(possibly spam: 6.9576) [JPL] Programming limited to name recognition

Eric Hines EHines at message.nmc.edu
Fri Jul 2 16:41:59 EDT 2004

The Good Doctor wrote:
I recently received this reply from a station to one of my "weekly

"We're not a new releases format, and artists who have no name
or connections to (our) listening area are a hard sell for our

I thought the purpose of programming a station was to bring to the
the best music available.  I have a problem with grasping the logic of

limiting the selection to artists that you believe your listeners
know, or that have geographic roots, or to previously released

Then me:
We are a "new releases" format here, and I generally agree that playing
new stuff is a good thing.
But I'd argue this from a different perspective: The purpose of my
programming is to present great music that will garner listeners and
supporters for the radio station. So for me there are three criteria any
programming has to meet: it has to be enriching to the people who do
listen and to the community I serve; it has to attract a fair number of
listeners; and, at the end of the day the programming should put the
station in a position where I can balance the books and even improve
things a bit every year.
My argument with only playing artists everyone already knows is that a)
it's boring and far less than you can be doing in terms of service; and
b) in the long run, this sort of "caveman-esque" application of focus
group data will lead to smaller audiences and less support. 
Why do I say that? Well, look at commercial radio, where they've been
ruthlessly applying the lessons of focus group data for years.
Listenership is going down. The excuse for this is that people's
lifestyles are changing and radio doesn't fit in as well as it used to.
That's crap. The reason commercial radio programming is doing badly is
because it's harder and harder to listen to. In other words focus groups
are a bad way to find out what people will actually want to listen to
over the long haul.
Focus groups are more or less the entrail reading of radio programming.
They're put forward as "objective data", but finding a marble in the
goat entrails is objective data, too. Just that there's absolutely no
connection between the marble in the goat entrails and whether or not we
should invade Belgium.
And I'd say there's far less connection between what a focus group dial
turner seems to tell you and what that person will listen to on his car
radio every morning than the consultants tell you.
The best objective data set we have about listening preferences is NOT
predictive, it goes to outcome: Ratings.
The ratings say commercial programming methods don't work in the long
run, in spite of consultants' constant repetition of the word
So how do the most successful public music stations choose their
programming? I doubt if they rely exclusively on arbitrary rules like
whether or not someone is already famous. And I doubt that the
subjective judgement of programmers is entirely absent from the process.
And I bet that having some interesting new stuff in there is essential
to the mix.
(Below is an essay by Roy Williams, a very successful ad man on this
general topic: that is why it is important to be creative and original
in your appeal, and not just to apply marketing data like a caveman
might apply his club.)
Chaos: a Higher Level of Order A Monday Morning Memo for the Clients
and Friends of the Wizard 
You hope to attract and hold the attention of another. Your goal is to
fascinate their mind. You are a teacher, a minister, a romantic suitor
or possibly an advertiser. Is there a successful model for attracting
and holding human attention that you can study? Indeed there is.

Humans are attracted by ocean waves, clouds, mountains, lightning and
logs burning in the fireplace and we think that snowflakes are
beautiful. We detect an elegant order in each of these that our minds
cannot fully grasp. This marvelous order is known to science as Chaos,
and it is beyond the ability of the mind to predict. (Predictability,
you will recall, is the mortal enemy of persuasion because
predictability triggers boredom. Remember Broca? Hmm... this could be

Using Business Problem Topology, let's begin by looking at the defining
characteristics of Chaotic systems: 

1. Chaotic systems are deterministic; they have something determining
their behavior.

2. Chaotic systems are very sensitive to their initial conditions. A
very slight change in the starting point can lead to enormously
different outcomes.

3. Chaotic systems may appear to be disorderly, even random, but they
are not. Beneath the random behavior is a pattern of elegant order.
Truly random systems are not chaotic.

If you consider that mapping a chaotic system can easily require tens
of millions of notations, you can easily see why it was only after we
had invented computers that we were able to see these patterns that had
previously been too big to comprehend. 

Using chaotic mathematical equations, computers today are producing
images that look exactly like the beauty we see in nature. Amazingly,
these chaotic maps also seem to mirror the behavior of the stock
exchange and population fluctuations and chemical reactions. We call
such Chaotic maps "fractals." Examples of natural fractals are clouds,
coastlines, lightning and mountains. 

Benoit Mandelbrot, a scientist at IBM, created additional fractal
images by mapping the variations in:

1. stock market prices,

2. the probabilities of words in English, and

3. the motion of turbulent fluids, because each of these has a pattern
that is slightly beyond our human ability to predict.

Are you now beginning to understand why people are drawn to: 

1. playing the stock market,

2. listening to the combination of rhythm and words in poetry and song,

3. staring at the sea?

Roy H Williams


Eric Hines
General Manager
1701 East Front St.
Traverse City, MI 49686

More information about the jazzproglist mailing list