RE: Gaia theropod follow-up: a "new" phylogeny

["Ken Kinman" <kinman@hotmail.com>]

Tracy,
    In all fairness, I think cladistic analyses done well are one of our 
most important systematic tools.  And generally the more characters a good 
systematist has, the better the result will be.  But it isn't just a numbers 
game, and quality of both character selection and analysis play an important 
part as well.
    Like any other segment of society, there are going to be some slackards 
out there, cutting corners and taking whatever the computer first spits out 
at them.  There might even be a few really bad apples who set out to achieve 
a certain end and manipulate the data until they get it (just like 
statistics, cladograms will say anything if you torture them long enough).
    But I don't think you are going to find such bad apples on this list, 
and most such people never make it out of graduate school, especially in a 
discipline like systematic biology.
    Purposeful manipulation of data towards a preconceived result is 
obviously a bad thing, but you can't therefore conclude that this makes 
"tweaking" and testing bad.  In fact, the other extreme of no tweaking and 
testing of results would be rather unscientific.  The best "tweakers" no 
doubt learn from experience how much is too little and how much is overdoing 
it.  In that way, it is probably just as much an art as it is a science.
   However much I might dislike the automatic conversion of cladograms into 
classification (and nagging at the strict cladists about it), I believe 
well-done cladistic analyses and cladograms are now indepensible to good 
systematic work and those who do it well should be applauded for it.  I 
suppose it could be possible to have too many characters (and diminishing 
returns set in), but I don't think I've ever seen that happen to date.  And 
it hard to imagine that having too many would be worse than having too few.  
That's my two cents worth.
                   Ken Kinman,
                      an Ashlockian cladist
********************************************************
> One thing I don't do (and which good systematists don't do) is recode and
> recode until I get a tree I like.  <<
> That's what I'm talking about. Say the person using Paup doesn't like the
> out come of the tree, THEY tweak it until it is something THEY like. Then 
> WE
> don't know if the person was right or the computer was right in the first
> place. If they always do that then how can we know (as George as said in 
> the
> past) which is right? It's like having a chocolate granola bar, it defeats
> it's purpose.
> Tracy

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 
http://profiles.msn.com.