If you would like to contact me, you can at email@example.com
or through the contact points on each page which automatically insert a reference.
About this site
This site contains most of my published work
including books, invited chapters,
scientific papers and software
as well as more general topics. It does
items as published, only their early forms to protect
Download what you like.
You may use all or part of these works provided you
acknowledge their author(s) and this site as their
source. The copyright remains with me.
I gave up with Wordpress as, although pretty, it is neither reliable nor secure.
I went back to this old system I wrote years ago which at least has the benefit that it works.
The equation I produced using CodeCogs with grateful thanks.
Selection of recent work up to April 2018:
Conservation of Information in all discrete systems
A complete parsimonious theory using Conservation of information in an
ergodic system to show that all discrete systems will have the same
length and alphabet distribution. This is then demonstrated to very
high statistical significance in the European Protein database, a very
large amount of open source software, a large amount of downloadable music,
a corpus of texts and the abundance of elements in the Universe.
Heiland Lecture, CSM, Golden CO, 07-Mar-2017
NIST Lecture, Boulder CO, 06-Mar-2017
Conservation of Information in Proteins
This is an application of Conservation of Information in the structure
of proteins by Greg Warr and myself. In essence, CoI applies a global
constraint on the action of natural selection by which all proteins in
a species tend to have the same average size and the average size of
a protein depends only on its unique amino acid alphabet including the
action of post-translational modification.
PLOS ONE download
How to build reproducibility packages
Software's Hidden Clockwork
This is a recent result showing that all software systems have
a hidden clockwork which forces them into the same component size distribution
whatever they do, whoever wrote them and whatever technology or
language was used. Just watch as 55 million lines of C, C++, Java, Fortran,
Ada and Tcl-Tk asymptote to the above formula in 1/2 million line chunks.
The flat bit on the left is related to fixed tokens in programming languages.
IEEE TSE preprint
NOTE: December 2016. This theory has been extended to all scales
and predicts that software and protein lengths are actually
distributed identically. This work is under review but
shows with overwhelming experimental support that neither human
volition in software construction nor natural selection in proteins
have anything to do with defining the length distribution of computer
programs or proteins.
Theory paper 1
Theory paper 2
Email Forensics: Eliminating Spam, Scams and Phishing
Get Email Forensics at Amazon.com
Get Email Forensics at Amazon.co.uk