[ppml] Policy Proposal: IPv4 Soft Landing

[Stephen Sprunk]

Thus spake <Dan.Thorson at seagate.com>
>> 255/8 is tough to unreserve in practice for the same reasons
>> as 0/8 and 127/8, so it's probably not 240/4 but rather 240-
>> 254/8.  If we're going to do that, we might as well switch 225-
>> 238/8 to unicast too -- only 224/8 and 239/8 see much use in
>> the (miniscule) multicast world, and the exceedingly few
>> oddballs using 225-238/8 could migrate to the remaining two
>> /8s with minimal hassle.
>
> There is danger in the assumption that there is a "(miniscule)
> multicast world".  Whereas this is likely true on the Big-I Internet,
> multicast is widely used in the corporate manufacturing world.

I'm aware of that; I've deployed plenty of it myself for IP/TV and stock 
trading systems.  However, _all_ of the registrations for public multicast 
use are in 224/8, and anyone using multicast privately should be in 239/8. 
I excluded both of those blocks.

I now see that 232/8 and 233/8 are assigned (I'd forgotten SSM and GLOP), so 
remove those from my list as well.

225-231/8 and 234-238/8 are still completely reserved and nobody should be 
using them; if folks are, they'll deserve whatever problems they have --  
just like folks who randomly chose reserved unicast addresses got burned 
when those addresses were later assigned.  There's one particularly famous 
case of that (Apple, IIRC).

> Who is prepared to pay the price of lost connectivity when
> traditional multicast IP's are suddenly deployed as end-user
> IP's?  Who will be modifying all my legacy system's IP
> stacks?

Ditto for the arguments in favor of defining 240/4 to be unicast.  It's 
either workable enough of an idea that we could deploy it in general use 
before v4 is otherwise exhausted, or it's so unworkable that it's not even 
worth standardizing for private use.  Private networks already have vast v4 
space available, both unicast and multicast, and if they're greenfields they 
can use v6 and get a virtually unlimited amount of space.

My proposal for repurposing multicast space was more of a reductio ad 
absurdum argument, hence my reference to publication on 1 Apr.  I should 
have made it clearer.

S

Stephen Sprunk      "Those people who think they know everything
CCIE #3723         are a great annoyance to those of us who do."
K5SSS                                             --Isaac Asimov