[discuss] IPv6 Deployment and IG

Thu Dec 26 20:05:30 UTC 2013

At 20:31 26/12/2013, Brian E Carpenter wrote:
>The same applies to the word "policy", for which I bear some of the blame,
>having used it when drafting the document that became the IETF-IANA
>memorandum of understanding published as RFC 2860.

Along Aristotle, policy is the art of commanding free men. Many may 
prefer it to being formated by a business doxa using a technical alibi.

My long standing position is that RFC 2860 is part of the solution 
(the other part is RFC 1958) when it states: "It is recognized that 
ICANN may, through the IANA, provide similar services to other 
organisations with respect to protocols not within IETF's scope (i.e. 
registries not created by IETF or IRTF action); nothing in this MOU 
limits ICANN's ability to do so."

IETF never limited ICANN. Their own lack of vision does it.

At 18:16 26/12/2013, Andrew Sullivan wrote:
>On Thu, Dec 26, 2013 at 12:06:52PM +0100, JFC Morfin wrote:
> > No, but it needs to be governed.

NB. ("it" being the technical community)

>Why?  What is the problem that needs to be solved here?
>
>A certain amount of this discussion appears to be taking as
>self-evident that there is a problem to be solved, and so far I've
>seen several mostly ill-defined competing descriptions of "the
>problem" or even "the problems".  Surely, if we cannot describe what
>the issues are, we are going to have a mighty hard time coming up with
>mechanisms to do something about them.

I suggest you do this. Read and compare IEN 48, RFC 1122, 1958, 3439, 
3835, 3869, 3935, 5890 and 6852.  They have been written by people 
you know, we respect and who are quite clever.

- If you consider your reading is fully consistant with what you 
think today, I will never convince you.
- Otherwise, I suggest that the BUG is not in the texts, nor in what 
IETF did, but in their "loose" reading (that some benefit from).

There is a thibetan say: if something is true, it can be 
demonstrated/discovered in several ways. I have my own solution. It 
can only be consolidated by yours, and the ones of others.

At 16:25 26/12/2013, John Curran (responding Nathalie Coupet) wrote:
>   The issue being discussed is far greater than "whether there 
> should be national IPv6
>   address registries via the ITU"...  (that particular topic has 
> been discussed at length
>   in various ITU Study Groups and has gone moribund due to 
> inability to find any actual
>   problem that would be solved as a result of proceeding accordingly)

Natahalie, the IPv6 debate suffers from the same global syndrom as 
everything affected by what I call the Internet BUG (bogged down in a 
uniform/unique globality). Its aficionados look for architectural 
light in markets economies (RFC 6852) fostering competition. Yet, 
their credo is monopolistic. They designed 10 IPv6 numbering plans 
(i.e. 8 spare RIR systems, and 35,635 DN spaces, i.e. ICANNs). They 
want only one. What ITU considered was only to manage one of these 
plans and see what people would chose, for what.

Today, 1NET wants one RossetaNet, under one GS1; as they wanted one 
IDNA under Unicode, as Unicode wanted one langtag system on the IANA. 
Because the real thing is the race for the IANA. Because they can 
only imagine one single political and economical "globality" that 
could work because NewsWeek, the NYT and The Grand Chessboard said so.

The internet test phase has fully proven that the catenet concept, 
under limited TCP/IP end to end experimentation, was robust (except 
that it would probably do far better security/service wise with a 
presentation layer). May be time now to fully implement the concept 
and enlarge it? Before a blunt "critical self-organization" quake of 
magnitude forces us to do it?

Cheers
jfc