Re: [PATCH] gnu: icecat: correct useragent to GNU GuixSD.

[ng0 <ng0@we.make.ritual.n0.is>]

Hi,

Brendan Tildesley <brendan.tildesley@openmailbox.org> writes:

> On 2016-09-10 19:18, Jan Nieuwenhuizen wrote:
>> Hi!
>>
>> Attached simple patch fixes spreading of theh terribly ignoring lie that
>> I would be running Windows NT.
>>
>> I'm probably displayng unimaginable ignorance here but it's beyond me
>> why a project like Firefox would hardcode such a default.
>>
> The user agent is a string that is freely provided to websites when you
> connect. Changing the default user agent in Guix's Icecat to Guix means
> that all Guix Icecat users will suddenly be identifiable by this datum,
> and automated surveillance networks will pick this up, bulking our
> browsing behaviour with other instances of the same  user id, using that
> to serve up advertising to us on sites or whatever, I haven't really
> read much on this.
> This is the current Icecat user string followed by the user string of my
> version of Tor browser:
>
> Mozilla/5.0 (Windows NT 6.1; rv:38.0) Gecko/20100101 Firefox/38.0
> Mozilla/5.0 (Windows NT 6.1; rv:45.0) Gecko/20100101 Firefox/45.0
>
> If we had Icecat 45 already, we would be helping Tor users be less
> trackable, but as you can see, just 4 characters are different, which is
> all that is needed to break anonymity. Perhaps there are still some
> people using Tor Browser 38 that we are assisting, but I suspect not
> many. I can't see how the version number would affect website behaviour
> realistically, so perhaps we could maintain our instance of Icecat
> patched with the latest Tor user agent instead, in order to assist Tor
> users? On the other hand it is kinda *bleh* to be advertising myself as
> a Windows user, skewing statistics, so It'd be nice to have a Linux,
> GNU/Linux, or Guix user agent, so that published OS usage statistics
> will display our mighty existence! User agents being changeable is
> meaningless unless there is at least one large cluster of users with
> some arbitrary string that can be chosen to hide with, so there may be
> greater value in keeping "Windows NT". I'm not sure, what do others think?
>
>

I agree and I am against applying this patch.
-- 
ng0
For non-prism friendly talk find me on http://www.psyced.org