2009/9/11 Bob W <[email protected]>: > I don't think this is necessarily a mistake in principle, and I don't think > your example is a mistake at all. The problem lies in the way people > (mis)use or misunderstand hierarchies. I recommend David Lorge Parnas's > essay "On a 'buzzword': hierarchical structure".
Thanks Bob, I'll try to google it. > Livestock and Hoofed animals are intersecting sets - some animals genuinely > belong in both. A hierarchy which includes Hoofed animals sounds to me as > though it is vaguely Linnaean - hooves indicating common descent - whereas a > hierarchy which includes Livestock suggests that it is based on the way > humans make use of the animals. It certainly doesn't indicate common > descent. So in the example given, the error is yours for expecting LR to > recognise this. It was a silly example. I couldn't think of anything better just then. But my point was exactly what you write, intersecting sets. > The lesson to be learnt is that you must think carefully about whether these > are genuine hierarchies, or the same type of hierarchy. Try to make your > hierarchies represent the real world. I try to think in terms of sets and > subsets - set theory is the secret of the universe - and recognise the > distinction between hierarchies and other types of relation. It's not possible to set up a single taxonomy for all keywords because they'll group differently according to context. Managing the taxonomies _and_ the photos can be either just a laissez-faire or a complete nightmare, depending on how you need to use those keywords. The way Dave Brooks and Tim Øsleby describe their keywording, for example, is blessedly simple and I'm sure it will work perfectly as a mind-map and retrieval aid. Keywording pictures for _others_ to retrieve the photos is a different matter. Magnum is in a special position because they carry enough authority to bend the customer to their own thinking. For microstock it would be the other way around. The agency and the photographer would have to design their taxonomy to maximise the convenience for the customer. That demands a lot more management of the taxonomy. Personally I hate this kind of stuff and have got it wrong two times already in other archive software, so I'd hoped to get it right from the beginning when starting with LR. :-) > By some peculiar coincidence, I was thinking about animals' feet during some > idle moments at work this afternoon. My office contains people with a number > of different eating taboos. People who don't eat beef, people who don't eat > pork, others who don't eat horse, rat, badger or dog. The only mammals that > everyone seems to eat are sheep, goats and deer of one sort or another, and > I think they all have the same sort of feet, don't they, which differ from > the controversial animals. They're all quite closely related aren't they? Or > am I just wasting company time thinking like this? Why don't people eat > badgers? In two words: religious superstition. :-) Jostein -- http://www.alunfoto.no/galleri/ http://alunfoto.blogspot.com -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
