On May 11, 2013, at 7:34 PM, SM wrote:
At 13:08 11-05-2013, Tom Vest wrote:
Sorry, but unless you can point to some relevant real-world examples of
self-executing, self-sustaining principles, or you're a nihilist and don't
really believe that such things as principles exist at all, this is a
patently false, bordering on nonsense statement.
I am not suggesting any self-executing or self-sustaining principles.
Fair enough; I will assume that we agree that policies and principles are
not mutually exclusive and incompatible phenomena, and that the class of
durable, self-executing, and self-sustaining principles is an empty set.
One is tempted to ask work for who? but that would entail giving this
statement more credence that it merits. Since TCP/IP is only useful to the
end of communication between two or more nodes, the proposed principle of
finitude would perfectly consistent with this, and almost every other IETF
addressing/attachment protocol *not* working at all.
Or did you mean to say that The principle for the above is to avoid set any
constraint unless it is necessary for IETF protocols to 'work' between two
virtual machines, under lab conditions?
What I meant was to leave policy (PDP, etc.) to the communities interested in
IP addressing. I'll quote part of a message posted on the thread:
'To date, the communities interested in IP addressing have established
policies
that dictate operational needs should be the primary constraint (as
opposed
to say constraining on geo-political boundaries, by ability to pay, etc).'
The message is at
http://www.ietf.org/mail-archive/web/ietf/current/msg79200.html in case what
I was quoted is misrepresented.
I certainly did not intend to misrepresent your position. But given the fact
that the part of a message that you reproduced was offered in response to
doubts that you yourself raised about the points covered therein (esp.
operational need), what is your position, exactly? As David said, to date
the communities have established policies that are broadly informed by the
practical implications of the finitude and uniqueness constraints on address
resource management. However, to conclude based on past observations that such
will always be true would be tantamount to claiming that management of those
constraints is assured by the operation of (unspecified self-executing,
self-sustaining) principles. Based on your views as expressed in/around
draft-moonesamy-rfc2050-historic-00, it's pretty clear that you don't see any
durable, much less timeless principles embodied therein -- but that only
makes your position on these matters all the more ambiguous.
Perhaps it would help if you would answer the following clarifying questions:
1. Is it your position that some other force or principle(s) outside of the
general mechanisms/practices documented in RFC2050 (and potentially,
RFC2050-bis) guarantees that IETF-defined addressing protocols will just work
as designed, in perpetuity, and thus the informational codification of matters
related to the management of finitude and uniqueness constraints is at best
unnecessary, at worst counterproductive? If so, what are those unnamed
forces/principles, exactly?
2. Is it your position that, if the traditional communities interested in IP
addressing one day elect to adopt policies that make it impossible for IETF
addressing protocols to fulfill even the basic just work test, then from the
IETF view that should be regarded as a perfectly appropriate and acceptable
outcome?
At 13:14 11-05-2013, Brian E Carpenter wrote:
It's up to the IETF to set boundary conditions for the address space
that it created (in the case of IPv6) or inherited (in the case of
IPv4), in order to protect the long-term viability of the Internet.
There is some text about Internet address architecture. It would cover that
if the relevant communities are agreeable to it.
Could you please clarify which passages about Internet address architecture you
are suggesting are sufficient to make the sections about distribution and
uniqueness constraints unnecessary?
Thanks,
TV
Regards,
-sm
