[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Comment-Aso] IPv8 Address Assignments for IPv6

As ICANN sorts out all of the various people that are
going to try to turn the IPv6 address space into their
own private gold mines, I suggest that people keep
in mind the following points:

1. The IPv6 address space is larger than anyone can
imagine but it has already been made very small by
those seeking to carry their IPv4 scarce resource and
high-priced IP address policies to the IPv6 world.

2. It does not take rocket scientists to assign IP addresses
contrary to what people try to tell you. The primary job is
to make sure that two parties do not get the same block
of address space. It is about as complex as an airline
stewardess making sure people are not in the same seats.

3. Routing is a separate issue from assignment. The
registries state that they do not guarantee routing. Routing
is often cited as one of the reasons why people claim an
expert "clerk" is required to make the assignments. What
people do not understand is why the clerk needs to give
people the third degree and try to learn everything they
can about the person's business just to get a simple block
of addresses.

Given this, the IPv8 and IPv16 plan is to pre-assign huge
allocations to the various TLD "communities" to then do with
as they please.This does not dictate a business plan but it
does recognize that address assignment could easily be a
small low-cost or no-cost side-line for most companies involved
in the domain registration business.

Unlike IPv6, IPv8 and IPv16 do not suffer from the IPv6 Privacy
Problem which ICANN and the Regional Registries are apparently
not concerned about. That problem is described in more detail
below. It is basically similar to the Pentium Serial Number problem.
In the case of IPv6, your packets will be tagged with an ID tied to
your computer. This will describe the manufacturer, model, etc.
Also, if you buy a system from someone, you could be buying their
"reputation" on the net. BE CAREFUL

The way that IPv8 and IPv16 get around the privacy problem is by
using the right-most 64 bits of the IPv6 address fields for addressing.
More detail of that is shown here:


In summary, IPv8 divides its address space into 2,048 Internets
which are EACH as large as the current IPv4 Internet which has
been around for over 30 years and still has plenty of addresses left.
IPv16 expands on this theme. That will be described later when needed.



From: Jim Fleming <jfleming@anet-chi.com>
To: deirdre@cdt.org <deirdre@cdt.org>
Cc: jberman@cdt.org <jberman@cdt.org>; lkessler@cdt.org <lkessler@cdt.org>;
commerce@mail.house.gov <commerce@mail.house.gov>; tom.bliley@mail.house.gov
Date: Sunday, July 18, 1999 10:58 AM
Subject: Open Letter to CDT

>The Center For Democracy & Technology
>1634 Eye Street NW, Suite 1100
>Washington, DC 20006
>(v) +1.202.637.9800
>(f) +1.202.637.0968
>@@@ http://www.cdt.org/testimony/mulligan071399.html
>JULY 13, 1999
>I read your testimony noted above before the U.S. House Commerce
>Committee regarding privacy on the Internet. I share many of the same
>concerns about privacy. I was somewhat surprised that you did not focus
>on one of the major Internet developments, IPv6, which has the potential
>to either end people's privacy or enhance people's privacy depending
>on how the technology is deployed.
>In the interest of making more money and improving their PR images
>the ICANN, ARIN, ISOC, and IETF, etc. are rushing to deploy IPv6 without
>concern for the privacy issues of their plan. They are encouraging that
>that people's NIC card addresses be encoded in each of the IPv6 packets.
>In my opinion, this is a potential violation of privacy because it exposes
>the type of equipment the person is using, as well as a unique identifier
>tied directly to that person's desktop. The NIC card address is assigned
>by the manufacturer. This is similar to the Pentium serial number issue.
>Apparently the regional registries who make millions of dollars
>selling IP addresses do not care about these issues. They appear
>to be proceeding as shown here. They are blindly following the IETF.
>@@@@ http://www.arin.net/ipv6/ipv6-regserv.html
>"Because all interface IDs are required to be in the EUI-64 format (as
>specified in RFC 2373 and RFC 2374) the boundary between the network and
>host portions is "hard" and ID address space cannot be further sub-divided.
>@@@@@ http://standards.ieee.org/regauth/oui/tutorials/EUI64.html
>"The IEEE defined 64-bit global identifier (EUI-64) is assigned by a
>manufacturer that has been assigned a company_id value by the IEEE
>Registration Authority. The 64-bit identifier is a concatenation of the
>24-bit company_id value assigned by the IEEE Registration Authority
>and a 40-bit extension identifier assigned by the organization with that
>company_id assignment."

>I suggest that people become more aware of these issues.
>Jim Fleming