# OONI Monthly Report: April 2024

Throughout April 2024, the OONI team’s work can be tracked through the various OONI GitHub repositories: https://github.com/ooni

Highlights are shared in this report below.

## Published research report on internet censorship in Tanzania

In April 2024, we published a new research report (“Tanzania: Surge in online LGBTIQ censorship and other targeted blocks”), documenting the blocking of LGBTIQ websites and other targeted blocks in Tanzania based on the analysis of OONI data. Thanks to community member Tori Francis, who translated the report, we also published it in Swahili to reach local communities.

Our research report is available in:
* English: https://ooni.org/post/2024-tanzania-lgbtiq-censorship-and-other-targeted-blocks/
* Swahili: https://ooni.org/sw/post/2024-tanzania-lgbtiq-censorship-and-other-targeted-blocks/

As part of this report, we analyzed OONI data collected from Tanzania between 1st January 2023 to 31st January 2024. Our analysis of OONI data collected from Tanzania shows:

* Blocking of many LGBTIQ websites, including LGBTIQ social networks (such as Grindr), LGBTIQ rights sites (such as OutRight International and ILGA), LGBTIQ news and culture sites (such as Queerty), and a LGBTIQ suicide prevention site (The Trevor Project);
* Blocking of online dating websites (such as Tinder and OKCupid);
* Targeted blocking of specific websites that defend human rights through grants and petitions (Change.org, Global Fund for Women, GlobalGiving, Open Society Foundations);
* Targeted blocking of specific social networking sites (Clubhouse and 4chan);
* Targeted blocking of ProtonVPN.

Our analysis reveals the extensive blocking of LGBTIQ sites, which correlates with the escalating discrimination and crackdown on LGBTIQ communities in Tanzania in recent years. Many other blocks identified as part of this study appear to be targeted, as they involve very specific websites, while other sites from the same categories (e.g. social media, human rights) were found accessible. For example, access to the Global Fund for Women website was found blocked in Tanzania, while Amnesty International's website was found accessible. Meanwhile, Tanzania recently started requiring users to report their use of VPNs. Out of tested VPNs, we only found ProtonVPN blocked in Tanzania during the analysis period.

Overall, the results of our analysis show that most ISPs in Tanzania appear to implement blocks by means of TLS interference, specifically by timing out the session after the Client Hello message during the TLS handshake. As the timing of the blocks and the types of URLs blocked are (mostly) consistent across (tested) networks, ISPs in Tanzania likely implement blocks in a coordinated manner (possibly through the use of Deep Packet Inspection technology).

## Published new reports on the OONI Censorship Findings page

In April 2024, we published 2 new reports on our Censorship Findings page:

* Kyrgyzstan blocked TikTok: https://explorer.ooni.org/findings/154621229001
* Senegal blocked TikTok: https://explorer.ooni.org/findings/144156914701

These reports were published in response to emergent censorship events, sharing relevant OONI data. Access to TikTok was temporarily blocked in Senegal amid political unrest, while access to TikTok was blocked in Kyrgyzstan following a government order to ban the platform.

## Research collaborations with partners on upcoming reports

We continued to coordinate with our partners on research efforts required for upcoming research reports. Specifically, we coordinated with our partners on extensive updates to the Citizen Lab test lists for Bangladesh and Iran.

In April 2024, Miaan Group contributed extensive updates to the test list for Iran: https://github.com/citizenlab/test-lists/pull/1702

## OONI Probe Mobile

In April 2024, we released OONI Probe 3.8.6 on Android (https://github.com/ooni/probe-android/releases/tag/v3.8.6) and iOS (https://github.com/ooni/probe-ios/releases/tag/v3.8.6).

This release includes a fix for the OONI Probe Signal experiment, as well as several other bug fixes and improvements.

## OONI Probe Desktop

In April 2024, we released OONI Probe Desktop 3.9.5: https://github.com/ooni/probe-desktop/releases/tag/v3.9.5

This release contained a fix that prevents the app from being disabled on Windows and uses OONI Probe CLI v3.21.0 (https://github.com/ooni/probe/issues/2699).

## OONI Probe CLI

We rationalized, documented, refactored, and improved the code used to communicate with the OONI backend: https://github.com/ooni/probe/issues/2700

As part of this work, we improved the code used for circumventing the blocking of the OONI backend, documented its design, and planned for future improvements:  https://github.com/ooni/probe/issues/2704

## OONI Run

As part of our work on creating the next generation version of OONI Run (“OONI Run v2”), we continued our efforts with QA testing for the Android version of the new OONI Probe app that will include both an improved UI and support for new OONI Run links. We spent time polishing the new flow that users will go through when adding OONI Run v2 links to their app dashboard, as well as ensuring that updating and removing links works as planned. You can explore related issues here: https://github.com/ooni/run/issues

Additionally, we continued to test and fine-tune the admin dashboard that users will use to create, edit and manage their OONI Run v2 links.

## Documented OONI methodology for measuring throttling

In April 2024, we published documentation for our methodology on measuring throttling: https://github.com/ooni/probe-cli/pull/1546/files

In previous months, we created an open methodology for measuring throttling. In line with our broader scope of work (which focuses on targeted cases of internet censorship), this methodology aims to detect cases of targeted throttling that impact specific online services (such as the throttling of Twitter/X). As part of this methodology, we measure cases of targeted throttling through the analysis of OONI Web Connectivity data (which is collected through the OONI Probe testing of URLs).

Specifically, OONI’s methodology for measuring targeted cases of throttling involves the analysis of timing information during HTTPS requests in Web Connectivity data. This methodology has been successful in measuring various cases of throttling, such as those documented as part of our previous research reports on throttling cases in Kazakhstan (https://ooni.org/post/2023-throttling-kz-elections/#throttling-of-sites), Russia (https://ooni.org/post/2022-russia-blocks-amid-ru-ua-conflict/#twitter-throttled), and Turkey (https://ooni.org/post/2023-turkey-throttling-blocking-twitter/). Our throttling methodology is also explained in these reports.

## OONI Explorer

### User research

Last month, we mentioned that we started user interviews with members of our community who use OONI Explorer (https://explorer.ooni.org/) as part of research and advocacy. The goal of these interviews was to supplement the qualitative data collected through the surveys, and to better understand how they use OONI Explorer and the challenges they encounter in discovering censorship findings through the platform.

In April 2024, we completed these interviews and distilled the information and insights gathered to help us inform the next phase of several projects we have on the go.

To collect community feedback for the improvement of the OONI Explorer navigation, we also did a user study (https://twitter.com/OpenObservatory/status/1778016715940536698). Our goal through this study (https://s900pyof.optimalworkshop.com/optimalsort/33tdoyag) was to collect community feedback on how to improve the structure of content on OONI Explorer and make it easier to navigate.  

All of these efforts are helping us make progress on two important projects:
* Presenting thematic censorship findings on OONI Explorer
* Revamping the OONI Explorer navigation

### Presenting thematic censorship findings on OONI Explorer

Based on community feedback collected as part of our user research studies, we worked with a designer on creating wireframes and mockups of potential ways to present thematic censorship findings on OONI Explorer (https://explorer.ooni.org/). In addition to enabling the internet freedom community to easily and quickly discover censorship findings that they care about, we aim to also ensure that information is presented in a logical fashion, building a stronger connection between the various pages and sections on OONI Explorer.

### Revamping the OONI Explorer navigation

Adding another section like the Thematic Censorship pages to OONI Explorer will require us to consider the navigation menu of the site. As is, the OONI Explorer menu is fairly crowded, and we want to ensure that we can continue to grow and expand the navigation as we grow the functionality within OONI Explorer. Based on community feedback collected as part of a user study on OONI Explorer navigation (https://s900pyof.optimalworkshop.com/optimalsort/33tdoyag), we completed some wireframes of different options for the navigation and had several brainstorming sessions to discuss.

### Other improvements

On the development side, we began the work to update a few sections of the OONI Explorer country pages that had previously been redesigned (https://github.com/ooni/explorer/issues/916, https://github.com/ooni/explorer/issues/914).  We fixed an issue with our language drop-down menu for people using Brave browser (https://github.com/ooni/explorer/issues/931). We also continued our work using Tailwind for our design library (https://github.com/ooni/design-system/pull/175).

## Automating censorship detection and characterization based on OONI measurements

In April 2024, we continued to make progress towards shipping our new OONI Data Pipeline into production.

Specifically, to improve maintainability going forward, we refactored OONI data into two separate packages (https://github.com/ooni/data/pull/60):  
* End user pip installable package to download and parse measurements;
* Pipeline to perform the analysis and processing of OONI data.

We also ported analysis and observation generation over to temporal and set up the production environment in the OONI devops repository (https://github.com/ooni/data/pull/61).

## General backend/devops

In April 2024, we worked on porting the oonifindings service to our new and improved ooniapi pattern (https://github.com/ooni/backend/issues/807). We also experimented with some proof of concepts to improve our devops processes, such as re-conceptualizing codepipeline triggers, and trying weighted target groups and scheduled blue/green deployment.

## Test list updates

Following the announcement of the suspension of the broadcasting of programs from several news media outlets in Burkina Faso (https://x.com/sergedanielinfo/status/1784676862083399728), we created a new test list for Burkina Faso with those news media websites: https://github.com/citizenlab/test-lists/pull/1714

We also added a few news media websites to the Togo test list: https://github.com/citizenlab/test-lists/pull/1715

## Planning the OONI Partner Gathering 2024

In preparation for the upcoming OONI Partner Gathering in Malaysia in May 2024, we continued to coordinate on numerous logistics (flights, shuttle service, hotel, catering, etc.). We also continued to assist participants as needed with their travel logistics, and we continued to coordinate with designers on event-related supplies and materials.

To help ensure that the OONI Partner Gathering 2024 agenda is as useful as possible to our partners, we previously shared a survey to collect their feedback. As our goal was to ensure that we create an agenda that is valuable to all participants, we requested their feedback on the types of sessions that they would find most useful, the types of skills and knowledge that they would like to learn, and the outcomes that would make their participation feel well spent. We also asked participants to propose sessions that they would like to facilitate at the event.

Based on the analysis of partner feedback, we created the final agenda based on the most pressing needs, interests, and requests identified in most survey responses, while taking into account the diversity in participant backgrounds. We also coordinated with partners who expressed interest (through our survey) in facilitating sessions to include their proposed sessions in the final agenda.

In April 2024 (several weeks in advance to the event), we shared the final Agenda with all OONI Partner Gathering participants. The detailed Agenda of the OONI Partner Gathering 2024 is available here: https://ooni.org/documents/OONI-Partner-Gathering-Agenda.pdf

We also created an additional, internal, and more detailed agenda for our team to enable coordination and the management of logistics during the event.

Prior to the event, we prepared and shared the following with participants:
* Code of Conduct: https://ooni.org/documents/Information-Package-CoC-Privacy-Policy.pdf
* Communication Guidelines: https://ooni.org/documents/Information-Package-Communication-Tips.pdf
* Privacy Policy
* OONI Partner Gathering Information Package (providing detailed information about the event, logistical details, and answers to Frequently Asked Questions (FAQ) about the event)
* Facilitation guidelines (shared with all OONI Partner Gathering facilitators)
* Participant list document
* COVID-19 Safety Guidelines
* Emergency Information document

Moreover, in preparation for the OONI Partner Gathering 2024, we:
* Created a dedicated OONI Partner Gathering 2024 mailing list;
* Created a dedicated OONI Partner Gathering 2024 Signal group;
* Created documents for note-taking during the OONI Partner Gathering sessions;
* Created slides for various OONI Partner Gathering sessions;
* Created certificates for each of the participants;
* Ordered OONI t-shirts, banners, stickers, flyers, tote bags, and lanyards for the event;
* Ordered COVID-19 masks, tests, and hand sanitizers for the event.

## Rapid response

### Blocking of TikTok in Kyrgyzstan

In response to the blocking of TikTok in Kyrgyzstan, we shared relevant OONI data and information on Twitter/X: https://twitter.com/OpenObservatory/status/1782005308342055256

We subsequently published a report on our Censorship Findings page, providing further information on the blocking of TikTok in Kyrgyzstan: https://explorer.ooni.org/findings/154621229001

## Community use of OONI data

### Report on the blocking of Grindr in Malaysia

Our partner, Sinar Project, published a report documenting the blocking of Grindr in Malaysia based on OONI data: https://imap.sinarproject.org/news/internet-censorship-update-blocking-of-grindr-com-website

They also encouraged OONI Probe testing of Grindr in Malaysia on Twitter/X: https://twitter.com/sinarproject/status/1781287369213395291

## Community activities

### DRL Implementers Meeting 2024

Between 1st-4th April 2024, OONI’s Arturo and Jessie traveled to Washington D.C to attend the DRL Implementers Meeting 2024.

As part of their participation, they facilitated sessions about community organization and governance, and communicating security risks to users.

### OONI training by Zaina Foundation for researchers in Tanzania

On 4th April 2024, our partner, Zaina Foundation (https://ooni.org/partners/zaina-foundation/), facilitated an OONI training for researchers in Tanzania (https://x.com/ZainaFoundation/status/1776243051339350226).

OONI’s Elizaveta joined the training through remote/online participation, provided a live demo of using OONI Explorer, and addressed the questions of the participants.

### Digital Rights & Inclusion Forum (DRIF) 2024

Between 23rd-25th April 2024, OONI’s Elizaveta traveled to Accra, Ghana, to participate in the Digital Rights & Inclusion Forum (DRIF) 2024.

As part of her participation, Elizaveta facilitated a session on strengthening community response to internet censorship (https://drif.paradigmhq.org/agenda/). While in Accra, Elizaveta also shared OONI’s work as part of a TV interview with Pan African TV: https://www.facebook.com/PANAFRICANTV/videos/979923349677358

## Measurement coverage

In April 2024, 57,990,058 OONI Probe measurements were collected from 2,887 networks in 167 countries around the world.

This information can also be found through our measurement stats on OONI Explorer (see chart on “monthly coverage worldwide”): https://explorer.ooni.org/