Re: new version of draft-narten-ipv6-3177bis-48boundary-03.txt

[Thomas Narten <narten@us.ibm.com>]

Hi Iljitsch.

> "The above-mentioned RFC3177 goals can easily be met by giving home  
> users a /56 assignment by default."

> I disagree. A /56 is still wasteful if we assume that home users are  
> only going to need a very small number of subnets.

To be clear, the wastefulness of a /56 was not a concern of
3177. Remember, it recommended a /48. So moving to /56 would be less
wasteful and I think the statement is correct. :-)

I understand your point  that a /60 would also be OK for home end
sites.

Couple of points.

The sentence in the document isn't technically recommending a /56. It
is simply saying it would meet the requirements. So, it could be
changed to something like "by giving home users a /56 or /60
assignment by default". I think I would be OK with that, but I suspect
others would be less willing to accept that.

I too think one could argue that a /60 for typical home sites is
sufficient _today_. But I also don't know that we really need to go
this far. Going from /48 to /56 is the big savings.

> Additionally, many people will be tempted to start using a /56 for  
> purposes that really need more subnets than about 250.
> Renumbering when a network needs its 257th subnet will be
> painful. It would be much better to a have a clear distinction
> between very small networks and all other networks, and not have a
> size that is still to large for most while at the same time being
> too small for some, even if they don't realize it at first.

> So I'd like to see /60 for consumers and /48 for anyone who feels /60  
> isn't enough.

Note: it is the RIRs/ISPs who make this decision, not the IETF. So I
don't think it is too useful to get too wrapped up over the exact
value specified in _this_ document.

Thomas