[tor-relays] BWauth no-consensus state in effect

[starlight.2015q3 at binnacle.cx]

Maybe geo-location would not be so great
because two networks in the same physical
area might have relatively poor connectivity
to each other.

Aggregated IP block might be the ticket.
CIDR-Report has both actual and suggested
netblock aggregations.

http://www.cidr-report.org/as2.0/#Gains

Shows 562259 reduced to 253420 advertisements.
Perhaps too fine-grained, but 250k is not
so much bigger than the 45k ASN count.

Idea is to anchor metrics to maximum
size BGP routable blocks, which ought to
correlate performance-wise down to all
the IPs in each block.


>>BWauths are continuously pairing relays for
>>measurements, and perhaps metrics from that
>>could be mapped to autonomous system numbers
>
>. . .scratch AS, geo-location is
>better and MaxMind specializes in that
>
>Pinging a FiOS relay in LA takes
>75ms while a close-by relay takes
>8ms.  Both are in AS 701.