[tor-bugs] #1610 [BridgeDB]: Turn mail requests into ’subscriptions’

[Tor Bug Tracker & Wiki]

#1610: Turn mail requests into ’subscriptions’
--------------------------+----------------------------
     Reporter:  phobos    |      Owner:  isis
         Type:  task      |     Status:  accepted
     Priority:  normal    |  Milestone:
    Component:  BridgeDB  |    Version:
   Resolution:            |   Keywords:  bridgedb-email
Actual Points:            |  Parent ID:
       Points:            |
--------------------------+----------------------------
Changes (by isis):

 * cc: isis (added)
 * keywords:  SponsorL => bridgedb-email
 * status:  new => accepted
 * owner:   => isis


Comment:

 I am against the idea of sending plaintext subscription emails to users
 for several reasons.

  1. I don't want a database of bridge users on the server. I would prefer
 not to have ''any'' information at all on BridgeDB's users. The
 SocialDistributor, a.k.a. rBridge, (see #7520) would be a significantly
 safer way -- for the bridges, bridge users, and for me maintaining the
 BridgeDB server and service -- to implement automatic bridge retrieval.

  2. [https://lists.torproject.org/pipermail/tor-talk/2014-July/033870.html
 In light of the recent revelations] of
 [http://daserste.ndr.de/panorama/xkeyscorerules100.txt XKEYSCORE targeting
 of BridgeDB and various torproject.org servers], I would prefer to send
 ''less'' emails. Not more. The idea now is "send only enough emails to get
 the user's Tor working, then encourage them to use the web interface".

  3. As arma [comment:9 pointed out], we don't really churn bridges like we
 used to, so with this approach, all the bridge in the email hashring would
 likely eventually get emailed to everyone, and they'd have the entire
 hashring. (And due to point !#2, so would the NSA and several other
 intelligence agencies.)

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/1610#comment:11>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online