[FoRK] [IP] more on INTERESTING ARGUMENTS DJF The Myth of Scarcity, or Verizon-MCI is Doomed (fwd from dave@farber.net)

Eugen Leitl eugen at leitl.org
Wed May 4 08:23:29 PDT 2005

----- Forwarded message from David Farber <dave at farber.net> -----

From: David Farber <dave at farber.net>
Date: Tue, 3 May 2005 14:22:57 -0400
To: Ip <ip at v2.listbox.com>
Subject: [IP] more on  INTERESTING ARGUMENTS DJF  The Myth of Scarcity, or Verizon-MCI is Doomed
X-Mailer: Apple Mail (2.728)
Reply-To: dave at farber.net

Begin forwarded message:

From: vijay gill <vgill at vijaygill.com>
Date: May 3, 2005 12:11:45 AM EDT
To: dave at farber.net
Cc: Ip <ip at v2.listbox.com>
Subject: Re: [IP] The Myth of Scarcity, or Verizon-MCI is Doomed

David Farber wrote:

>May 02, 2005
>The Myth of Scarcity
>Posted by Dana Blankenhorn
>The bidding war between Verizon and Qwest for MCI is based on a  
>myth  of scarcity. That is, both think they can make the deal pay  
>by  squeezing customers for the scarce resources represented by the  
>MCI  network.
>Moores Law of Fiber rendered that inoperative many years ago.  
>There  is no shortage of fiber backbone capacity. And there are  
>ample  replacements for Plain Old Telephone Service -- not just  
>cable but  wireless.

This is starting to read like Gilders Telecosm. Yes, there is no  
shortage of fiber backbone capacity. The problem is looking at the  
fiber in the ground the major component, when in reality, in building  
a backbone, this is only one of many major components. Lets go into  
some excruciating detail here (to backup my statements, I present the  
amount of traffic on a backbone I am responsible for: http:// 

1) There is no shortage of fiber backbone capacity.
True. There is plenty of fiber in the ground. There are even empty  
conduits you can blow more fiber through. However, lighting the fiber  
is non-trivial in terms of capital spend. Then you have huts you have  
to maintain, and cards in chassis that sometimes go bad, and it costs  
you to have 4 hour turnaround in a hut in the middle of Ohio.

Then there are pathing issues. You build a backbone and since there  
is infinite capacity of fiber in the ground, you turn up multiple  
thousands of gigabits. Then a backhoe or a tunnel fire cuts your  
network capacity in half and, well, you're in trouble. There is  
plenty of fiber, there are not plenty of PATHS, so to avoid that,  
you're doubling up capacity on the same routes and instead of n:1  
redundancy, you are now effectively operating a 1:1 redundancy or  
3:1, and your cost basis just doubled.

2) Any idiot can build a global backbone and at NANOG 34 (http:// 
www.nanog.org), yours truly will speak to that, using himself as an  
example. Building a backbone is _easy_. Getting people onto the  
backbone is _hard_. The issue is not even the last mile or the first  
mile. The issue is the last 100-200 feet. Laterals cost you money.  
They cost you time. Getting a permit to trench into a building from  
your fiber path in the street is non-trivial. Wireless local loop or  
lasers are NOT the answer, as many billions in failed ventures will  
tell you (Telegent, Monet, winstar et al).

3) Operating a backbone costs money. Everyone buys the same  
equipment, and their COGS tends to be about the same for the layer 3  
side of the house. Things like OSS/NMS will become the competitive  
advantage. Today, we're used to living in high margin situations.  
This is untenable and the companies will have to learn how to survive  
on 10%-15% margins. The SG&A will have to be cut down to the bone,  
and the way to do this is automation. And building automation is  
beyond hard, ask Mike O'Dell, who I believe is a reader, about his  
time at UUNET. It takes an amazing amount of force of will, time and  
talent to automate the network and that costs money. Unless you have  
deep pockets already, you cannot afford to fund this, and you will  
get crushed trying to make a business out of that "abundant" capacity.

4) Wireless. POTs is being supplanted by wireless, and the dominant  
players in wireless also just happen to be verizon/sbcs of the world.  
This is because building a wireless network is also hard. And did I  
mention it is expensive? Building a RAN is spendy enough that the  
cost of IP equipment to backhaul the packets/voice is trivial in  
comparison. Waving the magical wireless wand is easy. Circumventing  
laws of physics is hard. Circumventing community resistance to  
putting up towers, care and feeding of towers, backhaul capacity from  
towers and the maintenance of the equipment is harder still.

>The myth on which this deal is based is, simply, untrue.


I do not think so. Maybe I am too focused on the operational  
realities of networking that I am not seeing the forest for the  
trees, but I deal with spreadsheets and they almost never lie. At the  
end of the day, is the return on the spend on the network greater  
than my cost of capital? Yes/No.

>The same scarcity myth has long underlain the FCC's regulation of   
>broadcast content. The FCC regulates what is broadcast only  
>because  there is a limited frequency spectrum, thus limited choice  
>in each  market. Without that myth all content regulation is a  
>direct  violation of the First Amendment.
>And it is a myth. Moore's Law of Radios proves it. Cellular  
>telephony  proves it. Frequencies can be re-used, and transmit far  
>more than we  once thought. The resource is, in fact, practically  

Frequencies can be re-used. However, re-use looks great from the 100k  
foot view. On the ground, reality is a bit different. High re-use.  
Ok, that means smaller "cells" of coverage. Smaller cells means more  
cell-sites. More equipment. More backhaul. More maintenance, more  
capital spend. At this point we hear the mantra about moores law  
making the electronics cheaper every n months. That is true for the  
processing elements, but things like antennas, housings, power, and  
maintenance are  fixed and don't come down as quickly.

>All efforts at controlling the mind are subject to this myth of   
>scarcity. Try to control education and smart kids learn to reject   
>you. Try to control the Internet and Chinese people learn just how   
>precious a sip of freedom is, while the rest of us get bored by  
>the  firehose.
>Freedom is the answer to tyranny because freedom creates  
>abundance,  or discovers the abundance that is in fact all around us.
>To all those who feel oppressed, no matter their cause, I hope  
>this  brings some comfort. In the end, freedom will win out. Maybe  
>not  today, maybe not tomorrow, maybe not even in your lifetime,  
>but it  will. Freedom, knowledge, and truth are all plastic, all  
>powerful,  and all unlimited.
>The universe, whether we're talking about your mind, this planet,  
>or  the stars themselves, is far more infinite than you or I can  
>possibly  know.
>And the only way to get a taste of it is to open your mind, as   
>scientists do, and prepare always to have it changed.

This is definitely very pretty wording.


You are subscribed as eugen at leitl.org
To manage your subscription, go to

Archives at: http://www.interesting-people.org/archives/interesting-people/

----- End forwarded message -----
Eugen* Leitl <a href="http://leitl.org">leitl</a>
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net

More information about the FoRK mailing list