<div dir="ltr">It is very possible that I misinterpreted Brian's original email. If this is the case, my apologies. If/when there is something to test, let me know and I will be happy to bang on it for a bit.<br></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Sep 8, 2015 at 12:56 AM, David Stainton <span dir="ltr"><<a href="mailto:dstainton415@gmail.com" target="_blank">dstainton415@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Paul Rabahy,<br>
<br>
Thanks for running the public grid so users can test out Tahoe-LAFS...<br>
<br>
Btw both you and Zooko misunderstood Brian Warner's e-mail; he doesn't<br>
want to take away the autodetect feature... he just wants it to not be<br>
the default if no listening address is specified... keyword AUTO will<br>
be used to specify autodetect... As Daira pointed out, this makes<br>
"tahoe create-note" slightly more difficult to use.<br>
<br>
On the other hand these changes will make Tahoe-LAFS way easier to use<br>
with Tor and other transports (I2p, ipv6 etc) who wish to run storage<br>
nodes at home and need the transport's NAT penetration property.<br>
<br>
I think there are political, economic and technical consequences to<br>
the NAT penetration method chosen... so I have this work in progress<br>
document to help explain some of these tradeoffs:<br>
<br>
<a href="https://github.com/telekommunisten/nat-penetration-tradeoffs" rel="noreferrer" target="_blank">https://github.com/telekommunisten/nat-penetration-tradeoffs</a><br>
<br>
It's far from complete...<br>
<br>
<br>
Cheers,<br>
<br>
David<br>
</blockquote></div><br></div>