[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

[tor-talk] Tor Weekly News â August, 21st 2013



========================================================================
Tor Weekly News                                        August 21st, 2013
========================================================================

Welcome to the eighth issue of Tor Weekly News, the weekly newsletter
that covers what is happening in the great Tor community.

Future of pluggable transports integration
------------------------------------------

While David Fifield was busy updating the Pluggable Transports Bundle to
match the âclassicâ bundle version 0.2.4.16-beta-1Â[1], several
discussions took place on how to better integrate pluggable transports
in the future.

bastik opened #9444Â[2], pointing out that âcurrently TBB with Pluggable
Transports are build separately, thus lagging behindâ. Having two
separate bundles is also a long standing usability issue, as often users
have tried to add âobfsâ bridges to their normal TBBÂ[3].

Mike Perry is fully aware of the issue and stated in the discussion that
his âlong term goal is to try to cram all of the pluggable transports
into The One True Bundle.â

This will require modifications to the new âTor Launcherâ component of
the TBB 3.x series in order to allow users to select the bridges and
pluggable transports they wish to use. Compromises might be needed on
how users should input bridges. BridgeDB recently stopped having the
âbridgeâ keyword in front of the addresses it repliesÂ[4] with as
Vidalia would not understand it. Mike Perry was thinking in exactly the
opposite direction: âtake bridge lines directly from bridgedb [â]
verifying only that they start with âbridgeââ. Maybe the transition
could be easier if Florian Stinglmayrâs patch to VidaliaÂ[5] was merged
so that current bundles would ignore the âbridgeâ keyword when entering
bridgesÂ[6].

In any case, Mike wants to solve these issues âbefore we release as
beta/stable, to minimize user confusion.â

Another tricky part of the âOne True Bundleâ solution is the bundle
size, making it harder to circumvent download restrictions through
emailÂ[7]. But, as Mike said, âeven if they donât, weâll probably have
to find some other solution anyway for gettor, because the intersection
of gettor users and PT users is probably high.â

   [1]Âhttps://www.torproject.org/docs/pluggable-transports.html.en
   [2]Âhttps://bugs.torproject.org/9444
   [3]Âhttps://bugs.torproject.org/9156
   [4]Âhttps://gitweb.torproject.org/user/isis/bridgedb.git/commit/792cfd9
   [5]Âhttps://github.com/n0la/vidalia/tree/master-bug/6724
   [6]Âhttps://bugs.torproject.org/6724
   [7]Âhttps://www.torproject.org/projects/gettor.html

Extended ORPort land in tor 0.2.5
---------------------------------

After more than a year and a half in the making, the Extended ORPort
mechanismÂ[8] has been merged by Nick Mathewson into the tor master
branchÂ[9]. This will allow pluggable transport proxies to exchange
arbitrary operational information and metadata with tor clients and
bridges.

Such plumbing was needed in order to make some pluggable transports
easier to use or to allow Tor to gather more data about the state of
the transports used.

obfsproxy has supported this new communication channelÂ[10] for a
little while and was only waiting for tor to catch up. George
Kadianakis thus asked obfsbridge operators to upgrade their tor to git
master to enable client statisticsÂ[11].

Once they do, their bridges will send statistics on users per transport
to the bridge authority, and they will be published on
metrics.torproject.orgÂ[12]. This helps track deployment of pluggable
transports in the future.

   [8]Âhttps://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/196-transport-control-ports.txt
   [9]Âhttps://gitweb.torproject.org/tor.git/commit/74262f15
  [10]Âhttps://gitweb.torproject.org/pluggable-transports/obfsproxy.git/blob/HEAD:/obfsproxy/network/extended_orport.py
  [11]Âhttps://lists.torproject.org/pipermail/tor-relays/2013-August/002477.html
  [12]Âhttps://metrics.torproject.org/users.html?graph=userstats-bridge-transport&transport=obfs3#userstats-bridge-transport

A new implementation for the web side of check.torproject.org
-------------------------------------------------------------

Arlo Breault wrote a new implementation for the web component of
check.torproject.org in the Go programming languageÂ[13], in response
to Roger Dingledineâs appeal: âCheck could really use some love. Any
volunteers please?â.Â[14,15]

There is already a ticket to replace the check.torproject.org servers
with Arloâs Go versionÂ[16]. Andrew Lewman stated again that âAs for
check.tpo website, it shouldnât exist at allâ, as it is an architectural
issue to âhave the entire tor browser userbase hit a single website to
learn âTor or notââ. Until all clients are changed to stop using check,
deploying a new code base would only make sense if it was at least able
to handle â500 requests per second on really busy timesâ. More
benchmarks are probably needed with Arloâs implementation.

On another front, tup, the initial author of TorDNSELÂ[17], has
resurfaced to offerÂ[18] to update the code to work with newer Haskell
environments after many years of silence!

  [13]Âhttps://github.com/arlolra/check/
  [14]Âhttps://lists.torproject.org/pipermail/tor-talk/2013-August/029306.html
  [15]Âhttps://bugs.torproject.org/9529
  [16]Âhttps://gitweb.torproject.org/tordnsel.git/commit/99d490
  [17]Âhttps://bugs.torproject.org/9204#comment:14

Tor exit crowdfunding
---------------------

Moritz Bartl from torservers.netÂ[18] posted an updateÂ[19] on their
ongoing crowdfunding campaign to support Tor exit bandwidth. The fund
just went over â3000, and there are still a few days left!

For more information, and ways to contribute, please visit the Indiegogo
pageÂ[20].

  [18]Âhttps://www.torservers.net/
  [19]Âhttps://lists.torproject.org/pipermail/tor-talk/2013-August/029431.html
  [20]Âhttp://www.indiegogo.com/projects/tor-anti-censorship-and-anonymity-infrastructure/

A Flattr-like incentive for Tor relays?
---------------------------------------

While torservers.net is presently collecting euros, George Kadianakis
asked for comments from the Tor community about âa practical crowdfunded
Flattr-like incentive scheme for Tor relaysâ, dubbed FlattorÂ[21].

Georgeâs proposal is meant to solve âone of the problems of scaling Tor
to tens of millions of usersâ, that âTorâs bandwidth capacity is
finiteâ.  He observes that âlately the bandwidth coming out of
Tor-friendly organizations (like torservers.net, universities, etc.)
seems to increaseâ and is worried that âTor might end up looking like
the Bitcoin network â where a number of organizations (mining pools)
drive the network.â

What George would like to see is incentives for contributing to the
network.  After studying schemes proposed in the past, all deemed âhard
to implement and deployâ, George proposes a simple approach: users can
opt to spend a fixed amount of bitcoins to support the Tor network, and
their donation will be divided according to the bandwidth of each relay.
Obviously, relay operators who wish to receive such contributions would
need to publish a bitcoin address, probably in the âcontactâ field.

There might be some concerns with such scheme, or any monetary
incentives scheme, as George summarized: âIf relay operators start
getting money for their bandwidth, we might end up with relay operators
that are just in for the money.  It might then be easier for a
three-letter org to persuade those relay operators to snoop on their
users (by giving them double the money they are currently getting).â

Moritz Bartl commentedÂ[22] that the idea was already quite close to
torservers.net current plan, to the extent that donations
were distributed âacross all participating organizations based on [â]
advertised bandwidth and a country-specific factor.â Moritz also pointed
out that similar discussions had already happened in the past when a
sponsor wished to fund faster exit relaysÂ[23].

George concluded his mail by saying that he is ânot even sure if such an
incentive scheme is a good idea, but posting bad ideas to mailing lists
is what the Internet is for, right?â

Feel free to join the discussion, or hack wildly.

  [21]Âhttps://lists.torproject.org/pipermail/tor-talk/2013-August/029419.html
  [22]Âhttps://lists.torproject.org/pipermail/tor-talk/2013-August/029421.html
  [23]Âhttps://blog.torproject.org/blog/turning-funding-more-exit-relays

Miscellaneous news
------------------

The new release of Orbot 12.0.3 comes with a shiny new icon and
graphics, bugfixes, and Tor 0.2.4.16-rc. You can download the update via
Google PlayÂ[24] or straight from Guardian Projectâs websiteÂ[25].

  [24]Âhttps://play.google.com/store/apps/details?id=org.torproject.android
  [25]Âhttps://guardianproject.info/releases/orbot-latest.apk

Andrew Lewman has published the financial reports of the Tor Project for
the year 2012Â[26].

  [26]Âhttps://blog.torproject.org/blog/transparency-openness-and-our-2012-financial-docs

Arturo has sent his report for July 2013Â[27].

  [27]Âhttps://lists.torproject.org/pipermail/tor-reports/2013-August/000313.html

Runa Sandvik reported on her trip to Black Hat & DEF CONÂ[28]. She
managed to fill âthe Penn & Teller theater (~1500 people)â for a
talkÂ[29] about âthe safety of the Tor network which focused on network
diversity, relay operators, and misbehaving relays.â The former Tor GSoC
student Brandon Wiley also gave an updateÂ[30] on Dust â âan Internet
protocol designed to resist a number of attacks currently in active use
to censor Internet communication.âÂ[31]

  [28]Âhttps://lists.torproject.org/pipermail/tor-reports/2013-August/000312.html
  [29]Âhttps://www.defcon.org/html/defcon-21/dc-21-speakers.html#Sandvik
  [30]Âhttps://www.defcon.org/html/defcon-21/dc-21-speakers.html#Wiley
  [31]Âhttps://github.com/blanu/Dust/

Karsten Loesing has made progress on âexperimenting with a client and
private bridge connected over uTPâÂ[32]. The connection can be
established, but strange timing issues remain to be solved.

  [32]Âhttps://trac.torproject.org/projects/tor/ticket/9166#comment:23

George Kadianakis has sent two new proposals to improve hidden service
identity key securityÂ[33] and prevent address enumerationÂ[34]. TWN
will cover these proposals in detail once the draft deployment strategy
is published. Feel free to help refine the proposals in the meantime!

  [33]Âhttps://lists.torproject.org/pipermail/tor-dev/2013-August/005279.html
  [34]Âhttps://lists.torproject.org/pipermail/tor-dev/2013-August/005280.html

Help Desk Roundup
-----------------

Users experience confusion when trying to update the Tor Browser Bundle.
Users are not always aware that the Tor Browser Bundle does not have an
autoupdate function. Some users will download the latest release from
the Tor Project website, then ask âOk, what do I do now?â. We recommend
closing the browser, then deleting oneâs current Tor Browser folder
before unpacking the new download.

One person asked for help while using the Pirate Browser.
Torrent-sharing website The Pirate Bay released the Pirate Browser this
week as a fork of the Tor Browser Bundle. The Pirate Browser is not
endorsed or recommended by the Tor Project. It is unclear what the
advantages are compared to using the Tor Browser Bundle and no source
code is available.



This issue of Tor Weekly News has been assembled by Lunar, dope457,
malaparte, mttp, Karsten Loesing, and harmony.

Want to continue reading TWN? Please help us create this newsletter.
We still need more volunteers to watch the Tor community and report
important news. Please see the project pageÂ[35], write down your
name and subscribe to the team mailing-listÂ[36] if you want to
get involved!

  [35]Âhttps://trac.torproject.org/projects/tor/wiki/TorWeeklyNews
  [36]Âhttps://lists.torproject.org/cgi-bin/mailman/listinfo/news-team

Attachment: signature.asc
Description: Digital signature

-- 
tor-talk mailing list - tor-talk@xxxxxxxxxxxxxxxxxxxx
To unsusbscribe or change other settings go to
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk