Rohit, you never told me Hal Varian is studying usage-based pricing of
Internet router capacity!
> Having presented a qualified argument in my last notes against usage-based
> pricing of Internet router capacity, I should cite a brief, accessible
> summary of the arguments in favor of such schemes: Jeffrey K. MacKie-Mason
> and Hal R. Varian, Some FAQs about usage-based pricing, Computer Networks
> and ISDN Systems 28, 1995, pages 257-265. I do have one complaint about
> this article: it does not distinguish consistently between pricing the
> use of individual servers, a thoroughly decentralized mechanism which does
> not affect the basic architecture of the Internet, and pricing of packet
> routing, which has pervasive consequences for the architecture. It seems
> to me, as I pointed out in my notes, that congestion of particular servers
> is a completely different issue from congestion of the packet-switching
> network.
And look, a plug for kudos...
> (I find it puzzling, by the way, that I have never heard anybody discuss
> congestion of domain name servers in economic terms. That seems like
> easily the most remediable of the bottlenecks that disrupt my own use of
> the Web, particularly given that -- like many people -- I avoid loading
> images as much as possible.)
And what's a "serious economist"?
> I would also emphasize that MacKie-Mason and Varian, who are serious
> economists, only advocate usage-based pricing to support the variable
> costs of running the Internet, not the fixed cost. Moreover, as they
> point out, almost all of the costs of running the Internet are fixed
> costs. At the same time, they suggest that the revenues from usage-based
> pricing be used to create new capacity. I do not understand the rationale
> for that suggestion. It seems like, by their own logic, new capacity
> should be funded through the capital markets, with the investors getting
> their money back once the new capacity is online and the new fixed-rate
> fees start rolling in.
While we're recommending stuff...
> Recommended: Sharon Eisner Gillett and Mitchell Kapor, The self-governing
> Internet: Coordination by design, in Brian Kahin and James Keller, eds,
> Coordination of the Internet, Cambridge: MIT Press, 1997. Available on
> the Web at http://ccs.mit.edu/CCSWP197.html . This is a good introduction
> to the Internet's governance mechanisms, with a clear emphasis on keeping
> those mechanisms as decentralized as possible. While I certainly agree
> that the Internet's culture of decentralization and interoperability etc
> is a good thing and worth preserving, I would also argue that the kinds of
> formal mechanisms and cultural values that Gillett and Kapor describe are
> only one fairly weak force in favor of keeping the Internet decentralized.
> If the Internet is a natural monopoly -- a concept that many people simply
> dismiss out of hand with no substantive argument -- then that force will
> be swept away sooner or later unless some other, greater force -- like big
> bad government -- becomes aligned with it.
So is the Internet the immovable object in this scenario, or is it the
irresistible force?
----
adam@cs.caltech.edu
Iron rusts from disuse; stagnant water loses its purity and in cold
weather becomes frozen; even so does inaction sap the vigor of the mind.
-- Leonardo da Vinci, The Notebooks