<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On 9 June 2015 at 14:53, Edward Lewis <span dir="ltr"><<a href="mailto:edward.lewis@icann.org" target="_blank">edward.lewis@icann.org</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 6/9/15, 7:42, "Mark Andrews" <<a href="mailto:marka@isc.org">marka@isc.org</a>> wrote:<br></span>ents that can be referenced<br>
separately (like in RFP's and contracts). I found that trying to make<br>
code prefer newer technologies over old by fiat seems to backfire (like<br>
the way DNS used to prefer v6 over v4 and now seems to have reversed,<br>
looking at some observed behavoral studies).<br>
</blockquote></div><br></div><div class="gmail_extra">interesting this crops up, and in a thread with Mark. I am told he recently confirmed that there is no systematic deliberate biassing towards V4 in the code: its just shorter RTT selection. </div><div class="gmail_extra"><br></div><div class="gmail_extra">I wonder if there is a non-intentional bias against V6 eg the order the calls are made, or a lazy evaluated IF statement logic, because we see overwhelmingly more V4 than V6 on dual-stack NS with no cached state.</div><div class="gmail_extra"><br></div><div class="gmail_extra">-G</div></div>