Fw: On IPv4, IPv6, IPv8 and IPv16 Address Spaces

["Jim Fleming" <jfleming at anet.com>]

----- Original Message -----
From: "Jim Fleming" <jfleming at anet.com>
To: "deering at cisco. com" <deering at cisco.com>; "Lynn" <lynn at icann.org>; "vint cerf" <vcerf at MCI.NET>; <nvictory at ntia.doc.gov>;
<broseman at ix.netcom.com>; "CJ Wittbrodt" <cjw at groovy.com>; <dawn.martin at wcom.com>; "narten at raleigh. ibm. com"
<narten at raleigh.ibm.com>
Cc: "mueller at syr. edu" <mueller at syr.edu>; "froomkin at law. miami. edu" <froomkin at law.miami.edu>; <richard at vrx.net>; "Gordon Cook"
<cook at cookreport.com>; "Joanna Lane" <jo-uk at rcn.com>; <terastra at terabytz. co. nz>; "karl at cavebear. com" <karl at cavebear.com>;
"andy at ccc. de" <andy at ccc.de>; "Judith Oppenheimer" <joppenheimer at icbtollfree.com>; "Joe Baptista" <baptista at dot-god.com>;
"love at cptech. org" <love at cptech.org>; "tbyfield at panix. com" <tbyfield at panix.com>; "todd glassey" <todd.glassey at worldnet.att.net>
Sent: Tuesday, July 23, 2002 9:54 PM
Subject: On IPv4, IPv6, IPv8 and IPv16 Address Spaces


> http://www.icann.org/aso/ipv6-statement-11jul02.htm
> "Compared to IPv4, IPv6 has a seemingly endless amount of address space. While superficially true, short-sighted and wasteful
> allocation policies could also result in the adoption of practices that lead to premature exhaustion of the address space.
>
> It should be noted that the 128-bit address space is divided into three logical parts, with the usage of each component managed
> differently. The rightmost 64 bits, the Interface Identifier [RFC2373], will often be a globally-unique IEEE identifier (e.g., mac
> address). Although an "inefficient" way to use the Interface Identifier field from the perspective of maximizing the number of
> addressable nodes, the numbering scheme was explicitly chosen to simplify Stateless Address Autoconfiguration [RFC2462]."
> =====
>
> 1. Correction...."it should be noted that"...**the I* society's view** of the 128-bit address space is now apparently
> that it is divided into three logical parts. That seems to change from year to year. So much for stability. It is interesting
> to see that the IPv6 Privacy Problem has still not been addressed, with people being encouraged to put their
> hardware serial numbers ("mac address") in their packet's address fields.
>
> 2. It should also be noted that people seem to be allocating pieces of the IPv6 address space without the involvement
> of the so-called RIRs (Regional Registries), which only manage **part** of the IPv4 address space. Most large
> commercial companies have their own IPv4 address space and do not have to pay the RIRs for what should be
> a totally automated process of obtaining (or returning) an available block. There has been little or no motivation to
> automate the allocation process, because people's non-profit job security depends on the cash-flow from leasing
> the IPv4 address space, and now it looks like they expect to profit (or non-profit) from IPv6 leasing.
> http://www.iana.org/assignments/ipv4-address-space
> http://www.hs247.com
>
> 3. IPv8 and IPv16 addresses are encoded in the **right-most 64-bits** of the 128-bit DNS. The left-most 64-bits
> are used for transition mechanisms. Currently the 2002:<IPv4>:: style of addressing found in modern versions
> of Windows allows for one IPv4 address to be used as a site-id to help transition to the IPv8 Address Space.
> That is a more stable transition mechanism because people remain connected via the legacy (aging) IPv4 Internet
> as they evolve to a larger address space. Estimates show that there are still plenty of IPv4 addresses, and the
> 11 bits added via IPv8, increases the existing Internet by a factor of 2,048. In other words, a new Internet, the
> size of the entire IPv4 Internet could be added each year for the next 2,048 years. That is just for the added IPv8
> bits which can be carefully stuffed inside of existing IPv4 headers. IPv16 increases the address space for 4,096
> years. That can all happen now, and people do not have to wait. It is not a scace resource, control policy, as
> found in all I* society approaches to allocations.
>
> 4. The top 2,048 TLD managers can all be the equivalent of RIRs, by managing space via IN-ADDR.[T:LD] in
> the same manner as IN-ADDR.ARPA is handled. The TLD managers will no doubt want to automate the
> process, to minimize the cost of managing the address allocations. A non-profit has no motivation to minimize
> costs. Revenues help to pay for all of the employees to fly around all year to parties in Miami and Las Vegas.
> Customers will have to decide if that is what they want their money spent on. With the I* society, outsiders have
> no say on how the money is spent. It is all kept carefully inside the closed society.
>
> Jim Fleming
> http://www.ntia.doc.gov/ntiahome/domainname/130dftmail/unir.txt
>
>
>


-----
This is the afnog mailing list, managed by Majordomo 1.94.5

To send a message to this list, e-mail afnog at afnog.org
To send a request to majordomo, e-mail majordomo at afnog.org and put
your request in the body of the message (i.e use "help" for help)

This list is maintained by owner-afnog at afnog.org