Re: [Tails-dev] Please review and merge feature/obfsproxy

Delete this message

Reply to this message
Author: sajolida
Date:  
To: The Tails public development discussion list
Subject: Re: [Tails-dev] Please review and merge feature/obfsproxy
On 22/11/12 19:02, anonym wrote:
> 22/11/12 18:26, anonym wrote:
>> 13/11/12 20:25, Ague Mill wrote:
>>> anonym:
>>>> 12/11/12 15:11, anonym wrote:
>>>>> 03/11/12 09:08, intrigeri wrote:
>>>>>> Hi,
>>>>>>
>>>>>> anonym wrote (02 Nov 2012 20:26:34 GMT) :
>>>>>>> Basic (perhaps even experimental as it currently lacks documentation)
>>>>>>> support for obfsproxy has been added in the branch feature/obfsproxy.
>>>>>>> Please review and merge it into devel.
>>>>>>
>>>>>> We agreed at the Tails summit to not merge new features before their
>>>>>> documentation is ready. For the record, this is what allows us to
>>>>>> squeeze the delay before feature freeze + RC1 and RC2, because it's
>>>>>> now dedicated to translation work, rather than (like we used to do) to
>>>>>> doc writing + translations.
>>>>>
>>>>> Now done:
>>>>
>>>> I should perhaps have pointed out that I'd really to see this branch
>>>> merged for Tails 0.15.
>>>
>>> Confirmed working. Merged.
>>
>> I've pushed some further improvements to the docs after a discussion on
>> #tails. Please review and merge.
>
> There was some concerns raised against commit bb1c08e on #tails:
>
>     (18:35:29) velope: finally, i do not think it is an improvement
>     to change 'you think you need to use bridges' into the
>     recommendation 'you should use bridges'
>     (18:36:16) velope: using bridges is not necessarily better, and
>     users should not use them merely because they want an extra layer
>     to hide more.
>     (18:37:22) velope: bridges are less reliable and tend to have
>     lower performance than regular guards, so a user has to actively
>     keep getting and configuring them, as opposed to letting tor
>     choose guards.
>     [...]
>     (18:50:33) velope: i really believe that circumvention and hiding
>     are so different that they should not be combined in a single
>     recommendation.
>     (18:51:21) velope: and my concern with 'you should use bridges'
>     in anything except cut-and-dry circumvention is that you're
>     making a security decision for users where the facts are not
>     fully understood.
>     (18:52:11) velope: that is why i prefer something like 'you wish
>     to use bridges', which leaves the power and responsibility
>     completely with the user.
>     (18:54:15) velope: of course users tend to not want full power
>     and responsibility, they want a simple answer. mais c'est la vie.

>
> What do you think? My reading of the section in question is "if Tor is
> censored or dangerous, use bridges cause they make it harder for others
> to see that you use Tor which helps in those situations", which seems
> like a sound recommendation to me.


I agree with you. Still, I changed the "should" into a "might" and
increased the visibility of the warning explaining that bridges are not
a perfect protection to hide you as a Tor user.

I also mentioned the limitations in reliability and speed.

> I just thought that this could be some food for though for whoever
> reviews these doc changes.
>
> Cheers!


I pushed some work on that page. See 68069a3..22aedce.

I'm pretty such this is a bad timing for submitting documentation
changes after rc1 but I just found out about that while clean a bit my
inbox.