Hi Geoff,

On 17/08/11 3:19 PM, "Geoff Huston" <[email protected]> wrote:

> Thank you for your response.
> 
> So this draft is not really a draft about routing intent. Its a draft that in
> effect proposes a rewrite of the IPv4 registry, and potentially the IPv6

Nothing regarding who the allocations in the IPv4 and IPv6 registry are made
to will change. The dates won't change. And for the most part the prefix in
the entries won't change. I'm trying, but really struggling, to see this
called a re-write. 

>From an engineering perspective it is just, and only, about signaling
routing intent so people who read the registry either with human or machine
eyes can transparently see what prefixes are intended to be routed without
any ambiguity. This has benefits for the Security efforts in RPKI and
elsewhere and for anyone dealing with the burden of an allocation which has
fallen under the filtering bus[1].

It makes engineering such structures immensely simpler, and I certainly have
an interest in doing that given I will be implementing the RPKI system for
ICANN. Transparency in registry also aides in the engineering trust point.
eg "is what is represented in the registry a 1:1 of what is produced in
RPKI?"

[1] Approximately once a month or less I see an operator on a mailing list
speak up about how their recently assigned prefix is unusable due to
filtering. No silver bullets here.. but everything from the RIR's QA
procedures, to better documentation, and clear unambigious registries will
help.


> address registry. Frankly, in such a light, I'm unhappy that this is a GROW WG
> document, and I would prefer that this is an IAB document [RFC2860].

I will open a discussion with the grow chairs and the IAB in regard to this.

> 
> And no, I'm not confused any more. I object to this draft.

Actually, since it was clarifying the position in the IANA registries to
make the "world a better place" I really thought I would see 2 thumbs up
from you.

> 
> - it is out of scope of GROW's charter

The GROW charter says:

"The purpose of the GROW is to consider the operational problems
  associated with the IPv4 and IPv6 global routing systems"

now I'm confused! (easy to to do I guess ;)

> 
> - it is arguably part of the IAB's charter (well actually it could be part of

Possibly. But probably easier as a GROW item.

> rather than having this proposal come
> from an ICANN staff member, where there is some institutional overlap between
> the IANA contracting entity and the proponent of this work.

My institutional relationship at this stage should be considered moot. As
per the IETF Note Well I do this work as an IETF contributor and as an
engineer implementing systems that arguably will be required by the entire
operator community.

> 
> There is a substantive issue here about precisely what is recorded in these
> registries, and there is a case to be made for clear simplicity and

The PRI field is simple. (I don't like complex structures)
Putting the more specifics in the registry makes it simpler to use and read.

So this is clear simplicity. :)

> consistency, as distinct from constant adornment and tinkering in these
> registries, but as far as I can see this is a matter for the IAB to consider
> as a matter of registry policy as distinct from the GROW Working Group within
> the necessarily charter-limited scope of GROW's work.
> 

Again, a bit of confusion on my part. I thought that the IETF through the
working group' documents told the IANA what to put registries and how they
are to be structured? If that isn't the case, can someone please point me to
the clarifying documentation?

Cheers
Terry

_______________________________________________
GROW mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/grow

Reply via email to