Re: Metcalf and Other Net.Fogies

At 10:27 AM 9/5/96 -0700, Timothy C. May wrote:

>Robert Metcalfe, inventor of Ethernet and founder of 3COM, and how
>publisher of "Infoworld" and sailing enthusiast, was interviewed on CNBC a
>few minutes ago. He repeated his prediction of an "Internet collapse" in
>1996, based on overuse, on bad pricing models, on lack of controls, and on
>other concerns.

Some are griping just to gripe.  Metcalfe may have a valid prediction here.

I have a machine that I use for work that is connected to the net via
ethernet.  No 28.8k bottlenecks involved.  Most of the time I am lucky if I
can exceed 14K bps to anything outside the local area.  When I run
traceroutes, the blockage is in MCI or Sprintnet land.  (Except for the one
to ftp.funet.fi early this week where where two of the machines somewhere
in California were caught in some sort of weird DNS loop.)

The bandwidth to the net has been oversold.  If the government were
*Really* concerned about "protecting the net", they would be on MCI and
Sprint's cases, not looking for virtual terrorists.  (Virtual Terrorists
are to Terrorists the same way that Virtual Reality is to actual reality.)
There have been days where you could not move anything at any reasonable
speed from certain areas of the country.

Yes, people have been predicting the end of the net (GIF/JPEG/WAV at 11!)
since it was founded.  In this case, I think that the person has enough
network experience to be right.  With the way things are now with oversold
bandwidth, the DNS numbers getting close to being used up, many of the
routers needing to be replaced and/or upgraded, and software that uses
bandwidth like candy (phone conversations, video conferencing, huge
interactive web page animations (like shockwave), real audio, and more as
the marketing droids can sell you on it.), the chances of a west coast
power-system style collapse does not seem that far from reality.  (There
are many who I know in the industry that are amazed that it has lasted this
long.)

On a smaller scale, those collapses happen in a small regional area, get
fixed and things go on.  But just like the earthquakes in California,
everyone is waiting for "the Big One".  (At least this one is preventable.
Lets hope that the fixes can occur before the government gets involved,
otherwise the net *IS* really doomed.)

---
Alan Olsen -- alano@teleport.com -- Contract Web Design & Instruction
        `finger -l alano@teleport.com` for PGP 2.6.2 key 
                http://www.teleport.com/~alano/ 
  "We had to destroy the Internet in order to save it." - Sen. Exon
                "Microsoft -- Nothing but NT promises."

Alan Olsen writes:
> Metcalfe may have a valid prediction here.

Metcalfe is talking out his ass. He's reached the "old geezer who's
impeding his own field" stage. Many of his articles seem to be written
as though no one was trying to fix problems.

> When I run traceroutes, the blockage is in MCI or Sprintnet land.

How do you manage to determine where you are losing bandwidth using
traceroute? That must be a mighty powerful traceroute to do that --
most traceroutes I've seen are hard pressed just to find out what the
connectivity path is.

> The bandwidth to the net has been oversold.

Always the case. Big deal. Bandwidth is still increasing pretty
fast. There are, naturally, growing pains, but the outages and
bandwidth situation are pretty good, all things considered. Compared
to the way things were eight or nine years ago they are amazing;
compared to four years ago they are still astoundingly better.

Now if we could only go back in time and shoot the folks responsible
for HTTP before they thought of it we might even be able to do
something about the packet loss situation -- if HTTP just played nice
with TCP and Netscape didn't spawn simultaneous TCPs the situation
would be much improved.

Perry