Date: Tue, 30 May 2000 16:02:43 +0100
From: Chrystopher Nehaniv <c.l.nehaniv@herts.ac.uk>
To: Barry McMullin <mcmullin@eeng.dcu.ie>
Subject: Re: ALife7/Evolvability Workshop...

Dear Barry,

I am pleased to accept your submission "Some Remarks on
Evolvability and Genetic Relativism" by Barry McMullin" subject
to revisions as outlined in the enclosed referee report(s).  I
suggest you mention von Neumman self-reproducers in the title to
help quickly orient the reader.

Please note that workshop proceedings papers are limited to FOUR
PAGES.  Suggestions for how to achieve this are made by referee 2
(see below).

Additionally, we would like to include this or a longer version
of your paper in a University of Hertfordshire Technical Report
arising from the workshop for which a greater length is fine.

Please send me the revised version(s) of your paper to arrive in
in electronic form by 10 June 2000 to C.L.Nehaniv@herts.ac.uk

Best Regards,

Chrystopher Nehaniv
Program Chair "Evolvability Workshop", Artificial Life 7
http://www.cs.herts.ac.uk/~nehaniv/EVOLVABILITY.html
-------------------------

Referee's Review:

The paper is relevant and interesting for the evolvability
workshop and participants. It relates von Neumann's
self-reproducing CA work with issues of evolvability and
flexibility of genotype-phenotype relations.

The writing needs some clarification and explanation to make this
understandable as a stand alone paper. Since you are currently
under the 4 page limit, you should have room to expand the paper
to make it more readable:

Abstract: Please add an abstract summarizing the results and key
issues of the paper.


p. 1 col 1:

Remind the reader of von Neumann's work by quickly explaining the
components of his self-reproducing automata, including the
notation you introduce in the 2nd paragraph.


You discuss "mutation".  I and most readers are unaware that von
Neumann considered mutations of his self-reproducing automaton.
This topic should be introduced.  - Are mutations allowed in a CA
?  - Could these affect the "genotype-phenotype map" ?

column 2:

"Working with such a framework, it is necessary to exhibit"
^^^^^^^^^^^^^

For what purpose is it necessary? clarify.

p. 2 column 1:

"we are not restricted to considering the range of complexities
of any single set of self-reproducers, but can include the union
of the sets":

It would be better to insert "with a fixed language" (or
something to that effect) after "self-reproducers" since the
union of sets of self-reproducers is still a (single) set of
self-reproducers.


next paragraph:

"The only benefit of any sort in this approach..." :

Another benefit is that the approach is more biological tenable
from the viewpoint of the evolution of genetic systems!

next paragraph:

"von Neumann's problem, as I have stated it" 
                       ^^^^^^^^^^^^^^^^^^^^

I couldn't find where you have stated it, although you referred
to this on p. 1, here, and in the next column.  Please do state
it!

column 2

spelling: automataon -> automaton

consistency: turing or Turing (you use both spellings for turing
machine and Turing computatble, it seems that it would be better
to pick one)

p. 3

Codd quote: insert a space before "Codd (1968.."