Sponsor3 2018 outcomes

Tor is a privacy and security system that helps people around the
world use the internet in safety. The Tor network is made up of 8000
volunteer-run relays, which carry traffic for millions of daily users,
including ordinary citizens who want protection from identity theft and
prying corporations, corporations who want to look at a competitor's
website in private, people around the world whose internet connections
are censored, and even governments and law enforcement.

However, the internal architecture of Tor is complicated. This
complexity creates performance problems, and it makes security analysis
difficult. While Tor is by far the most studied internet privacy network
in terms of attention from academic researchers, there is still room
for improvement.

This research project studied and implemented ways to simplify the
underlying Tor software, as well as ways to make it easier for external
researchers to participate in analyzing and improving the Tor design. We
highlight three areas of progress here.

Our first area focused on making the design and implementation of the core
Tor software more modular. Dividing the software into different modules
makes it easier to analyze each module by itself, for example to improve
its performance or to improve its security. These steps have in turn made
it easier for external researchers to contribute to a particular module
without needing to worry as much that their work will end up impacting
other pieces of Tor.

One concrete area where modularity has helped is in a part of the Tor
protocol called the circuit handshake, which uses cryptography for two
purposes: first to let Tor users make sure that they are connecting via
the relays they chose ("authentication"), and second to make sure that no
single relay gets to learn what websites the user visits on the internet
("confidentiality"). Tor's current circuit handshake design is not strong
enough to protect against future quantum computers; but mathematicians are
still working on inventing techniques that are both quantum-safe and also
practical to deploy. So while we wait to see which new technique emerges
as the consensus choice, in the mean time we have generalized Tor's
circuit handshake design to make it easy to switch to new techniques.

Our second research area focused on establishing and publishing
guidelines for external research groups who propose major changes to
the Tor design. Historically many different research groups have written
high-profile papers suggesting Tor changes, but the incentive structure
in academia doesn't reward authors for following through on implementing
their ideas and making sure they work for real users in the real world. We
have accumulated a set of advice and useful tips for researchers hoping
to see their Tor research transitioned into the deployed Tor software.

More broadly, as part of our outreach to interdisciplinary research
groups, we directly worked with other research groups in the form of
site visits: we gave guest lectures on various aspects of Tor, mentored
graduate students doing Tor research, and generally helped make sure
privacy research at other institutions is solving real and important
problems.

Our third focus area was on establishing the Tor Research Safety Board,
which is a group of researchers (both internal and external) who study
Tor, and who want to minimize privacy risks while fostering a better
understanding of the Tor network and its users. We aim to accomplish this
goal in three ways: developing and maintaining a set of guidelines that
researchers can use to assess the safety of their Tor research; giving
feedback to researchers who use our guidelines to assess the safety
of their planned research; and teaching program committees about our
guidelines, and encouraging reviewers to consider research safety when
reviewing Tor papers. More than a dozen research groups have voluntarily
sought our advice to make sure their Tor research project is as safe as
they can make it.

In addition to these three areas of technical improvement, we also
continue to gather and provide statistics about the live Tor network,
so researchers can better understand the right problems to solve. We make
all of the raw data available to the public so anybody can do analysis.

Looking forward, contributions beyond science and technology will
continue to be a core component of the Tor Project's work. The Tor
network draws a diverse set of users, ranging from ordinary citizens
in Western countries to corporations, law enforcement and military, but
also bloggers in oppressive regimes around the world. Tor gives people
a voice when otherwise they wouldn't be able to communicate or interact
with the broader world. Anonymity online also levels the playing field
with respect to race, culture, and background. In a world where private
and state surveillance and censorship are increasingly growing concerns,
systems like Tor are critical tools for ensuring safe communications.