Re: New Tor distribution for testing: Tor Browser Bundle

[Scott Bennett <bennett@xxxxxxxxxx>]

     On Sun, 03 Feb 2008 16:37:20 -0800 Jacob Appelbaum <jacob@xxxxxxxxxxxxx>
wrote:
>Scott Bennett wrote:
>>      On Sun, 3 Feb 2008 22:19:54 +0100 "Michael Schmidt"
>> <schmidtm524@xxxxxxxxxxxxxx> top-posted:
>>> Steven, i suggest to make it hardcoded default and a Must, that each user,
                                 ^^^^^^^^^^^^^^^^^       ^^^^
>>> using this browser, is as well running an tor **exit** node,
>>> tit for tat. like emule partials: upload is a MUST.
>>> That would help a lot to have more tor-exit nodes.
>>> Thanks.
>>>
>>      That seems a tad impractical for a portable tor bundle, don't you
>> think?  If you take your computer to your friendly neighborhood coffee
>> shop, I doubt that you will find them so friendly as to be willing to
>> reconfigure their wireless router to let you run tor in server mode at
>> all, much less as an exit server.
>>      And who would determine the exit policy?  The exit policy would
>> probably have to be hardcoded as well.
>>      Where were you thinking that these changes would be made?  In the
>> tor source tree?  Or would a modified tor be packaged into the bundle?
>> Perhaps with a different version/release number?
>> 
>> 
>
>While it may be impractical because of NAT, I certainly don't think it's
>impractical to become a Tor server. It's a few lines in the torrc and
>you're off.

     Please reread what he wrote.  He said, "hardcoded" and "must".  Now,
it's true that tor's self-reachability testing would keep it from actually
running as a server if the appropriate NAT and RDR rules weren't in place,
it does seem wasteful to have it continually trying to complete the test
in situations where it never can.  It would be much better simply to let
it run as just a client in those cases.
     There is also the ethical consideration of tying up the available
bandwidth in a private business's services provided to its customers, but
for the benefit of people scattered at unknown locations all around the
globe, rather than of those customers.
>
>However, without the ability to punch a hole in your upstream NAT ( With
>UPNP or something else), I don't think it's very /useful/ for the Tor
>network. Though it's certainly possible without changes to the Tor source.

     But then those changes would not be hardcoded either, as the top-poster
demanded.
>
>The bundle ships with Vidalia, so in theory, it's just a check box away
>if the user didn't wish to be a server. Still, it seems impractical
>unless there's some way to actually *reach* the Tor server.
>
     But a check box, by the top-poster's request, would have to do nothing,
so that the server mode with some unspecified exit policy would be a "must".


                                  Scott Bennett, Comm. ASMELG, CFIAG
**********************************************************************
* Internet:       bennett at cs.niu.edu                              *
*--------------------------------------------------------------------*
* "A well regulated and disciplined militia, is at all times a good  *
* objection to the introduction of that bane of all free governments *
* -- a standing army."                                               *
*    -- Gov. John Hancock, New York Journal, 28 January 1790         *
**********************************************************************