Creating the New Public Network

Codeine writes: "Tom Lyons argues persuasively that the incumbent competitors might be incapable of delivering an utility IP network. Competition in such commodity markets encourages the breaking of connectivity, ``Connectivity is the fundamental service of the Internet, yet it is connectivity that suffers first when network providers compete for users and services.'' Thus he proposes the Institute for the Promotion of the Internet Protocol Utility."

Comments

it's the age old question

[MORTAR_COMBAT!]

adam smith's economics and capitalism, or the promise of cheap, reliable broadband for everyone. how often has the promise of "public involvement for the public good" sounded so, well, good, but in the delivery it all goes bad. the USA has always had this attempt of having their cake and eating it too. when you try to have BOTH free markets AND public regulation, what exactly are you trying to do? either have one or the other, with both, you are playing tug of war with yourself.

Stable enough?

[webword]

Has the "IP level of the Internet" stabilized enough to consider making it a "publicly supported & controlled utility"? Sure, it has been around for a while, but are we really ready for it to be a utility? Connectivity is important but do we need the IP level of the Internet to be a utility to guarantee stability? Further, aren't there some benefits to instability, such as innovation? The article is good, but it doesn't convince me that we are ready to this kind of commitment.

Re:Stable enough?

[4of12]

are we really ready for it to be a utility?

I'm ready for it to be a utility.

aren't there some benefits to instability, such as innovation?

Indeed there are, but you don't need to dissolve all standards and semblance of order to create a nurturing environment for creativity.

Roads are an example. They were used for foot traffic and for horse and buggy and yet the innovation of the automobile was able to build upon the network of roads quite handily.

And, that innovation of the automobile still doesn't preclude other innovations (robotic aerial drones for package delivery is my favorite) that do not make use of the road network.

Likewise with digitial communications. I think the IP utility is a good idea with plenty of mileage left in it for new ideas. If someone does comes up with a dramatically better idea for digital communications networks, then I suspect that one of the hallmarks of its success will be the ability to easily piggyback legacy IP on it.

I'm not sure I understand the basis for the authors criticism of NAT. I think it's a matter of degree; that NAT can be a sufficiently good multiplexer of services that my view of the rest of the network is not overly inhibited. OTOH, security considerations for firewall NAT tend to start with the assumption of "start with it disabled", so possibly inhibiting new services.

Where has he been?

[b.foster]

Connectivity on the modern internet has been broken for many years, and will continue to stay that way as long as it is in providers' best interest to do so. Let's take a look at some examples to see why the problem is intractable:

• The internet is global. Although America has a bit of a stranglehold on most of the network, large portions of the internet are controlled by different governments, many of whom do not cooperate with the others. Saudi Arabia, China, and Iraq all firewall off most of the American internet hosts to suppress democracy. What makes My. Lyons think that these nations would be open to creating a "new public network" that allows free and open access?
• Blackhole lists are the rule, not the exception. Remember MAPS and ORBS, who asked participants to load ACLs onto their routers that killed off the class Cs and class Bs of suspected spammers? Well, these almost always resulted in collateral damage to unsuspecting customers of spammer-friendly ISPs. This created a dark underbelly of the internet: redlined addresses that were like the bastard half-brothers of the other hosts on the network, unable to access many important sites.
• Rogue nations need to be dealt with. Some nations, such as Korea and Russia, are widely acknowledged to have a preposterously bad record in dealing with security issues. Part of the problem is that their WHOIS system is unfriendly to English speakers; part of the problem is that their system administrators are severely overburdened and do not have time to fix r00ted systems. The problem arises in that it makes sense for Western hosts and ISPs to block traffic to and from these nations, in order to protect their own interests.

Balancing rights

[nuggz]

As a private company with a contract with individuals they can allow and prohibit anything they want.

It seems that the key arguement towards making internet access a utility is to remove onerous clauses from the contract, similar to consumer protection laws, existing utility legislation, and tenant rights laws.

I think this is good, charge for speed or data transfer.
But what about spammers flooding and other hostile attacks?
Removing the ability for the corporation to limit user behaviour would requite the government to limit user behaviour, with the current situation (MPAA, RIAA, DMCA, and others issues of course) we may want to be careful what we wish for.

heh

[Cave Dweller]

The Institute for the Promotion of the Internet Protocol Utility (IPIPU) will be a non-profit association of users of Internet connectivity [...]
How do you pronounce IPIPU? I-Pee-Poo?

</joke>

Public Utility

[idfrsr]

A public utility will have to be centrally regulated.
Any such regulation, will also have to regulate things that are not in the public interest, because the public utility is for the benefit of the public.
If this were to happen, how are we going do to decide what is in the public interest? We have a real hard time even with the sample of people that is slashdot deciding what is in the public interest. We could find that many things we enjoy about the internet (its anonymity, its freedom, its ability to share information) might become regulated for the public interest. We have all heard this argument before and what is happening in Australia is a perfect example

This may sound like a paranoid rant, but I think its is something people should consider, before we make this kind of decision. Many bad ideas in the world started out as good ideas....

Re:The way the world works

[Nomad7674]

Competition is a bad motivation.

The great strength of both Capitalism and Republican Democracy is their ability to turn our Sin Nature around and use it for the collective good.

(For those unfamiliy with the term "Sin Nature," it is a Christian theological term. It refers to human being's innate tendency to do that which is evil - to tend to lie, cheat, steal, destroy, etc. in pursuit of their own desires at the cost of the desires of their fellows. In this belief system, it is a fundamental building block of human nature. I think most (though not all) modern thinkers agree with some mutation of this idea, though they may decide to call it by a different name.)

Both of these systems, which dominate the world scene now due to their success, work because they channel our natural tendency to do what is good for me into doing what is good for all of us. The capitalist wants money or power or recognition. He/She does this by building a company (which provides jobs which benefits other) which produces a product (which meets a need of others) and brings in money (which pays for the other stuff and is also collected in taxes to support the many). The key is to make sure the structure is put together correctly so that this works. It is the strengths which lead to the unprecedented wealth and prosperity seen in First-World countries. It is the weaknesses which lead to problems like Enron, etc.

This post is a bit long. The basic point is that competition is not bad. It is simply a force which our society has harnessed for good. Don't try to end the competition. Try to put a structure in place which will make the compeition work for society to produce the Good you are looking for.

One word

[Rupert]

Latency.

You don't notice it on your XM radio, because it's all one way. The various satellite IP systems I've seen have played rather scary games with the network stack in order to get some semblance of performance (and even then, not nearly as good as cable or DSL).