I have objected to the prefix/locator translation name because some
versions of the document, including the current one, also modify the
identifier/IID portion of the address.
I think the important distinction here that Remi wants to make (if I
understand correctly) is not in what part of the address is
translated, but in the fact that this translation is algorithmic and
reversable and does not require any sort of per-flow state. The
devices is _not_ stateless, however, because as others have pointed
out there is configured state.
So, if it is felt we need a special qualifier in the title of the
document to distinguish this from other types of NAT, I could go for
"Algorithmic NAT66"... Would that address the concerns that you and
Remi have about calling it simply "NAT66"?
Margaret
On Oct 29, 2010, at 7:07 PM, Fred Baker wrote:
On Oct 29, 2010, at 3:53 PM, Margaret Wasserman wrote:
- Some contributors advocate that "NAT66" should only designate
*stateless NAT66*.
You are the only person I have heard make that statement. If
others hold this opinion, they should share it with me.
I do. The draft in question is draft-mrw-nat66-nn.txt, and it
describes a stateless NAT.
Why do I want that limitation? First, because that is the discussion
this list was created to have. Second, because every argument I hear
that uses the acronym is about stateful NAT, and the volume of "I
have to have..." and "I am absolutely apposed to ..." prevents any
discussion of the proposal actually on the table.
As I have said to you many times, I wish you had called it what it
is, which is prefix translation or locator translation. Keith thinks
that since it changes the locator it is equivalent to network
address translation, and I will give him 10% of that point. But
unless and until we can have a discussion on the topic that is
actually on-topic, I continue to think that the topic we have
actually proposed is not being discussed.
There is an alternative. We can create a different list for
discussion of stateless network prefix translation, named
[email protected]
and abandon this one to "abandon hope all ye who enter here". If we
do, I will sign off this list. Except that's what we thought we were
building here.
Oh, the 6ai BOF had the same problem. Did you notice that you
weren't actually permitted to speak to the topic on the table, other
speakers were brought in to confuse the discussion including Remi,
who wanted to talk about 6rd tunneling and as always jumped on every
venue that opened up with a view to talking about it, and the l-o-n-
g line of people at the mike complaining about NATs were all
complaining about stateful NAT and often specifically IPv4/IPv4 NAT?
The filibuster is getting Really Old. It would be nice to be able to
have an actual technical discussion of the proposal that is actually
on the table. For the first time since it was brought up.
_______________________________________________
nat66 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nat66
