>> well, currently, that means that we have to break up what is a simple single >> archetype otherwise into a set of archetypes, and we have poor binding >> between >> them. > > I don't think Thomas was suggesting multiple archetypes. I think he was > saying that you would have multiple data instances of the one OBSERVATION > archetype.
That's what I thought he meant. And that would mean that we'd need to take what is one logical and actual archetype now and break it into several parts to allow this to happen. Grahame
