[tor-dev] Fwd: Orbot roadmap feedback

[Georg Koppen]

Nathan Freitas:
> On Tue, Jan 12, 2016, at 10:48 AM, Georg Koppen wrote:
>> Nathan Freitas:
>>> - Overall improved configuration / settings UI to make tuning Orbot a
>>> better, simpler experience... this is an expansion of the new exit
>>> country selector in Orbot v15.1, but also includes managing things like
>>> network usage and so on.
>>
>> Could you explain that point a bit more, what you currently have and
>> what you plan to do? Especially the exit country selector seems kind of
>> scary to me but maybe I am just missing some bits. I looked at the
>> changelog you sent to tor-talk (in the alpha 1 release mail) but did not
>> find anything related.
> 
> The broader idea is to determine which Tor torrc settings are relevant
> to the mobile environment, and that could use a more intuitive user
> interface than the empty text input we currently offer in our Settings
> panel. This may also mean implement a slider type interface mechanism
> similar to Tor Browser. In addition, users are interested in being able
> to control which network types (wifi vs 3g) Orbot runs on, in order to
> conserve spending on bandwidth.

Sounds good to me.

> The Exit country selector option (basically a list of countries that
> starts with "World" as the option) is definitely controversial, but it
> is also something that comes up over and over again as a feature
> request. It is expected especially of a VPN-style app, which Orbot now

True. We see these requests with Tor Browser as well.

> is. It may be we only enable it when the VPN feature is enabled, and
> disable it when Tor is used directly via SOCKS from the Orfox browser.
> We may also need to just sufficiently warn and inform the user so they
> can decide for themselves what to do.

Well, warning/informing the users about possibly harming themselves is
one thing (which is already tricky in this case given the complexity of
the matter). But how do you convey the problem that all the other users
of the Tor network could be affected as well if a bunch of users is
tweaking the exit node selection based on whatever reasoning? And this
does not only have anonymity ramifications but
performance/load-balancing implications as well. How is a user supposed
to make an informed decision in this case?

Georg


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: OpenPGP digital signature
URL: <http://lists.torproject.org/pipermail/tor-dev/attachments/20160114/073398bd/attachment.sig>