Re: I-D Action: draft-carpenter-6man-lap-00.txt

[Mikael Abrahamsson <[e-mail]>]

On Wed, 13 Jun 2018, Lorenzo Colitti wrote:

> I don't see much point in pursuing this document, unfortunately. This is
> because:
>
>   1. I don't think there is value of this document unless it precisely
>   defines what a LAP is.
>   2. I suspect the working group will not be able to reach consensus on
>   any definition of LAP. This is because there will be participants that
>   claim that the prefix length for address assignment is already clearly
>   defined in RFC 4291 as being "64 bits long", and there will be participants
>   to whom that definition is unacceptable.
>
> The document says the goal of the document is not to rehash those opinions,
> but it does look very much like a rehash to me.

I interpret the document as trying to create a name for what we're 
typically arguing about, and naming it LAP. It doesn't try to define 
anything else or propose any changes.

While I think it helps to have agreed upon terminology when discussing 
things, the LAP might not help. There have been proposal in 5ganip mailing 
list about address mapping systems that would make non-contiguous blocks 
of addresses (prefix) available to the device (for instance a bunch of 
/96:es), which means this LAP concept doesn't apply, since it might now be 
of more interest to discuss the amount of aggregate addresses available to 
the device instead of the prefix length of the single prefix on-link.

I have no idea how to do that though... :)

-- 
Mikael Abrahamsson    email: [e-mail]