[tor-talk] Persistent Tor start in Tails vs location aware Tor entry guards (LATEG)

[Spencer]

Hi,

> 
> sajolida:
> Unless I missed something, right now the entry guard
> concept is not explained nor advertised to the user at
> all.
> 

That is correct (:

> 
> As a user of Tor Browser or Tails, the only wait to
> learn about entry guards is, first to be aware
> of this behavior of Tor, and then to infer its status
> from the Tor Browser circuit view or Vidalia in Tails.
> 

Even then, the FAQ is the place they end up before bothering developers 
for more info.

> 
> location and entry guard
> 

No clear definition here, either, only context.

> 
> I'll keep this idea in mind while working on
> https://tails.boum.org/blueprint/network_connection/.
> 

Upon request for more detailed information, George Kadianakis awesomely 
shared these:

https://blog.torproject.org/blog/improving-tors-anonymity-changing-guard-parameters

https://gitweb.torproject.org/torspec.git/tree/proposals/236-single-guard-node.txt

https://gitweb.torproject.org/torspec.git/tree/proposals/259-guard-selection.txt

https://gitweb.torproject.org/tor.git/tree/doc/state-contents.txt

And said: "Unfortunately, there are not many resources about how Tor is 
picking guards. The source code and the emails above are your best bet.  
We hope that with proposal 259 we will formalize this part of Tor 
further and make it easier to analyze and reason about."

I have been reading as much as I can to understand, for the startup 
process, of course (:

It seems that wiping the contents of the 'state' file and restarting Tor 
forces a new guard selection.

And then there is this, where someone received similar info:
https://lists.torproject.org/pipermail/tor-talk/2016-January/039932.html

There are two more interesting reads there.

Wordlife,
Spencer