As written before, I believe it would be beneficial to have a common TDWG publication standard. It should be both relatively simple (simpler than MARC, and hopefully also than MOD, albeit the latter is already available as a standard) and functional enough to provide a simple "replacement" or "Proxy" functionality in the absence of a true external database. Essentially, integrating semi-publication/reference managers is what almost any biodiversity product from Specify to Linneaus to BioLink does.
I here argue against the approach taken by Taxon concept, fixing the schema to a proprietary and rather strange software product. The EndNote schema used is not any kind of standard, even the company producing EndNote also has two other reference manager products which use different and incompatible data models. EndNote changes the semantics of fields dynamically according to the record type, which makes it difficult to support. The same (but with different fields and record types) approach is taken by ReferenceManager. While it certainly makes sense for these products, it makes any reuse or sharing of data almost impossible - which may be part of the motivation producing the software.
There is a discussion in 
http://efgblade.cs.umb.edu/twiki/bin/view/UBIF/ProxyDataPublication on the choice, which largely suffers from lack of ongoing criticism. Please do help us by arguing in favor of the TaxonConcept Publication model, or argue what is wrong with any of the options proposed in UBIF!
Gregor Hagedorn -- 12. August 2004
Yes, please continue this discussion. In some respects our Publication structure is a place holder until there is a standard (and until GUIDS make an impact). However, we believe the elements/fields that are cribbbed from Endnote can probably hold most data that might be in existing databases. They are just Text fields with explicit Element names, this does not tie them to either Endnote's dynamic datamodel or its application. TrevorPaterson 12 Aug