[metrics-team] OnionStats - roadmap?

[Anathema]

On 08/08/2016 15:01, David Fifield wrote:
> [...] Also, the choice
> of relays/bridges seems arbitrary--why not return the first 10 results
> from each country, or the first 10 from each AS, or the first 10 from
> each distinct Tor version? Those don't seem any more or less useful to
> me than relays/bridges.
> 

I don't know about that, the protocol is already built this way, and as
a logical entity I think we should think about it as relay/bridge, but
maybe Karsten (or the one who worked on the protocol) may know more.

> 
> So what does ElasticSearch do, when comparing two arrays, say?
>

It should compare char by char and sorts accordingly.

> 
> I think that this search feature is much less useful if it matches
> anywhere other than at the beginning of the fingerprint. It's just going
> to be a 2^−16 random sample of the search space, on top of the result
> you want.

It's not a big deal for ElasticSearch since it's the reason why it has
been built. Plus, I think it's useful (sometimes you may have only 1 hex
block and don't know which block it is).


-- 
Anathema

+------------------------------------------------------------------+
   GPG/PGP KeyID: CFF94F0A available on http://pgpkeys.mit.edu:11371/
   Fingerprint: 80CE EC23 2D16 143F 6B25  6776 1960 F6B4 CFF9 4F0A

   https://keybase.io/davbarbato
+------------------------------------------------------------------+