KPCOFGS

[wzdd]

Is the hierarchical classification scheme for organisms really useful any more, or are people just doing it for fun?

Comments

[roslyn]

Well - yes. New species are still being discovered all the time, and separating them into little drawers (literally, figuratively) makes them easier to organise. Even if it does become a bit bureaucratic. Plus it makes naming easier. :)

[wzdd]

It seems like it's gone crazy! Check them all out here: http://en.wikipedia.org/wiki/Taxonomic_rank

What about if instead we just allocated all creatures a unique ID and tagged them with interesting identifying features, such as "canine teeth" or "lays eggs" or "pentadactyl limb"?

Then we could provide links to their closest relations, providing some kind of web of life.

(I haven't had much sleep)

[roslyn]

I can't tell if you're being facetious, but on the assumption that you're not, check it out!: http://tolweb.org/tree/

Also a new discovery: http://species.wikimedia.org

ALSO it is such a nice change to read about something that isn't economics! Even if hyperinflation and Icelandic bankrupcy and the catastrophic failure of capitalism are among the most interesting things ever.

PS. Did you like my bureaucracy pun.
PPS. The answer is "yes".

[wzdd]

I didn't get it. Now I do. Yes, that's very funny. :-)

I was being a bit facetious but I do think that something that indicates species by closeness to other species rather than imposing a hierarchy (i.e. a web rather than a tree) would be a better fit nowadays. It seems weird, for example, to split sea squirts off the tree so early just because they have a backbone but are otherwise nothing like other chordates. But IANAB. Wikispecies looks pretty cool.

The trouble with the Iceland articles is they're always accompanied by fantastic pictures of Iceland (eg people bathing in warm pools surrounded by snow) which makes me think more about visiting Iceland and less about the terrible economic situation over there. If they just showed people looking unhappy, or something, I'd be more able to focus on the issues.

[jasongrossman]

Er ... the point is that it's unbelievably helpful to know which species are descended from which other species. A web won't tell you that.

Keyword: cladistics.

Not like me to defend the scientific orthodoxy! And yes, there are lots of things wrong with the details.

[wzdd]

Thank you for the keyword! Cladistics seems much more sensible than Linnaean taxonomy. It doesn't seem to be the be-all classification system, though -- you might want to talk about species that have a particular morphology for some reason, or perhaps species that were subjected to similar evolutionary pressures. You could do this if each organism was a leaf in a directed graph.

[jasongrossman]

True

[wzdd]

Sooo if I wanted to introduce any interesting classification systems (beyond putting a cladogram and the Linnaean trees on the same graph, since they seem be kinda-sorta-mostly talking about the same leaves) I'd have to use, in addition to nodes-which-are-species, nodes-which-are-bigger-than-species and nodes-which-are-smaller-than-species. But (*shakes sleeve*) that seems okay. I mean cladograms (if I've got this right) also have nodes-which-are-bigger-than-species in order to have branches in the graph

[couldtry]

I enjoyed reading this thread.