Le 2013-02-02 à 16:02, Fernando Gont <[email protected]> a écrit :

> On 02/02/2013 07:40 AM, Rémi Després wrote:
>>> Both are *experiments*.  Neither is a standard.
>>> 
>>> With respect to Softwire, they decided to standardise
>>> something other than 4rd.  Had they decided to 
>>> standardise 4rd, my views would be different, 
>>> but that wasn't the decision actually made in Softwire.
>>> 
>>> For *experiments*, a small shared use experimental 
>>> allocation *under RFC-3692 rules* should be both 
>>> sufficient and an appropriate approach -- and also
>>> can be used with any other experiments that might arise 
>>> in future.
>> 
>> Different understanding.
>> 
>> Making the reservation for one design, experimental or not, is a guarantee 
>> that no future design with conflict with it, experimental or not.
> 
> The point is that if't an experiment, you don't really want to
> permanently-asign any resources to it. -- that's probably why run
> suggested a "shared allocation".
> 
> If the experiment turns out to be successful, you get the actual assignment.

As answered to Ran:
- Making the reservation for one design, experimental or not, is a guarantee 
that no future design will conflict with it, experimental or not.
- A central registry is the appropriate tool for this.

Besides reserving 1/2^14 of the unused space, for which there hasn't been any 
request in more than 6 years now, shouldn't be considered a big deal (IMHO).

Regards,
RD

 
> 
> Cheers,
> -- 
> Fernando Gont
> SI6 Networks
> e-mail: [email protected]
> PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492
> 
> 
> 
> 
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to