[Taxacom] iSpecies with Wikipedia

Richard Pyle deepreef at bishopmuseum.org
Thu Mar 27 17:08:56 CDT 2008

> There is another: compromise and consensus settling on the 
> least unacceptable option.  This the approach used by the 
> Australian botanical community in developing the Australian 
> Plant Census.  So far I think we have had to resort to coin 
> flipping arbitration in only a couple of cases out of a 
> vascular flora of c. 20k.  That is not to say there has not 
> been shouting, blood and no-speakies, but it is
> progressing far better than anyone expected it to.   Give it a go -
> put the players together and tell them no-one is allowed out 
> until there is white smoke... :)

It seems to me that this is exactly what I've been advocating -- minus the
arbitrary coin flipping and unnecessary shouting, blood, and no-speakies.
The arguments would shift away from what particular classification to use,
over to what a reasonable mechanism would be to arrive at a "least
unacceptable option".  I think the end-point is the same in both cases; it's
just a matter of getting there via different means. The important difference
in my mind is that I oppose the notion of (sometimes grudgingly) arriving at
one classification at the expense of suppressing others, and instead embrace
all classifications, and allow the "list unacceptable" among them to emerge
from it.

The trick of course -- and the thing I'm very surprised that nobody has
really called me on yet (at least not that my sleep-deprived brain could
pick up on) -- is how you go about developing such a magic algorithm.  I
certainly don't have it, and I don't think anyone else does, either.
However, I bet if you put a handful of experienced but open-minded
taxonomists in a room with a handful of biophilic algorithm specialsits, you
could come up with a good start.  And then you would tweak and calibrate it
against established "least unacceptable" consensus classifications such as
the one you describe.

I'm not suggesting it would be easy -- just not terribly difficult.  But as
I said, the important thing is that it would allow taxonomy-nerds to have
their cake while the rest of the world can be happily eating it -- all at
the same time.

I also missed Paul's comment (Heaven forbid I fail to get the last word on
every contribution to this thread...):

> part of the key (perhaps the most important part) to mange 
> the uncertainty in the example is a back-end robust 
> nomenclator of the botanical type - with 
> homotypic/obligate/nomenclatural names linked together and 
> parent-child relationships. I'm not suggesting this will 
> solve all problems but it often goes a long way.

I completely agree (as I do with almost everything Paul writes).  In my
view, nomenclators should avoid asserting classifications. Rather,
nomenclatural units could/would/should provide the scaffolding upon which
all of the myriad classifications can be cross-linked and compared.

On a final (I hope!) note: it is not my intention to come off as a defender
of libertarian classification chaos and instability; nor as objecting to the
value of a single consensus classification for many/most applications.  My
main point (and the one I began with), is that I see no reason why we can't
have the best of both.


More information about the Taxacom mailing list