The thing that concerns me about 3a is the conditional definition of scientificName (i.e., it excludes authorship when that information is pre-parsed, but includes it when it's not pre-parsed).
right. That is where we are today. We need to test the contents every time. One thing we are developing requirements for is a DarwinCore Archive "Normaliser" which would be a web service/web app client that can accepted a DwC-A as input and would output the same data as a new DwC-A that conforms to a set of rules based on those requirements. So one thing would be to parse the names into the authorship field for simpler consumption.
The thing that concerns me about 3b is that scientificName is (supposedly?) required; so if someone with unparsed information provides scientificNameWithAuthorship, then are they supposed to leave scientificName blank?
The short answer is yes, they would leave it blank. if they could parse it then they would have put the parts into name and author elements in the first place. I also think the inverse is true. If it is already split and they put it into name and author fields - they won't concatenate and put a new, merged copy into a name+authorship field.
In regard to the requirement confusion - either we keep dwc:scientificName alone and have a simple requirement test or we add a new term and make it a conditional requirement. My point was, if we add a new term, then the set (scientificNameWithAuthorship, scientificName, scientificNameAuthorship) is more clear than (canonicalName, scientificName, scientificNameAuthorship)
DR
Rich