Re: The embarrassment of American broadband

[David Farber <dave () farber net>]

Begin forwarded message:

From: "David P. Reed" <dpreed () reed com>
Date: April 28, 2009 11:12:09 AM EDT
To: dave () farber net
Cc: ip <ip () v2 listbox com>
Subject: Re: [IP] Re:   The embarrassment of American broadband

Here's why American "broadband" is an embarrassment.  Ignorance at all  
levels.  Ignorance piled on ignorance.  Even those who claim expertise  
seem to want to add to the ignorance... Lest we get confused by  
another incorrect statement being added to the IP fray:

Brett Glass wrote:
> Actually, as an engineer, I understand that in truth NO service  
> except cable modem service is truly "broadband." ("Broadband" -- the  
> antonym is "baseband" -- refers to unidirectional and/or  
> bidirectional communications which are done via frequency division  
> multiplexing over a shared, wideband medium with multiple taps.)
As a technical term, broadband is not the opposite of "baseband", nor  
does it have anything to do specifically with multiplexing, FDM or  
otherwise. Broadband is the opposite of Narrowband - to an engineer.   
(at the end, I will explain that to a policy wonk or wonkette, the  
term has become meaninglessly converted to mean something like "triple  
play").

The bandwidth of a band-limited message is the range of frequencies  
into which the message is coded by some modulation technique,  
specifically F[H] - F[L] where F[H] is the highest frequency used to  
code the message, and F[L] is the lowest. We usually use the symbol W  
to represent the bandwidth of a message, signal, or system - and it is  
measured in Hertz.

For a particular message, signal, or system, one can choose among  
modulation techniques that uses a wide range of frequencies or a  
narrow range of frequencies.  Broadband techniques can be compared  
with narrowband techniques on a number of "quality", "cost" and so  
forth dimensions.

In a condition where noise is Additive, Gaussian and White, there is  
one perfect transmitter and one perfect receiver, a fixed limit on the  
power of the transmitter, and  no boundary conditions on the EM field,  
the capacity theorem from information theory says that theoretical  
limit of capacity of a radio system or a coaxial cable is C = W log  
( (N+S)/N ), where N and S are the energy of the Noise and Signal at  
the receiver.

Ultrawideband techniques, for example, are the extreme form of  
broadband techniques, where W is quasi infinite, so the signal energy  
can be quite small compared to noise.

However, this has nothing whatever to do with the policy debate  
surrounding the term "Broadband".   In that debate, Broadband means  
whatever the cable and telcos want it to mean.  Sometimes it means  
High-Speed Internet Access.  Sometimes it means lots of TV channels.

In general, it appears to be being defined by policymakers as "triple  
play" with faster than dialup Internet on the side to keep those who  
want to be hip and cool Internetters "happy".

To me this is the ignorant leading the blind.  But what do we expect  
when people claiming to be engineers and computer scientists persist  
in adapting terms to mean nothing at all but "please subsidize me, Mr.  
Congressman?"










-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com