[Taxacom] Usefulness vs. convenience (Protista)

[Stephen Thorpe]

realistically, I doubt any two scientists definitions of "science in itself" 
would agree perfectly. You can't take the man (or woman) out of the scientist,  
and that complicates matters. But, surely you can grasp that if what you want to 
do is to organize and make accessible biodiversity information, the 
classification that you should choose has at least as much to do with 
information management as it has with "science", and if you have to file 
something in your "biodiversity filing cabinet", it really doesn't help if 
scientist X insists that you file it under P because he/she has/claims to have 
80% support using his/her methods that it should go there, whereas scientist Y 
insists that you file it under A because he/she has/claims to have 85% support 
using his/her methods...




________________________________
From: Richard Zander <Richard.Zander at mobot.org>
To: Stephen Thorpe <stephen_thorpe at yahoo.co.nz>; Curtis Clark 
<lists at curtisclark.org>; taxacom at mailman.nhm.ku.edu
Sent: Tue, 21 December, 2010 1:00:48 PM
Subject: RE: [Taxacom] Usefulness vs. convenience (Protista)


Again we have a little problem. “80% bootstrap support” is not science in 
itself. It is a measure of the apparent certainty of a cluster (where 83% 
bootstrap is about 95% posterior probability from my simulations, depending on 
branch length). We use the theory of evolution to make scientific inferences 
about descent with modification of taxa. “80% bootstrap” is about the evidence, 
not the theory. 

 
You say you have difficulty in understanding what I’m saying? Well, it’s partly 
me (editors have twice independently labeled my imperishable prose as “dense and 
obscure”), but also partly the structuralist box of “tree thinking” you can’t 
quite break out of.
 
Bioinformatics naturally has a problem with reporting/recording the results of 
science. The results change and are commonly controversial. A theory of the 
macroevolution of a group is a result of science. Well-supported patterns of 
EVIDENCE of macroevolution are not terribly controversial, but that is all they 
are, evidence. _Of course_ phylogenetics fits bioinformatics better than that 
messy, controversial science stuff.
 
* * * * * * * * * * * * 
Richard H. Zander 
Missouri Botanical Garden, PO Box 299, St. Louis, MO 63166-0299 USA 
Web sites: http://www.mobot.org/plantscience/resbot/ and 
http://www.mobot.org/plantscience/bfna/bfnamenu.htm
Modern Evolutionary Systematics Web site: 
http://www.mobot.org/plantscience/resbot/21EvSy.htm

________________________________

From:Stephen Thorpe [mailto:stephen_thorpe at yahoo.co.nz] 
Sent: Monday, December 20, 2010 5:22 PM
To: Richard Zander ; Curtis Clark; taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] Usefulness vs. convenience (Protista)
 
well, it is a bit difficult to know exactly what you are saying sometimes 
(perhaps that's just me?), but I was objecting to what I thought you were 
saying, which was something like "artificial (=allowing paraphyly) 
classifications are not science and so nobody (scientist or bioinformatician) 
should use them". Maybe you aren't saying that? I agree that it would be bad for 
scientists to pass off artificial classifications as science, but I haven't 
really thought of that as a problem. On the other hand, you seem to reiterate 
that classification should go hand in hand with your conception of "science", 
but I say that would make bioinformatics too unstable to be useful. If you want 
to do "science", then do science, but I wouldn't say that people doing something 
slightly different, such as biodiversity information management, must follow 
your lead. "80% bootstrap support for clade X", or whatever, just isn't really 
that relevant to their activities ...
Stephen