[IP] more on Breaking America's grip on the net
Begin forwarded message:
From: Russell Nelson <nelson@xxxxxxxxxx>
Date: October 9, 2005 1:31:43 AM EDT
To: dave@xxxxxxxxxx
Cc: Bob2-19-0501@xxxxxxxxxxxxxxxxxx, rob@xxxxxxxxx,
kieren@xxxxxxxxxxxxxxxxxxxxxxxxxxxx
Subject: Re: [IP] Breaking America's grip on the net
For IP if you wish, Dave. Courtesy CC to the people mentioned herein.
Breaking America's grip on the net
After troubled negotiations in Geneva, the US may be forced to
relinquish control of the internet to a coalition of governments
Kieren McCarthy
I think that enough people have written in to IP to firmly correct
Kieren. I want to say something different. You see, I finally
(FINALLY) understand what Bob Frankston says about endpoints. If
you've ever listened to Bob Frankston, you quickly realize that he is
so much smarter than everyone else that it's hard to make sense of
what he says. Bob sounds confused or insane. But Bob goes beyond
mere genius to true discernment.
How do you know who owns property?
Different countries have different solutions. In the US, states are
split up into counties. Each county has an office which registers
deeds, run by the county clerk. In theory at least, there is a
one-to-one and onto relationship between land and deeds. Property is
described in relationship to well-known benchmarks (or so goes the
theory). Many things can go wrong, which leads to conflicting
property claims. There is a whole branch of law which deals with
those times when ownership of property is not clear.
How do you know who owns the name of a product (a trademark)?
Again, different countries have different solutions. In the US, the
US Patent and Trademark Office (USPTO) says that the first party to
use a trademark in commerce is the owner. This leads to interesting
conflicts like So-and-so's Pretzels of Kunkletown, PA, and So-and-so's
Pretzels of Lancaster, PA, when both So-and-so's sold pretzels without
any conflict until one expanded into the other's territory. To avoid
these kinds of conflicts, you can register your trademark nationally,
which gives you the presumptive right of ownership (but first use
trumps registration).
How do you know how to get to somebody's website?
Before you can go somewhere, you need an address. Back in the good
old days before websites there was HOSTS.TXT, which was a single
listing of every possible hostname on the Internet. That was the
official listing of hostnames and the only way to make a
correspondance between name and address. Towards the end they had to
restrict entrance into that table simply because it had become so
unwieldy.
This system was replaced by the domain name system. Paul Mockapetris'
genius was to replace one authoritative listing with one authoritative
list of lists. This is the list that ICANN controls, and which
contains .COM, .NET, .ORG, the two-letter countries, etc. Everyone
believes this list, just as everyone believes the country clerk when
they say that somebody owns property, just as everyone believes the
USPTO when they say that somebody owns a trademark.
It is this belief, this faith, that gives ICANN its authority. It is
also why ICANN has no more control over the Internet than does the
parade marshall leading a parade. If the marchers choose not to follow,
then parade marshall has two choices: recognize that his authority has
vanished, or get back in front of the parade and continue to pretend
that he is leading them.
But what about Bob?
Bob Frankston's insight is to recognize that there are other sources
of belief.
Let's say that you want to use a P2P network to find a file being
shared. You log onto the network using a cache of IP addresses that
have previously been on the network. You start running through the
cache looking for addresses that are still connected to the network.
Once you find a few, you are connected to the network again. You can
issue a search for a name (much like a domain name), and you will be
offered several names of files. If you find several hits with the
same name, file size, and file hash, it's likely that those are all
the correct file.
Let's say that you want to find my friend Rob Logan's website. You
can go to rob.com or logan.com, or you can go to any search engine (I
tested Google, Yahoo, A9, and MSN) and type "Rob Logan". You'll get
one or the other of Rob's websites in domain name form. What if he
wasn't lucky enough to have domain names that match his names? What
if, instead, he didn't have any domain name at all? The search
engines would show him as being at 66.94.81.250.
The Internet is already usable without domain names. The UN and EU
think they're taking over control of the Internet? They're actually
taking control over a wet noodle.
--
NOTE: I am eliding some details for the sake of explanation. For
example don't bother writing to tell me that there are 45 classes of
trademarks. I already know that, and the readers of this don't need
to in order to understand my point. Also, Bob makes the point that
well-known ports such as 80 for http and 25 for email are also not
necessary, since a web server endpoint could be specified as
1.2.3.4:80 just as easily as 1.2.3.4. Or email could be addressed to
nelson@xxxxxxxxxxxxx:25. Nobody would ever need to publish those
numbers; people would say instead "search for Farber" or "search for
Russ", or "search for McDonald's".
--
--my blog is at blog.russnelson.com |
Crynwr sells support for free software | PGPok | There ought to be a
law
521 Pleasant Valley Rd. | +1 315-323-1241 | against calling for
more
Potsdam, NY 13676-3213 | Sheepdog | regulations!
-------------------------------------
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/