[arin-ppml] ARIN-prop-167 Removal of Renumbering Requirement for Small Multihomers

[Joe Maimon]

Jo Rhett wrote:
> On Apr 30, 2012, at 7:38 PM, William Herrin wrote:
>> it'll have to renumber. Which means I personally will have to deal
>> with the pain.
>
> Now I'm going to go back to the devil's advocate side again, and also
> point out that the pain of renumbering is not the horror, the horror,
> the horror that everyone makes it out to be. Painful, and sometimes slow
> when there are lots of external dependancies, but it's really not as bad
> as people make it out to be ;-)
>
> --
> Jo Rhett
> Net Consonance : net philanthropy to improve open source and internet
> projects.
>
>

I suspect it would be more significant to note which of us have not 
interacted with a significant renumbering event, rather than the opposite.

The less often people renumber the more painful it is likely to be when 
they do so, not just for themselves, but for everyone else who may be 
faced with it at one point.

This is due to the pain being wholly self inflicted, in that 
dependencies are formed and grown under the assumption of address longevity.

To a much larger extent than normally practiced, varying amounts of this 
self inflicted pain is avoidable with forethought, planning, and self 
discipline.

Attributes which are usually far less prevalent then they should be.

Attributes which are not recognized as valuable without the 
understanding and experience of renumbering, and that it is neither 
always predictable nor avoidable.

The greater the emphasis placed on preparing for renumbering as opposed 
to avoiding it, the more likely that workable technology that properly 
abstracts the task is developed and deployed.

How renumber-able are the typical IPv6 deployments?


Joe