<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Tue, Oct 20, 2015 at 3:15 PM, Tom Hill <span dir="ltr"><<a href="mailto:tom.hill@bytemark.co.uk" target="_blank">tom.hill@bytemark.co.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On 20/10/15 14:02, James Blessing wrote:<br>
> Can we limit it this to a /X to see what the impact is before<br>
> throwing the entire remaining v4 space under a bus?<br>
<br>
</span>I was thinking the very same, actually.<br>
<br>
If we're thinking that the current policy has been too conservative,<br>
it seems like we should be cautious not to swing too far in the other<br>
direction (too liberal).<br><br></blockquote></div><div><br></div><div>Here's a thought experiment:</div><div><br></div><div>Set aside a /12 pool for this particular purpose.</div><div><br></div><div>This means that up to 1024 additional allocation requests may be made.</div><div><br></div><div>It means that it is predictable, and according to those who complain the most about the strict policy, should be more than ample enough to handle those who think they need more IPv4 space.</div><div><br></div><div>There would not need to be any further restrictions than those that are already in the policy and this proposal.</div><div><br></div><div>Pro:</div><div><br></div><div>- ensures that we don't accidentally "liberate" our RIR of its current pool</div><div>- ensures that small actors get a bit more</div><div><br></div><div>Con:</div><div><br></div><div>- still unfair to greater LIRs</div><div>- only a small pool, which risks being a "oh, cool, it's gone" experiment</div><div><br></div>-- <br><div class="gmail_signature">Jan</div>
</div></div>