<<< Date Index >>>     <<< Thread Index >>>

[IP] A REPLY FROM Vixie,( is there a new DNS? Yes and No.)





Begin forwarded message:

From: Peter Bachman <peterb@xxxxxxxxx>
Date: October 2, 2005 11:06:55 AM EDT
To: dave@xxxxxxxxxx
Subject: A REPLY FROM Vixie,( is there a new DNS? Yes and No.)
Reply-To: Peter Bachman <peterb@xxxxxxxxx>


I'd hazard a guess that out of the multi-millions of people who use the Internet, probably a total of three will go back and read the IRC discussion between Paul and others regarding http://www.hyperorg.com/ blogger/misc/madrid_irc1.txt the issues that are underlying discussion of root DNS servers and democracy. Like Bob Khan's Handle System which Bob discussed with Vint Cerf in the ACM Turing Awards, there are both technical adoption issues, and social issues with which we need to be concerned in regards to the infrastructure of the Next Generation Internet. Looking at their statement below,

http://cyber.law.harvard.edu:8080/globalvoices/wiki/index.php/ Recommendations

which examines the linkages between democracy and infrastructure we can see that the participants probably realize that the government at this moment does not have the inside track on advanced theories regarding democracy, or an enviable track record overall in openess in regards to information. Cicero put it well in that officials by nature need to be constrained, but citizens should not be constrained in the same manner because the former acts in a role full of repetitive tasks, and the citizen needs to act in a number of unpredictable and thus potentially useful manners. Thus we need to be very careful with actions that restrict liberty, and sending people down the wrong road with a DNS query that comes back with the wrong answer would be fundamentally interfering with the citizen's need to improve government.

Thus it probably makes sense to keep them (the U.S. government) organizational component, at a specific part of the network DNS tree structure about where they are currently. They were happy there, but seem to have gotten OCD in attempting to control things which are beyond the scope of limited government. Expanding that scope will entail certain risks. If you dig into the MOU far enough I think there is nothing there and it's time to write a document which respects the rights of the end users in terms of identity.

For now the gov is happy to co-opt various commercial interests to proxy this for them, but we have not yet reached the next level in which we act as an "Intelligent Nation", although all the neurons and dendrites are in place for this evolution aided by the press, and on- line discussions, meetings, conferences, blogosphere etc. all wired together. However, beyond the blue door, there are a whole set of different problems regarding tapping into our collective intelligence potential, what appears to be emerging, hopefully is a design by contract where the individual can get situational awareness in terms of geolocation, services, weather, information, etc, in a location free manner.

This gives the U.S. government .GOV and .MIL and some delegations of .US

As far as them having a "national security" interest in the Internet, beyond scooping all the information up and storing it somewhere for analysis, IMHO they are not really qualified for the job. If the data needs to be meta-tagged for better semantic analysis, with identity, etc., then we should talk about that openly, with a clear risk analysis.

Right now any computer hooked up to the computer can be tracked down using the Globally Unique User ID, anyway, and as we proceed down further layers of indirection, even telephone numbers will probably be a thing of the past. Most of this number and naming is due for an overhaul to reflect the fact that we are more virtual, and more mobile.

The question is what container do you as a U.S. citizen or global internet user, want to put that meta-data information? It could be a purl, http://purl.oclc.org/docs/new_purl_summary.html a unique id like Bob F. suggests, an object identifier like the Handle System uses, or your http:// address, or an Infocard like Microsoft is developing or it could be part of a Directory.

However any individual or organization chooses to federate their identity, (and the discussion in Madrid looks to be an Identity Management discussion more than a root discussion) it certainly would be great if it had the success of the current domain name system which Mockpetris cleverly exploited in acheiving explosive growth curves by incorporating vanity license plates of E-Mail
and Web.

If TCP-IP is the information superhighway, then the equivalent of driver's licenses and plates is what we are talking about here, as well as the signage that lets us get to where we need to get to. The majority of drivers are well behaved and go about their business, but our current government is attempting to figure out ways to use information technology to track and prevent the use of those roads (the Internet) in ways that would injure the national interest. Certainly accurate and authoritative information is in everyone's best interest.

From the X.500/LDAP perspective, the government should be at an organization level, i.e. o=U.S. Government, not at the national level, which encompasses the entire U.S. people, We have a republic and wish to keep it. The government is purely an organization which we created to serve and represent our national needs, if they start on the path of a great firewall, we have a problem.

If you want to go back far enough in Internet history, I can point out where *all* the Internet domains were delegated from the ISO/OSI interunational structure and not the Department of Commerce MOU, but who cares really since it was the Internet's rapid adoption worldwide with the .com boomlet that captured people's imagination, and not the work of standards and infrastructure developers. Now that infrastructure needs to continue to adapt to support all the services that have been put on top of it. In some respects it is the old adage of 10 lbs in a 5 lb sack. Thus a call for a more robust container, or a small decentralized load.

The rapid adoption was due to the ready availability of cheap cyberspatial domain names and the resultant services that were in turn bundled with them. The resultant attempt to layer on intellectual property concerns, http://www.wired.com/wired/archive/ 2.10/mcdonalds.html to resolve disputes regarding domain names beyond the first come, first served model, certainly far outstripped anything that Jon Postel had in mind, and in fact did create the sort of nexus of control between government and private organizations


my 2cents.

Peter Bachman

peterb@xxxxxxxxx



-------------------------------------
You are subscribed as roessler@xxxxxxxxxxxxxxxxxx
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/