We have tried to investigate the issue of relationships as I mentioned
> before - if some other taxonomists wants to find what's missing that
> would be great.

We'll find them when we thrown big datasets at the schema.

> How often does the code change at that sort of level?
> I'm sure there will be many versions of all standards and
> implementations before the code changes this sort of thing - but I could
> be wrong...

Well....check out this article in tomorrow's issue of Nature:

http://www.nature.com/nature/journal/v437/n7058/edsumm/e050922-08.html

This article has *HIGH* relevance to TaxonNames, TaxonConcepts, GUIDs, etc.
A more technical follow-up article is being drafted as we speak
(err..write).

Aloha,
Rich




More information about the Tcs-lc mailing list