From: Remo Pini <rp@rpini.com>
Date: Sun, 21 Jul 1996 09:01:21 +0800
To: cypherpunks@toad.com
Subject: RE: ABC news on Internet Telephony
Message-ID: <1.5.4.32.19960720223758.0097ee74@193.246.3.200>
MIME-Version: 1.0
Content-Type: text/plain


At 01:54 PM 7/18/96 -0700, you wrote:

--- all the following points are based on swiss circumstances, they may not
apply to US ---

>point to point circuits are more efficiently handled by circuit
>switching rather than packet switching networks. Nicholas
>Negroponte wrote an interesting piece about asynchronous vs
>synchronous, I believe it is in his book "Being Digital." 

Well, from a users point of view, sending packet data over a packet mode
bearer service is more efficient (and cheaper). An interesting developement
in this direction is the PMBS-A/B modes of ISDN (packet switching to the
public switch). The existance of this service suggests its usability.

>ADSL is an interesting attempt at digital telephony but expensive
>and basically would mean replacing existing central office
>switches. (backbone bandwidth) 

We have a well developed DQDB-MAN and ATM net around, and bandwidth is
available (and getting cheaper by the minute). Currently, a onetime
investment of around $2500 per client is necessary to provide >5MBit/s
transfer volume (via the cable TV networks or the existing broadband networks)

>In a packet network you have to either dedicate a portion of the
>bandwidth for a synchronous circuit, or you have to have a very
>fast network and use very small packets (ATM), expensive either
>way.

Not if you have a dedicated packet switching network for asynchronous packet
transfer only. If you use it for both you don't have to have a very fast
network, you have to have a network with predictable and constant packet
delay. (that's not the same as fast!)

>A single central office has many times the bandwidth of the widest
>part of the internet, and the average state has hundreds of CO's.
>If even a small portion of the Internets current users tried
>placing a call things would grind to a halt. A huge increase in the
>number of backbones and their bandwidth would solve this, but who
>will pay the bill? 

I guess Internet-telephony is one of the bandwidth killers.

>TANSTAAFL
>
>Sometime ago the discussion was on the cost of laying new fiber,
>may I suggest  the realworld heuristic of "a million dollars a
>mile."

There are of course a lot of alternatives:
- Existing wiring (5 MBit/s over 6 copper wires is possible)
- Usage of the cable networks
- Radio transmissions (RITL - radio in the loop)
- Satellite transmissions

>Please note I am not trying to make fun of anyone personnally, I am
>in the words of Jubal Harshaw "heaping scorn upon an inexcuseably
>silly idea, a practice I shall always follow."

Neither am I, but isn't anyone?

----------< fate favors the prepared mind >----------
Remo Pini                      Fon 1: +41 1 350 28 82
mailto:rp@rpini.com            Fon 2: +41 1 465 31 90
http://www.rpini.com/remopini/ Fax:   +41 1 350 28 84
soon:PGP: http://www.rpini.com/remopini/rpcrypto.html
--------< words are what reality is made of >--------