In response to Jim Campbell:
The whole issue of signal versus noise is critical to the effective
operation of learning systems.
In biology we have genes that are a chemical method for encoding
information and transmitting it from generation to generation. The
mutations (and other changes resulting from a variety of reproductive
approaches) that occur in that encoded information is responsible for
evolution. These mutations occur at random -- in essence, they are noise
that in one sense "degrades" the information that is being transmitted.
Generally, the rate of mutation is fairly low -- in genetic algorithms it
is typically 5% or less.
If we were to look at this noise we may try to reduce it (eliminating
error) in an effort to ensure that the information is passed accurately
from generation to generation.
Yet there is a paradox in that this "degradation" can improve the chances
of most (but not all) of that information surviving for many generations.
This is because an organism that is able to adapt will often survive with
a significant portion of the original genetic information intact.
In the area of information and ideas, we have the concept of "memes".
These can been seen as paralleling genes in the world of biology. I
believe that much of what we see in the realm of biology can also be
applied in the realm for information and ideas.
A certain amount of "noise" is critical to the transmission of information
in a learning organization. Without that randomness, organizational
learning would not occur. Rather, information and ideas will remain static
-- identical copies that never change -- and ultimately the organization
would die.
Brainstorming is a classic example of memes and "noise". The idea behind
brainstorming is to toss out an idea. Then an associated (mutated?) idea
is tossed out. From that mutated idea, yet another mutation occurs until
something entirely different from the first idea emerges.
One of the problems with modern organizations is that they attempt to
eliminate the "noise" as much as possible from all work processes, from
products, and from the people involved in the organization. I've been at
plenty of meetings where, for the sake of efficiency, an agenda has been
developed and tightly adhered to in order to keep the meeting "focused"
when what was really needed was more "noise".
The drive for efficiency (which we see in such things as business process
reengineering) is removing noise from the system. That is not necessarily
bad in itself, so long as we remember that somewhere we need to include
the opportunity for "noise" to interfere with the "signals" that are being
transmitted within our organizations.
-- Regards, Peter von Stackelberg Applied Futures, Inc.