[CAnet - news] The Wealth of Networks

[Highly recommended reading. From a posting by Tom Vest on Gordon Cook's list -reprinted with permission from Tom Vest.  ---BSA

Some excerpts from Tom Vest papers on his web site:

The Wealth of Nations
“The annual labour of every nation is the fund which originally
supplies it with all the necessaries and conveniences of life
which it annually consumes, and which consist always either in
the immediate produce of that labour, or in what is purchased
with that produce from other nations.”

The Wealth of Networks
The Internet production of all networks is the fund which
originally presents each Internet user with information access
and consumption possibilities, and which consist always either in
local Internet production, or in what can secured by
interconnecting with other networks.


Policies and institutions that foster provider diversity
(e.g., IXPs) can contribute to growth and local diffusion of
Internet resources (users and content) that are the
primary objects of public interest claims about the
Internet.

Although the "economics of light" may only be clearly understood, in Hegelian fashion, once its dominion has completely passed, the "economics of land" has been exhaustively studied since the eighteenth century. This dissertation research draws on the body of classical economic scholarship to posit a framework for understanding how various territorially rooted infrastructure systems (land) are transformed and multiplexed by
diverse IP network operators (labor) to produce critical network resources (capital) -- in the form usage and content -- that represent the Internet's primary goods. This theoretical framework is used to interpret publicly available, globally scoped data illustrating the changing distribution of network infrastructure and the Internet resources that it supports at the enterprise, national, and global levels.



Tom Vest says...

My dissertation, "The Wealth of Networks," uses archived Internet  
routing tables -- each a nominally global scope view of the universe  
of important public policy goods (aka "Internet production, or the  
sum of Internet "users," "uses," and "usage") -- from 1997-present to  
illustrate how the public Internet is evolving at the enterprise,  
national, and global levels. I compare numbers of logical "ends," or  
routed public IP (attributed to the originating AS, and thence to the  
AS' ISO country code) to the underlying physical ends for the  
corresponding country -- that is, to the summed copper and coax  
infrastructure recorded annually by the ITU. This comparison helps  
reveal, for example, how fast the Internet is growing relative to the  
underlying infrastructure -- and helps to frame cases to address  
questions like "all things remaining equal, what makes the Internet  
grow fast or slow?", "Why is this country growing much faster,  
getting more concentrated, more diversified, etc ., than that  
country, which faces the same observable/relevant circumstances?"

This approach provides a consistent, defensible framework for  
international comparison that parallels recurring public policy  
interests, is consistent with the assumptions of both critics and  
defenders of current commercial arrangements (e.g., that  
infrastructure is paramount, and IP transit the most expensive / most  
valuable factor for international operators), and provides -- I hope  
-- an idiom that might help to bridge the Bellhead-Nethead divide.  
The OECD picked up the data and methodology for their upcoming  
report,  "Internet Traffic Exchange: Market Developments and  
Measurement" (public release: Spring 2006), and it may be  
incorporated into future releases of the OECD Communications Record  
on an ongoing basis.

The work began as a rejoinder to the ICAIS debates of the late 1990s,  
as an attempt to to reveal whether underdevelopment in the aggrieved  
countries was more plausibly explained by domestic vs. international  
factors. Preliminary results suggest that provider diversity has an  
independent, statistically significant positive effect on the growth  
of those easy-to-understand public goods -- users, uses, usage. I  
explain this effect by analogy to statistical multiplexing --  
multiple overlapping logical systems are better able to leverage any  
infrastructure platform than a single logical system of similar  
size....I call this (somewhat optimistically) "logical multiplexing."

Although the "stickiness" of number resources makes it's easier to  
study network economy growth and harder to study atrophy/contraction,  
I use the results to suggest that eliminating the causes of growth is  
likely to curtail growth also, even if it takes longer to observe  
statistically. Have given this version of the talk in a couple of  
places, but it needs more work.

Am already on the hot seat to get the time series data up and finish  
my dissertation in the next couple of months; will be happy to share  
if there is interest. For now, some past iterations of the work are  
online at:

http://www.pch.net/resources/papers/the-wealth-of-networks/

Questions/comments welcome!

[NOTE: before the network engineers on the list pounce, note that  
I've been able to observe very big global networks from the inside,  
am close to several RIRs, and have taken great pains to to try to  
make explicit how changes over time in both routing technology and  
number resource policies color any/every claim that one might be able  
to make using this research. To the best of my knowledge I'm not  
interpreting anything naively -- and the curve traced by the data  
matches my first-hand observations as an operator fairly closely.]

TV

-----------
[email protected]
www.canarie.ca/~bstarn
skype: pocketpro
SkypeIn: +1 614 441-9603


_______________________________________________
news mailing list
[email protected]
http://lists.canarie.ca/mailman/listinfo/news