[tor-talk] Tor Weekly News — October 2nd, 2013

Joe Btfsplk joebtfsplk at gmx.com
Wed Oct 2 15:44:16 UTC 2013

On 10/2/2013 7:00 AM, harmony wrote:
> ========================================================================
> Tor Weekly News                                        October 2nd, 2013
> ========================================================================
> On September 28th, Mike Perry released the fourth alpha of the new Tor
> Browser Bundle 3.0 series [1].... It also fixes a fingerprinting issue
> by randomizing the TIMESTAMP sent when establishing an HTTPS connection.
Why limit the randomization of browser characteristics or other data to 
just the timestamp?

Why not randomize certain other data or browser characteristics (that 
wouldn't break functionality)?  This goes back to my post containing the 
concept of it being easier to achieve chaos than perfection.

I'm not sure what would / wouldn't surely break some functionality; but 
things like user agent (or parts of it); time zone; randomly spoofing 
system fonts, even w/ Flash & Java disabled, etc.  Or other things that 
are easy to randomize & aren't likely to break functionality.

More information about the tor-talk mailing list