Message 2004-10-0060: Re: Fwd: Re: Panstems

Tue, 14 Sep 2004 14:33:07 +0100

[Previous by date - Apomorphy-based clades; was Re: Panstems]
[Next by date - Re: Fwd: Re: Panstems]
[Previous by subject - Re: Fwd: Re: My classification of coelurosaurs]
[Next by subject - Re: Fwd: Re: Panstems]

Date: Tue, 14 Sep 2004 14:33:07 +0100
From: [unknown]
To: phylocode@ouvaxa.cats.ohiou.edu
Subject: Re: Fwd: Re: Panstems

> Date: Mon, 13 Sep 2004 13:42:52 +0200
> From: Michel Laurin <laurin@ccr.jussieu.fr>
>=20
> =09I would like to point out that my greatest objection against
> this Pan-naming scheme is that I know that several of my colleagues
> that are respectable and reasonable systematists, that came to the
> "First International Phylogenetic Nomenclature Meeting" or that
> could consider using the PhyloCode, are very much opposed to it.  W=
e
> risk loosing a large basis of support among systematists if we
> implement an idea.  Let's face it, the PhyloCode faces a long, hard
> uphill battle for acceptance in the systematic community.  Let's no=
t
> drastically reduce the chances of the PhyloCode by introducing an
> unnecessary clause into this perfectly good code.
>=20
> =09Finally, I also note that the PhyloCode is mostly about how to
> define names, not how to name clades; the Pan-naming scheme in this
> respect would differ from most of the other clauses of this code
> that only set minimal requirements for the names (like the use of
> the Latin alphabet).

I've stayed quiet through most of this discussion, but I would just
like to agree wholeheartedly with Michel Laurin's comments here.  Eve=
n
if the problems raised by the putative Pan-naming convention are
solvable, these problems will have been raised and addressed in an
area that is not really within the PhyloCode's remit.

My background is in software engineering.  In the development of
software, the absolute immutable number one rule of design is that
each tool should do exactly one thing, and do it well.  When a
software tool starts to suffer "feature creep" -- the addition of
features unrelated to its primary task -- that is the death knell for
it as a simple, comprehensible and useful component that can be used
in the context of larger designs.  This is true _even when the
specific features added are useful ones_.

The application of this should be clear.  The PhyloCode is a fine
system for producing clade definitions.  So long as it doesn't
overreach itself (by making nomenclatural stipulations as well as
definitional ones) it ought to be, and probably will be, widely
adopted.  But if it sprawls into other areas, then it will surely fai=
l
to have any impact on the wider world -- for two important reasons.

First, the code itself needs to be a short enough and focussed enough
that people can actually read it all and understand it in a reasonabl=
e
period -- say one solid evening's reading.  If the prose of the code
becomes bloated, then people will not do that; and what they don't
read, they can't use.

Second, and more important, the PhyloCode's greatest utility and
applicability are to be found when it is used as a part of a wider
framework of phylogeny reconstruction, taxonomy and nomenclature.  By
keeping it "lean and mean", we maximise its usefulness.  But if we le=
t
it intrude on these separate but related areas as well as stipulating
how to define clades, then adoption of the PhyloCode becomes an
all-or-nothing affair; and experience suggests that the great majorit=
y
of workers, faced with such a choice, will choose nothing --
preferring to stick with what they know.

Instead of a clade-naming rule or recommendation, I commend a simple
suggestion that has been made on this list (perhaps by Mike Keesey?  =
I
don't recall).  When the name of a clade is written or spoken in a wa=
y
that might be ambiguous, it should be disambiguated simply by saying
"crown" or "panstem" first.  Thus we might say that _Confuciusornis_
is a member of "panstem Aves", but not of "crown Aves".

In conclusion, we should remember that the classic military mistake i=
s
to fight on more than one front.  In the PhyloCode we have a system
well able to prevail on the single front of clarifying clade
definition.  Let's fight -- and win -- just that battle.

Thanks for listening.

 _/|_=09 ____________________________________________________________=
___
/o ) \/  Mike Taylor  <mike@indexdata.com>  http://www.miketaylor.org=
.uk
)_v__/\  "Don't Live with Broken Windows" -- Tip #4 in Hunt & Thomas'=
s
=09 "The Pragmatic Programmer"

--
Listen to free demos of soundtrack music for film, TV and radio
=09http://www.pipedreaming.org.uk/soundtrack/


  

Feedback to <mike@indexdata.com> is welcome!