Re: information transmission

t (
Thu, 11 Mar 1999 11:44:39 +0100

From: <>
Date: Thu, 11 Mar 1999 11:44:39 +0100
Subject: Re: information transmission

Bruce Edmonds wrote:

> Shannon's definition of information was invented for a very specific
> purpose, it is *not* a good definition for information in general.
> I define information as such:
> "Anything uniquely identifiable that can be duplicated with no
> lower bound on the cost"
> This means that although information is always held by physical tokens
> of some kind, it is necessarily independent of these tokens. As
> technology improves this paragraph can be duplicated using less and less
> energy, but this paragraph is still identifiable as the *same*
> paragraph.
> Thus information can, of course, be transmitted _using_ tokens. If I
> send an email I can trace the causal chain of tokens that were the
> mechanism for the transmission of information. But the information is
> not the token.
> Take an analogy. I may have a desiese, this may be tranmitted via a
> virus. However the desiese is not the virus, it is a macroscopic
> collection of syptoms etc. that is caused by the virus. The desiese can
> be passed on to someone else _via_ the virus. So with information.
> Regards.
> --------------------------------------------------
> Bruce Edmonds,

I previously posted my favourite 'definition', one of Gregory Bateson:
Information is 'A difference which makes a difference'. This includes the
context sensitivity of the meaning of information. (Information is a
concept, it is not some'THING');

Take a gene and its allele, which differs by a single mutation. This
mutation may be a silent mutation, which means that the resulting amino acid
and thus the resulting protein of both wild type gene and mutant allele will
be identical: we have here a nucleotide difference which does not make a
difference at the cellular functional level. This nucleotide difference is
not informative.
Take the same two: wild type gene and mutant allele. Now they are studied by
a biologist trying to infer phylogenetic relationships between organisms and
discovering that there is a point mutation between the two sequences. In
this case the nucleotide difference is a difference which makes a difference
and it is information.

After looking for years for a useful definition of what information is, this
concept of Bateson is the most convenient thus far for me. Shannon's
definition is more about how to package information, it is not about the
'essence' of the concept.


Mario Vaneechoutte
Department Clinical Chemistry, Microbiology & Immunology
University Hospital
De Pintelaan 185
9000 GENT
Phone: +32 9 240 36 92
Fax: +32 9 240 36 59


Symposium 'Water and Human Evolution'. April 30th , Ghent, Belgium
Information at:

This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)