[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #28531 [Community/Outreach]: Publish a snapshot of what PTs are needed for successful Tor use in each country
#28531: Publish a snapshot of what PTs are needed for successful Tor use in each
country
-------------------------------------------------+-------------------------
Reporter: arma | Owner: phw
Type: task | Status:
| needs_review
Priority: Medium | Milestone:
Component: Community/Outreach | Version:
Severity: Normal | Resolution:
Keywords: anti-censorship-roadmap-2020Q1 , | Actual Points:
s30-o22a2, s30-o23a2, ux-team |
Parent ID: #31279 | Points:
Reviewer: cohosh | Sponsor:
| Sponsor30-must
-------------------------------------------------+-------------------------
Comment (by cohosh):
Replying to [comment:18 phw]:
> I think YAML is more suitable for this task. How about the following
structure?
>
> {{{
> location_id: # Two-letter country code or ASN; in
this case "cn".
> infrastructure:
> website: false # Is torproject.org reachable?
> bridgedb_www: false # Is bridges.torproject.org reachable?
> bridgedb_moat: true # Is moat's cloud front reachable?
> relays: false # Are (a subset of) relays reachable?
> dirauths: false # Are directory authorities reachable?
> bridges:
> default: false # Are default bridges reachable?
> published_vanilla: false # Are vanilla bridges in BridgeDB
reachable?
> unpublished_vanilla: false # Are unpublished vanilla bridges
reachable?
> published_obfs4: false # Are obfs4 bridges in BridgeDB
reachable?
> unpublished_obfs4: true # Are unpublished obfs4 bridges
reachable?
> meek-azure: true # Does meek over Azure work?
> }}}
>
> * What other information do we want to capture in this snapshot? At the
very least, we want to collect information that can help Tor Browser make
informed circumvention decisions.
- I like the simplicity of binary true/false values for each of these
because it will help people make decisions on what to use. But I'm worried
it will be difficult for us actually decide which it should be, and we
should be clear about how we make the decision to finally say reachability
is `false`. Is it when 50% of bridgedb bridges are unreachable? For
unpublished bridges is it when we do our own ad-hoc tests on some bridges
we set up like in #29279?
- Can we add one for downloading Tor Browser? Might be nice to see which
places need GetTor. If we want to really expand our understanding of
GetTor we could include each of the link providers and our distribution
methods for the links.
> * Referring to bridges as "published" and "unpublished" is somewhat
misleading because bridges are never public. Any ideas for better terms?
Perhaps "bridge_obfs4" vs "unlisted_obfs4"? or listed vs. unlisted?
>
> * Censorship isn't necessarily uniform across a country and may differ
from AS to AS: `location_id` is either a country code (e.g., cn) or an ASN
(e.g., 1234). An ASN has preference over a country code.
Nice!
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/28531#comment:19>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs