Sponsor 30 - Empowering Communities in the Global South to Bypass Censorship

This is the master ticket for the whole Sponsor 30 - Anti-censorship technologies.

Project Period: September 20, 2019 - November 30, 2021

The goal of the proposed project is to empower human rights defenders in the Global South by giving them the tools they need to do their important work.

With this project we will:

1. Improve censorship event detection and reporting. We will improve OONI Probe to better understand how, when, and where censorship is taking place.

2. Ensure users have the best option for their needs to bypass censorship. We will determine how to distribute bridges that are optimized for performance and speed to activists trying to access the internet in places where the Tor network is blocked.

3. Inform human rights defenders when censorship is happening and how to bypass it. We will combine the censorship event data from OONI with our circumvention technology to make it easier for activists to know what is going on and what to do.

Objectives

• Objective 1: Support censorship circumvention through the improvement of network measurement methodologies aimed at detecting the blocking of circumvention tools

• Objective 2: Ensure users in target countries have access to the best Tor bridge options for circumventing censorship

• Objective 3: Improve Tor Browser and bridges.torproject.org experience for human rights defenders under censorship

Expanding on the objectives 2 & 3

Objective 2: Ensure users in target countries have access to the best Tor bridge options for circumventing censorship. Activities under this objective aim to improve aspects of bridge distribution and selection methods. We seek to automate bridge distribution for our users as much as possible, but some distribution mechanisms require minimal user interaction, like solving CAPTCHAs. One example of an active distribution mechanism that is as automated as possible is “moat,” which is already integrated into Tor Browser. The fact that this distribution method is integrated into Tor Browser minimizes the number of hoops our users have to jump through. When possible, we hope to integrate distribution methods developed in this project into Tor Browser, minimizing the impact on the user. With special focus on human rights defenders in the Global South, we will increase access to Tor bridges and expand the network. Specifically, we will:

O2.1 - Create an evaluation framework and collect data to better monitor and evaluate current bridge selection and distribution processes.

• Create an evaluation framework for bridge distribution and selection methods.
• Evaluate distribution and selection methods for human rights defenders in target regions.
• Identify which bridge selection and distribution methods are most used in targeted regions.

O2.2 - Improve user experience and user interface of bridges.torproject.org.

• Evaluate the user experience of our bridge download site, bridges.torproject.org, and provide a report with problems encountered by human rights defenders on the ground.
• Identify solutions for user problems reported above and implement them to make the portal easier to understand and use successfully. We will extend bridgeDB to keep track of the number of failed user interactions, which will allow us to understand how the user experience of bridgeDB evolved over time.
• Support localization of bridges.torproject.org. Tailor additional localization strategies based on community feedback, requests, and data collected during the project timeframe.

O2.3 - Develop new and/or improve existing bridge selection and distribution strategies.

• Develop new and/or improve existing bridge selection and distribution strategies based on data collected about successful, effective methods per evaluation during O1.1.
• Develop methods to present bridges to users based on their location, potentially incorporating relevant censorship data published by OONI. If successful, we may be able to see an increase in the number of bridge users per country.
• Improve ability for bridgeDB/authority to test bridges that expose a pluggable transport.
• Update bridgeDB/gettor1 to give region-specific recommendations for PT and bridges.

O2.4 - Boost security by increasing the number of bridges run by volunteers and collective entities through improvements to onboarding and better communications. We aim to attract volunteers to run at least 100 new bridges, increasing the number of bridges by approximately 10%, which means a load decrease of approximately 11% across all bridges. In addition to taking load off of existing bridges, these new bridges will be unblocked and have a dramatic benefit for censored users. Our Tor Metrics portal tracks the number of bridges and the number of bridge users over time. We will refer to these numbers throughout the project’s life span to track our progress. To complete this activity, we will:

• Improve documentation on how to set up a bridge server and different pluggable transport bridge servers.
• Create scripts and configuration code for setting up a bridge on cloud providers to make it easier for operators to launch a new bridge.
• Promote workshops on how to set up a bridge at relay operator meetups.
• Improve documentation of bridgeDB--the code behind selecting and distributing bridges.
• Increase stability and resilience of bridge authority and bridgeDB by exploring and implementing decentralization of those services.

Objective 3: Improve Tor Browser experience for human rights defenders under censorship. As it stands at the writing of this proposal, there is no notification or error system within Tor Browser that tells a user that they may be experiencing censorship (vs. any other host of connectivity, speed, or hardware problems when Tor fails to connect). This lack of clarity can lead to frustration and possible abandonment of the tool. Under this objective, we will create easy-to-understand notifications within Tor Browser for when we detect censorship. The goal of this objective is to better alert users of detected censorship and then to provide clear pathways for human rights defenders to effectively circumvent this censorship, with methods that are as automated as possible. Specifically, we will:

O3.1 - Research different scenarios of how human rights defenders experience censorship.

• Condense, analyze, and map usability research we have conducted in the Global South prior to the execution of this project; combine this research with feedback gathered from trusted Tor and OONI community members.
• With this research, develop Tor user personas as a tool to understand contexts and mental models.
• Define which censorship scenarios we are going to address.
• Use Tor user personas to define user flows that will empower human rights defenders to effectively circumvent censorship based on the scenarios identified above. O3.2 - Design the flow of how human rights defenders can bypass the scenarios of O3.1.
• User flow part 1: User learns they are under censorship. ◦ Utilize Tor user personas completed in O3.1 to guide the creation wireflows, mockups, and static prototypes on how to visually notify Tor Browser users about detected censorship. ◦ Iterate designs based on peer design reviews with Tor Browser developers. Designs are modified based on development scope, for example: what changes are feasible within the project timeframe, what is realistically possible within the software.
• User flow part 2: Once censorship is acknowledged by the user, we then guide them on how to bypass the detected censorship. ◦ Utilize Tor users personas completed in O3.1 to guide the creation create wireflows, mockups and static prototypes on how to better guide our users to bypass censorship. ◦ Iterate designs based on peer design reviews with Tor Browser developers. Designs are modified based on development scope, for example: what changes are feasible within the project timeframe, what is realistically possible within the software.
• Collect feedback on user flow parts 1 and 2 from an OONI partner. Iterate designs as needed.

O3.3 - Implement final design of user flows created on O3.2.

• Tor Browser developers implement final design of user flows part 1 and 2 into Tor Browser.
• Test and refine solutions developed under O3.2, include them on nightly releases so target users can give us feedback early in the release process.

Archive in https://trac.torproject.org/projects/tor/wiki/org/sponsors/Sponsor30