<div dir="ltr"><div dir="ltr"><div>+1 to ULA should go extinct. Why do they even exist? How have they helped global native IPv6 adoption (without NAT66/NPTv6)?<br clear="all"></div><div><br></div><div>+1 to GUA being the default standard if IPv6 is called into play.<br></div><div><br></div><div>+1 to using GUA on "private" networks – future-proofed with GUA for future <b>inter</b>connectivity<b>.</b><br></div><div><b><br></b></div><div>+1 to IPv6 being about the number of networks (/64s) and not the number of "hosts"; the latter is, without a doubt, IPv4-centric.<br></div><div><div dir="ltr" class="gmail_signature"><div dir="ltr"><font color="#000000" face="arial, sans-serif"><br><b>--</b><br></font><div><font color="#000000" face="arial, sans-serif">Best Regards</font></div><div><font color="#000000" face="arial, sans-serif">Daryll Swer</font></div><div><font color="#000000" face="arial, sans-serif">Website: <a href="https://l.shortlink.es/l/a1abea4cf890b4fe45d3ca3253ffe607875c37ed?u=2153471" target="_blank">daryllswer.com</a></font></div></div></div></div><br></div><br><img width="0" height="0" class="mailtrack-img" alt="" style="display:flex" src="https://mailtrack.io/trace/mail/36fc7269db318773c815200dd71c395b3f20b578.png?u=2153471"><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Mon, 29 Jun 2026 at 23:16, John Osmon <<a href="mailto:josmon@rigozsaurus.com">josmon@rigozsaurus.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">> Current qualifications for ARIN to assign an IPv6 /48 block for<br>
> off-Internet use is roughly:<br>
> <br>
> 1. Have 2000 hosts, or<br>
> 2. Submit a network plan explaining why ULA addresses are not good enough.<br>
> <br>
> More than a single /48 requires multiple sites and the above.<br>
<br>
When using IPv6 the trend has been to discourage counting hosts, but<br>
rather counting networks. Is it too much to ask the policy to start<br>
drifting in the direction?<br>
<br>
<br>
> There's more nuance and some corner cases (see NRPM 6.5.8), but that's<br>
> the general shape of it.<br>
<br>
Some of the explanatory notes in 6.5.8.1 seem to reinforce the 2000<br>
hosts idea, but also has a couple of important "loopholes" that make<br>
justification easy if your network needs unique addresses. (This is a<br>
GOOD THING in my eyes.)<br>
<br>
<br>
> Is that about right for what the policy should be? Should the rules be<br>
> more restrictive? Less? Where would you set them and why?<br>
<br>
I think unique addressing is important, so I'd set the rules pretty<br>
loose.<br>
<br>
We could plan to re-evaluate when allocation of 2003::/3 hits something<br>
like 25% and 50%?<br>
<br>
<br>
_______________________________________________<br>
ARIN-PPML<br>
You are receiving this message because you are subscribed to<br>
the ARIN Public Policy Mailing List (<a href="mailto:ARIN-PPML@arin.net" target="_blank">ARIN-PPML@arin.net</a>).<br>
Unsubscribe or manage your mailing list subscription at:<br>
<a href="https://lists.arin.net/mailman/listinfo/arin-ppml" rel="noreferrer" target="_blank">https://lists.arin.net/mailman/listinfo/arin-ppml</a><br>
Please contact <a href="mailto:info@arin.net" target="_blank">info@arin.net</a> if you experience any issues.<br>
</blockquote></div></div>