I'm taking a short break from simulation and matching work to post a
couple of messages stimulated by recent discussions and web pages
visted about Gross Progress Indicators and about population.
I started out to address those topics, but instead wrote a preamble
discussing a related question. For now I'll just post this preamble,
and will get to the actual amble itself next message.
On my web pages I talk about "true bandwidth" or "net baud rate" by
which I mean the amount of actual communication on a channel between
two people, as opposed to the number of bytes sent and received. I
won't worry here about the technical details, but there is a major
concept here that everyone should be aware of.
Suppose it takes you one minute to read a 6000 byte message -- that
would be 100 bytes a second or 800 bits per second, roughly 800 baud.
But that is a gross baud rate, measuring only the flow of bits from
screen or page to your eyes. What I'm trying to communicate to you
here is the idea of "net baud rate", the actual amount of
communication taking place.
For example, suppose you can only read and write in an unrelated
language that uses the same character set, such as Malay, or Finnish.
You could still run yours eyes over this page and see each character
of the text, but there would be no actual communication taking place:
a net baud rate of zero.
Personally I'm not very good at communicating with people, and I
expect that for many of you, even though you can read English
perfectly, the net baud rate for this text is very low, near zero. In
part this is because of my deficiencies as a writer, but mostly its my
choice of topics and underlying assumptions.
If I to was write a cheerful account of my last trip to Vancouver
mentioning only the places visited and people seen, the net baud rate
would be much higher.
I can think of a few people out there who know me and my favourite
topics quite well, people who can read most of my text quickly and
still understand it almost perfectly. But for those people the net
baud for a piece of text like this one is much lower than you might
think, because the people I have in mind already have some
understanding of what I am saying here. In information theory
'information' is really 'news', what is new, and to people who know me
this text is "just more of Wilson's crazy ideas" -- not news.
The maximum net baud rate for this message would be for someone who
has a good understanding of English, an interest and understanding of
technical matters, BUT has never heard of or seen these ideas before.
Or we could turn that around, and say that if one of you writes a
message that I can understand easily, but which contains a lot of new
ideas I've never encountered before, then the net baud rate for me
reading that message would be quite high.
What I most want to do is to increase the net baud rate for all the
communication I partake in. I want to send my messages to people who
will understand them and will find them interesting and newsworthy.
And I want to receive messages from people who will write stuff I can
understand, and be interested in, and will find newsworthy (new,
novel, etc.)
Your own motives for reading and writing messages are unknown to me,
but I can guess that what I just said in that last paragraph holds
true for you as well. Doesn't it? Don't you also want to increase or
maximize the net baud rate of your communication? Think about it,
please!
One of the ideas that I have been working on for many years involves
network optimization to maximize the effective or net baud rate of all
communication on a network (the "net net baud rate", so to speak).
I've written both about doing this for messages flying about the
internet, and for interpersonal communication on what I call the
social network, the network of interacting human beings, mediated
mostly by speech.
For anyone who has read my web pages or earlier messages, it won't
come as any surprise (news) to read that this network optimization
problem is a combinatorial optimization problem, and that I propose to
solve it by matching people to each other based on personality,
interest, and education profiles. But let me just emphasize here that
this is a very solid plank in my platform -- I have written a great
deal of nonsense over the years, but I think this one point is
completely solid, susceptible to mathematical proof.
Of all the things I have to say, if there is only one thing you take
seriously let it be this:
We can maximize the global net baud rate for interpersonal
communications by using combinatorial optimization to match people
based on personality, interest, and education profiles.
I'd appreciate your comments on this, whether you agree or not. More
important, perhaps, I'd appreciate your advice on who might be
interested in what I have to say. Many of the people reading this
will be taking it in at a low net baud rate for one reason or another,
lacking either appropriate educational background or interest in the
topic, or perhaps being already familiar with the ideas. But even so,
you could probably recommend a person or a mailing list of people that
might absorb it more completely. If so, please let me know!
dpw
Douglas P. Wilson [EMAIL PROTECTED]
http://www.island.net/~dpwilson/index.html
