(a) Contiguous address block allocation is not ensured by APNIC when
       an organization goes back to APNIC for further allocation
       (reapplying after more than one year)

I support addressing this potential problem -- e.g. through propositions 98 or 99.

 

I agree with this point.  I believe that with little to no changes, prop-98 and/or prop-99 will provide most of the benefit that the proposer of prop-100 is seeking.

From the feedback on the list so far, I would advise that the proposer take a look at these other proposals and determine if they fulfil his requirements.  It may be the case that through small changes, these other proposals (which have received less negative feedback) could satisfy the requirements equally well.

I agree.  I still oppose prop-100.

I'm also going to say something slightly heretical:

I'm not sure I buy into the "minimising global routing table" goal anymore.  We say it, then we deaggregate because no one has the teeth to go and say "don't do that". 

If someone has two IPv6 prefixes rather than one.  Or three or four instead of one.  How's that going to change the world if they're deaggregating down to /32s or longer anyway?   

Currently I work for an organisation with a whole lot of different IPv4 allocations over time.  Managing that internally to work out what customers get what isn't that much harder than a single allocation might be.  It's just entries in a database.  If we got a second different IPv6 prefix we'd just chop that up and put it in the allocator according to where there was demand that needed it.  It's the same work.

MMC