[e2e] evolution of bandwidth as a term

Ted Faber faber at ISI.EDU
Thu Oct 2 15:48:37 PDT 2003


On Thu, Oct 02, 2003 at 05:34:09PM -0400, John Day wrote:
> At 12:32 -0700 10/2/03, Loki Jorgenson wrote:
> > [Evoloution (!?) deleted]
> >Agreed?
> 
> NO!
> 
> The Internet nor much else of the modern world could not have been 
> built without carefully defining and adhering to the definitions of 
> such things as:  bandwidth, mass, joule, kilogram, power, charge, 
> bit, watt, etc.  Getting terms right and sticking to them is crucial 
> for all scientific pursuits, if not intellectual pursuits.  Although, 
> I am aware that certain "disciplines" have made much in the last few 
> years about ignoring such things.  But as far as I am concerned they 
> can all go Fish!

Gentlemen.

I understand and empathize with your desire for clarity of thought and
expression.  I further agree that because signal processing and
networking are closely related disciplines, one must frequently
draw a careful distinction between the bandwidth of a signal and the
data rate of a system sending information using that signal.

However, let's not be dogmatic to the point of stifling metaphor and
economy of expression, both of which are essential for understanding.

Let me make three points:

1) While the language of science is far and away more precise than that
of laymen or soft disciplines, claiming that it is without ambiguity is
disengenuous.  For example, an ounce can refer to voulme or weight; a
tonne and a ton are different; the square root of negative one is j to
engineers and i to mathematicians; and even how much a billion is can
depend on the nationality of the speaker.  Certainly all these can be
resolved by careful specification, but so can signal bandwidth and
information bandwidth.

2) Other scientific endeavors have seen similar semantic drift when an
apt metaphor helped understanding or as a field was beginning to
understand new things.  Ocean currents and electrical currents are very
different things, but no one objects to using the same word to name and
different units to measure them.  That's probably untrue - the good
folks who lay underwater power cable probably have to express their
measures carefully.  Similar extentions have given us Hamming distance,
relativistic mass, and gene splices.

3) The particular metaphor under fire - information rate as bandwidth -
is an enlightenting description in several circumstances and the two are
often directly related.  For a fixed protocol and transmission system,
the two are usually directly related, and therefore straightforward to
interchange.  As a result, one naturally uses whichever representation
is easiest.  One can wind up with "10 lb.s of mass" kinds of
misstatements, but in a broad range of circumstances, there's no harm in
interchanging the quantities.  Done properly, capacity planning this way
can even enhance the understanding of the planners; knowing that every
300Hz is a phone call worth a certain amount of money can make tradeoffs
clear.  I wouldn't be surprised to see some bandwith allocation problems
calculated in dollars.  Why stifle that when no clarity is lost?

To restate, I think people should strive for clairity in their
expression, but it's also important to recognize that language is
inherently ambiguous, especially when ideas are in flux.

-- 
Ted Faber
http://www.isi.edu/~faber           PGP: http://www.isi.edu/~faber/pubkeys.asc
Unexpected attachment on this mail? See http://www.isi.edu/~faber/FAQ.html#SIG 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://www.postel.org/pipermail/end2end-interest/attachments/20031002/a50e7a32/attachment.bin


More information about the end2end-interest mailing list