Re: [RRG] some musings on PI v. PA, and assumptions, requirements, and tradeoffs

[jnc@mercury.lcs.mit.edu (Noel Chiappa)]

    > From: David Meyer <dmm@1-4-5.net>

    >> If you have a site which is providing content to a large part of the
    >> Internet, it makes sense (in a system-wide analysis) to have the
    >> routing support multi-homing for that site, because while everyone
    >> pays part of the costs, everyone also benefits.

    > That makes sense, although in the current regime even in
    > the case of a large multihomed content provider, the core
    > (network) still bears the burden while the site gains the
    > benefit.

This doesn't make that much economic sense to me. I reason as follows:

Assume there were no large content providers, just lots of small sites, each
making money. In that case, the network would have to carry as much overall
traffic, but somehow I don't think people would have this reaction 'oh, those
sites are getting something for free, and so the network providers should get
more money from them'.

Yes, a large content provider does use lots of bandwidth - but it has to get
that bandwidth from someone (unless they own their own fiber everywhere).

I tend to view this as a straight economic tussle - like that between a
manufacturer and a union. There's a pie (the income of the content provider),
and the network providers and the content providers are fighting over it.

I might further observe that it's only this way because the network business
has no pie to fight over. If Google/etc were barely breaking even, and the
network providers were raking in money hand over fist, I bet you'd see the
exact reverse of this argument: 'oh, those network guys are only making money
because of our content, so we deserve a share of their pie'.

I am not impressed by either argument.

Last I heard, the telephone companies weren't asking for a share of the
profits of mail-order companies who take their orders over the phone, either.

Right now, network services are not 'properly' valued by the market (in part
because there was overspending on capital investments like fiber). The same
thing happens in other markets too (e.g. commercial buildings). Eventually
things will return to a reasonable mean.


    > In some sense everyone benefits (as you say), but that cost/benefit
    > tradeoff isn't rationalized (i.e., there's no economy that assigns
    > resources accordingly).

I'm not sure I quite understand this? Are you saying that that costs aren't
perfectly assigned, that some people are getting something for free?


    > another problem with all of this is getting *there* (where ever theren
    > might be) from here. If the solution isn't incrementally deployable
    > (and those do deploy the solution reap the benefits), then we have a
    > tough sell.

That is *the* problem in Internet engineering these days. Any old fool (well,
sort of :-) can design a better network, or a jet airplane; but it takes a
real genius to figure out how to turn a fabric biplane into a jet while it's
flying! :-)


    >> hosts would have to e.g. notice that there's a service interruption,
    >> and try a different locator for the service.

    > there are the latency issues with obtaining the mappings and with
    > locator liveness.

I'll bet that pairwise liveness detection, and switching to an alternative,
is a lot quicker than having a giant distributed algorithm notice that
something has changed, and react and stabilize...

    > the mapping latency might not be an issue at "switch over time", since
    > you might have gotten all the mappings at once during the initial query
    > to the mapping system

Well, that's certainly one good option.


	Noel

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg