Bringing up a database of species is one thing:-|.

But what about setting op two pools of data: Objects on the one hand, and keys on the other.

For objects I’d say the Wikispecies-I may serve as (one of) the startingpoint, if objects get extended with properties describing them (e.g. Not just Paris quadrifolia, but also data on flower, leaves, biometrics…). I explicitly don't refer to them as species (because then the matching with a key -see below- apparently was already done, and my suggestion is to draw a line between these two) 

For keys I think a whole new Wiki-project needs to be created, e.g. Wikikeys, consisting of key-engines that can be elaborated by Wiki-users.

So on the one hand we have the collectors (filling Wikispecies, but as far as I’m concerned it may also apply to stamp-collecters…), on the other hand the classifiers, developing keys (in Wikikeys) to classify those objects.

And now comes the hard part: For Wikikeys to work it should be construct where the key-developers enter (sets of) rules (may even be programs) that can evaluate into a likelyhood of Yes or No. E.g. if item’s property flower-colour = Red, offering this specimen to the Flora-dichotomous-Linnaeus-key will encouter it with a rule on flower-colour which rules out Paris quadrifolia for 99%

I can elaborate on this but it may be usefull to first stick to this, check how much sense I made and if there’s anybody interested to pick this thread up.

I'm new to wiki, and please let me know if I have to address other people.

Cheers

Wim