Jeremy wrote:

] * Acknowledge that non-Internet communications equipment ...
] is economically obsolete ...

I'm interested in hearing what other people on the site think about this.

I agree with many of the things being said here, but I've stopped short of recommending this in the past. There seems to be an irrational undercurrent in this that wants to say "my stuff is better then your stuff" in an absolutist and unthinking way. There are certainly serious problems and limitations with the Internet Protocol suite, and in the last few years the quality of the "standards" the IETF and similar bodies have been producing has dropped dramatically.

IP didn't win because its the most capable solution. It won because it existed in an environment where the telecom monopolies were actively trying to stall the development of digital networks, and IP was the hardest solution to control.

Hard to control doesn't always been optimal. Gnutella is not more efficient then napster.

Having said that, I think its clear that IP has "won" and that with the deployment of IP being so ubiquitous that any future development must, at least, interoperate with it in order to be useful. I don't think we're going to replace it with something else anytime soon.

I raise this counter point only to provide the perspective needed to see what I think the FCC ought to be doing, which is that they ought to be agnostic. The FCC should no more prefer IP based solutions then other solutions. What it should do is create an environment where its possible for different solutions to be made available... an environment where it is possible to innovate. I don't want the government choosing a technological direction, and a future where non-IP based solutions are simply not possible is as much a threat to innovation as a future where the only technologies that are allowed are the ones that benefit the telecom companies.

Decius wrote:
] I don't think we're going to replace [IP]
] with something else anytime soon.

The next issue of the Cook Report will include an article
from Farooq Hussain (who used to run the Sprint NAP) in which
he will argue that IPv6 will never be widely deployed.

] ... they ought to be agnostic. The FCC should no more prefer
] IP based solutions than other solutions. ... I don't want
] the government choosing a technological direction, ...

Ideally, the FCC would write policy that encourages innovation
without inherently selecting the winning technology. However,
the government is also a consumer in this marketplace and must
make choices to support its own operations. To the extent that
the impact of this selection affects the marketplace, then the
influence of the government seems unavoidable.

Jeremy wrote:
] However, the government is also a consumer in this
] marketplace and must make choices to support its own
] operations. To the extent that the impact of this
] selection affects the marketplace, then the
] influence of the government seems unavoidable.

Well, thats pandora's box.

In general, there is a separation between policy makers and operational people, and my comments mostly reflect policy as opposed to operations. However, you raise an interesting question. What is the government's responsibility with respect to the influence of their purchasing decisions on technological direction....

There is no clear answer to this I don't think...

This has come up in a number of contexts. For example, the government hasn't invested social security funds for fear of influencing the market with their investments. On the other hand, the government tried (and failed) to promulgate clipper chips by requiring them for government contractors.

In general there is going to be little or no effect, and this shouldn't be a concern. When the purchase is extremely large and there is a significant fear that there might be a social impact, then this impact should be considered from a moral standpoint, but you really get into a space where the only reasonable place to make that kind of decision is in the legislature, as its democratically controlled. Unfortunately, there are lots of problems with doing that from a practical standpoint. Currently, that doesn't happen for the most part, and so we see situations where these concerns are taken into account, and other situations where they aren't...

This is messy....

Its possible to grab onto a purely anarchocapitalist perspective that government when purchasing is simply an actor in the marketplace and it should simply pursue its interests even in situations where the purchase is extremely large. This provides the comfort of an easy answer, but I dunno if it makes practical sense in every case...

Can we let the government make market investments if we set things up such that the people who make those investments are purely incented to make a profit?

Decius wrote:
] Jeremy wrote:
]
] ] * Acknowledge that non-Internet communications equipment ...
]
] ] is economically obsolete ...
]
] I'm interested in hearing what other people on the site think
] about this.
]
] I agree with many of the things being said here, but I've
] stopped short of recommending this in the past. There seems to
] be an irrational undercurrent in this that wants to say "my
] stuff is better then your stuff" in an absolutist and
] unthinking way. There are certainly serious problems and
] limitations with the Internet Protocol suite, and in the last
] few years the quality of the "standards" the IETF and similar
] bodies have been producing has dropped dramatically.

you're correct in this regard. Simply replacing old stuff with new stuff doesn't help anyone. If anything, it causes prices to go up, because telecom infrastructure is expensive and the depreciation rates are much faster than they've ever been due to new technology cycles. I don't agree with sitting on 100 year old plant and milking it, but I don't agree with pulling fiber to the home either. Both are a waste of application and BOTH are responsible for the current problems in telecom.

The article misses the most key point in the whole situation. Investor sentiment. The Fail Fast philosophy might be helpful to get rid of the pain quicker, but ultimately, the same problem will exist, and that is that investors are not likely to return to the telecom industry because of all the pain and suffering that has already occured.

Expectations were too high for new competitors entering the market (and time frames too short), and now investors are fleeing RBOC investment because of their shoddy debt and eroding margins. Face it: investors aren't going to get excited about spending Billions of dollars on plant and infrastructure, only to get gross margins in the teens in 5 years. Fail fast is not an answer in and of itself.

] IP didn't win because its the most capable solution. It won
] because it existed in an environment where the telecom
] monopolies were actively trying to stall the development of
] digital networks, and IP was the hardest solution to control.

it also worked with everything as a design philosophy (ie, it was the opposite of proprietary).

] Hard to control doesn't always been optimal. Gnutella is not
] more efficient then napster.

neither does 'winner'. Look at everything around you. The optimal solution is rarely the 'winner' in the marketplace. This is why I don't believe in the standard (and inaccurate) interpretation of Darwin's theories. The 'best' answer to a problem rarely ends up being the one implemented. Look at Microsoft. Look at the automotive industry. The music industry. Even the HTTP protocol sucks (from an engineering standpoint) but a huge revolution has been built upon it.

] Hav... [ Read More (0.2k in body) ]