tor-project January 2023

tor-project@lists.torproject.org

15 participants
17 discussions

UX Team Meeting cancelled today
by Duncan Larsen-Russell 31 Jan '23

31 Jan '23

Hi folks, The UX Team meeting is cancelled today due to a conflict. But fear not, we shall meet again next week on Tuesday the 7th of Feb. Thanks, Duncan

1 0

Anti-censorship team meeting notes, 2023-01-26
by Shelikhoo 26 Jan '23

26 Jan '23

Hey everyone! Here are our meeting logs: http://meetbot.debian.net/tor-meeting/2023/tor-meeting.2023-01-26-16.01.html And our meeting pad: Anti-censorship work meeting pad -------------------------------- ------------------------------------------------------------------------------------ THIS IS A PUBLIC PAD ------------------------------------------------------------------------------------ Anti-censorship team meeting pad -------------------------------- Next meeting: Thursday, February 2 16:00 UTC Weekly meetings, every Thursday at 16:00 UTC, in #tor-meeting at OFTC (channel is logged while meetings are in progress) == Goal of this meeting == Weekly check-in about the status of anti-censorship work at Tor. Coordinate collaboration between people/teams on anti-censorship at the Tor Project and Tor community. == Links to Useful documents == * Our anti-censorship roadmap: * Roadmap: https://gitlab.torproject.org/groups/tpo/anti-censorship/-/boards * The anti-censorship team's wiki page: * https://gitlab.torproject.org/tpo/anti-censorship/team/-/wikis/home * Past meeting notes can be found at: * https://lists.torproject.org/pipermail/tor-project/ * Tickets that need reviews: from sponsors, we are working on: * All needs review tickets: * https://gitlab.torproject.org/groups/tpo/anti-censorship/-/merge_requests?s… * Sponsor 28 * must-do tickets: https://gitlab.torproject.org/groups/tpo/-/milestones/10 * possible tickets: https://gitlab.torproject.org/groups/tpo/-/issues?scope=all&utf8=%E2%9C%93&… * Sponsor 96 * https://gitlab.torproject.org/groups/tpo/-/milestones/24 * Sponsor 139 <-- hackerncoder, irl, joydeep, meskio, emmapeel working on it * https://pad.riseup.net/p/sponsor139-meeting-pad == Announcements == * == Discussion == * * ln5 asks: does the anti-censorship team want a paid-for host to run STUN/TURN servers on? This is something that could be made part of a grant/fund request. * A server we run ourselves is, in principle, easy to block. The intent is either (not fully sure at this point): * Run a server that is open to the public, so that there are collateral users besides just Snowflake users. * Run a server just for Snowflake, and let it be blocked by the censors that know to block it, simply to reduce load on the other servers we use, in the places where it does not get blocked. (Like the default obfs4 bridges, which are also easy to block, but are not in many places.) * Not a lot of excitement for either idea, since connecting to a distinguished STUN/TURN server is a protocol identification risk. * Suggestion is to run a public STUN/TURN server, but only for Snowflake, and only for proxies (not clients), so that it does not become an identifier for clients. * Proxies-only would work for STUN, but if it were TURN, the connection would still be identifiable by censors (because both the proxy and client would relay through the same TURN IP address). * Will ask ln5 and cohosh for more context. * snowflake fallback from domain fronting to amp cache, how/whether/when/etc to implement (https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow…) * arma is going to check how hard would be to change c-tor to only connect to two bridges instead of all of them at once. So maybe we can duplicate bridges with domain fronting and amp cache * we need to check the consecuences of using amp cache, and if we are ok having half of our users using it * use ampcache for snowflake in IR? * https://gitlab.torproject.org/tpo/anti-censorship/team/-/issues/115 * https://gitlab.torproject.org/tpo/anti-censorship/rdsys-admin/-/merge_reque… * our domain front for both snowflake and circumvention settings is blocked in Iran * that will mean configuring amp cache in Circumvention Settings, the people that will get this configuration will still be able to use domain fronting to access snowflake, but they will keep == Actions == * We should make a ticket for pion to cache its stun answers when possible, because right now it surprises us by asking way more stun questions than it actually needs to. * We might want to be able to spin up our own stun servers, on our own ip/port, for debugging. We should talk to TPA about that goal at some point (not urgent). * Roger will look more at https://gitlab.torproject.org/tpo/core/tor/-/issues/40578 ("only contact the first few working bridges on your bridge list" and plan to have a sense of whether it will be an easy hack or a hard one, for next week. == Interesting links == == Reading group == * We will discuss "" on * * Questions to ask and goals to have: * What aspects of the paper are questionable? * Are there immediate actions we can take based on this work? * Are there long-term actions we can take based on this work? * Is there future work that we want to call out in hopes that others will pick it up? == Updates == Name: This week: - What you worked on this week. Next week: - What you are planning to work on next week. Help with: - Something you need help with. cecylia (cohosh): last updated 2023-01-26 Last week: - FOCI workshop prep - some progress on rust library for rdsys backend - https://gitlab.torproject.org/cohosh/rdsys-backend-api This week: - wrap up conjure documentation and write an announcement asking for testers - finish rust library for rdsys backend - continue working on lox client integration in Tor Browser - take a look at the dead on arrival rotating bridge problem - https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/… Needs help with: dcf: 2023-01-19 Last week: - made merge request to bring sample snowflake client torrc up to date https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - more review on abbreviating ice specifications https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - promoted the patch to remove Hello Verify Request and asked for testing https://ntc.party/t/second-snowflake-bridge-available-for-testing/3445/11 https://ntc.party/t/in-case-snowflake-rendezvous-gets-blocked/1857/25 - hacking on snowflake-graphs to do per-country graphs Next week: - migrate goptlib to gitlab https://gitlab.torproject.org/tpo/anti-censorship/team/-/issues/86#note_282… - try Conjure PT development version https://forum.torproject.net/t/tor-dev-introducing-a-conjure-pt-for-tor/4429 Help with: meskio: 2023-01-26 Last week: - give support to bridgesanner (rdsys#143) - review gettor whatsapp implementation (rdsys!66) - investigate the situation of meek in uzbekistan (censorship-analysis#40031) - write S96 report - prepare a Circumvention Settings configuration with ampcache for IR (rdsys-admin!13) - explore snowflake proxy docker options to avoid using the host network (docker-snowflake-proxy#11) Next week: - implement bridgescanner needs (rdsys#143) Shelikhoo: 2023-01-26 Last Week: - [Merge Request Awaiting] Add SOCKS5 forward proxy support to snowflake (snowflake!64) - [Discussion & Deployment] Rollout of Distributed Snowflake Support - [Coding & Deployment] Proposal: Centralized Probe Result Collector (anti-censorship/team#54) - [Research] HTTPT Planning https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/http… - reply a lot of tickets - [Merge Request] Automated Container Image Building in Continuous integration (https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/…) Next Week: - [Research] WebTunnel planning (Continue) - push the chunked upload raw data upload change to vantage points - research snowflake's performance issue in China Itchy Onion: 2023-1-26 Last week: - Investigate whether stun over TLS is beneficial to us (issues#40240) - Looking at options for standalone proxy runners to specify an interface (issue#40108) This week: - Lunar New Year break - Continue working on issue #40108 (standalone proxy bind specific IP) - Review MRs hackerncoder: 2023-01-12 last week: - figure out what makes ooni-exporter put all reports from a country in either success or failure (I still don't know why. But I got it to work) Next week: - getting ooni-exporter to work with torsf (snowflake) - work on monitoring bridges health cece: 2022-12-22 This week: - working on creating a dummy WhatsApp bot Next week: - My bot is not yet working as expected s still trying to figure that out Help with: - resources

1 0

PSA: noisy GitLab bot on the loose
by Antoine Beaupré 26 Jan '23

26 Jan '23

Hi everyone, You might have noticed getting a little more email from GitLab than usual this week. It's because I have deployed the fancy 🤖Triage Bot🤖 that was already in operation inside the "team" TPA project to all TPA and web projects over the week. This has first generated a flurry of email as the bot went through all the stale issues we had lying around (and we had a lot!). It also was pinging the author of the ticket instead of the assignee. That was partly fixed: it doesn't explicitly ping anyone now, but still makes a comment. A discussion is ongoing on how to improve that workflow here: https://gitlab.torproject.org/tpo/tpa/triage-ops/-/issues/11 More stale issues are bound to come up in the future: the trick here is that, when you have an issue assigned to you and it's marked Next, Doing, or Needs Information, you MUST update it once in a while, at *least* once every other week. Technically, the bot should be even more aggressive: the ~Doing label is supposed to be "work done this week": https://gitlab.torproject.org/tpo/tpa/team/-/wikis/policy/tpa-rfc-5-gitlab#… (On the other hand, the ~Next label is "work done this cycle, typically a month", so it's a bit too aggressive on that.. I personnally do not mind the reminder after two weeks, personnally, but I'm happy to tweak this if people find that too noisy.) I should probably have made a TPA-RFC for this since it seems to affect more people than I expected, but for now this announcement will have to do. Comments welcome in the above issue or please make a new issue in the triage-ops repository if you have more concerns about the bot! https://gitlab.torproject.org/tpo/tpa/triage-ops/-/issues Have a nice day and thanks for flying TPA. :) -- Antoine Beaupré torproject.org system administration

1 1

Notification of my blog posts and internship progress
by Cecilia Orji 25 Jan '23

25 Jan '23

Hi, My name is Orji Cecilia, an Outreachy intern working to Extend GetTor to distribute TorBrowser on censored networks over IM and social media <https://www.outreachy.org/outreachy-december-2022-internship-round/communit…>. My mentors are meskio and Gaba. I am writing to notify you of my blog posts which talk about my internship and how it is going. Below are links to my blog post which I will like to be added to the archives if possible; - https://dev.to/vivcis/introducing-cecilia-35jc - https://dev.to/vivcis/everybody-struggles-1085 - https://dev.to/vivcis/think-about-your-audience-extend-gettor-browser-2904 Regards, Cecilia

2 2

Proxy Leak Detector/Preventor (Outreachy)
by Robert Mindo 17 Jan '23

17 Jan '23

Greetings! I wanted to share an update on the progress of the project. So far, the ability to detect the IP address and port that socket system calls are sending data to has been successfully implemented. Additionally, the capability to block system calls that are sending data to an IP address and port that is not the desired proxy. An option to kill the application if a proxy leak occurs has been added, which is useful for manual QA testing of applications. Furthermore, an option to allow proxy leaks but log any that occur has also been added, which is useful for automated testing of applications. Additionally, the capability to use the environment variables that Tor Browser uses, such as TOR_SOCKS_PORT, to decide what the desired proxy is, has been included. SOCKSification is currently being implemented. The approach is to intercept the connect syscall's entry point and modify the destination IP/Port by manipulating the syscall. The exit of the connect syscall is then intercepted in order to capture the returning file descriptor representing the established socket. The file descriptor is duplicated using pidfd getfd, then a SOCKS5 handshake is done. Overall, the project is progressing well and any suggestions are much welcomed. Github Repository <https://github.com/namecoin/heteronculous-horklump> Project Summary <https://www.namecoin.org/2022/11/29/introducing-intern-robert-nganga.html> Thank You, Robert.

1 0

Next Tor Browser Meeting Moved to 2023-01-17
by Richard Pospesel 13 Jan '23

13 Jan '23

Hi Everyone, The 16th is a US Federal Holiday and therefore an official Tor Project holiday as well. We will therefore move the weekly applications dev meeting to Tuesday the 17th at 1500 UTC in #tor-meeting on OFTC IRC. best, -Richard

1 0

Anti-censorship team meeting notes, 2023-01-12
by Shelikhoo 12 Jan '23

12 Jan '23

Hey everyone! Here are our meeting logs: http://meetbot.debian.net/tor-meeting/2023/tor-meeting.2023-01-12-15.58.html And our meeting pad: Anti-censorship work meeting pad -------------------------------- ------------------------------------------------------------------------------------ THIS IS A PUBLIC PAD ------------------------------------------------------------------------------------ Anti-censorship team meeting pad -------------------------------- Next meeting: Thursday, January 19 16:00 UTC Weekly meetings, every Thursday at 16:00 UTC, in #tor-meeting at OFTC (channel is logged while meetings are in progress) == Goal of this meeting == Weekly check-in about the status of anti-censorship work at Tor. Coordinate collaboration between people/teams on anti-censorship at the Tor Project and Tor community. == Links to Useful documents == * Our anti-censorship roadmap: * Roadmap: https://gitlab.torproject.org/groups/tpo/anti-censorship/-/boards * The anti-censorship team's wiki page: * https://gitlab.torproject.org/tpo/anti-censorship/team/-/wikis/home * Past meeting notes can be found at: * https://lists.torproject.org/pipermail/tor-project/ * Tickets that need reviews: from sponsors, we are working on: * All needs review tickets: * https://gitlab.torproject.org/groups/tpo/anti-censorship/-/merge_requests?s… * Sponsor 28 * must-do tickets: https://gitlab.torproject.org/groups/tpo/-/milestones/10 * possible tickets: https://gitlab.torproject.org/groups/tpo/-/issues?scope=all&utf8=%E2%9C%93&… * Sponsor 96 * https://gitlab.torproject.org/groups/tpo/-/milestones/24 * Sponsor 139 <-- hackerncoder, irl, joydeep, meskio, emmapeel working on it * https://pad.riseup.net/p/sponsor139-meeting-pad == Announcements == * Open Collective funding for Snowflake bridge operations is open * https://opencollective.com/censorship-circumvention/projects/snowflake-dail… * First update post: https://opencollective.com/censorship-circumvention/projects/snowflake-dail… == Discussion == * Enable snowflake-02 in Orbot * snowflake-02 (enabled in Tor Browser only) currently gets only about 5% the traffic of snowflake-01 * at this moment: * snowflake-01: 445 MiB/s * snowflake-02: 21 MiB/s * snowflake-01 reaching its CPU limit * discuss with topic with guardian project at next S96 meeting * deofuscation obfs4 issues are public * https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/obfs… * https://gitlab.torproject.org/tpo/anti-censorship/team/-/issues/91 * is the stun.stunprotocol.org situation resolved to everyone's satisfaction, or is more attention required? * https://lists.torproject.org/pipermail/anti-censorship-team/2022-December/0… * removed from default lists https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… * it is resolved * (From last week) arti-based obfs4 quick reachability monitor https://gitlab.torproject.org/tpo/core/arti/-/issues/717#note_2866528 * feel free to comment on it * snowflake blocking in Russia (maybe TSPU only) by Hello Verify Request (since about 2022-07-20) * https://ntc.party/t/second-snowflake-bridge-available-for-testing/3445/7 * https://ntc.party/t/in-case-snowflake-rendezvous-gets-blocked/1857/9 * https://gitlab.torproject.org/tpo/anti-censorship/censorship-analysis/-/iss… * https://explorer.ooni.org/chart/circumvention?since=2022-07-08&until=2022-0… * Side note: see how correlated the number of daily reports is between CA and RU. Originally dcf included CA on this graph to highlight a potential problem in the OONI torsf test, since Snowflake should not be blocked in Canada. But a more likely explanation may be geolocation errors; i.e., that IPs in Russia are being mistakenly attributed to Canada, and therefore Canda shows an unexpectedly high rate of anomalies. We suspect something similar happened with IR->US in https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow…. * shelikhoo will look into it == Actions == * == Interesting links == == Reading group == * We will discuss "" on * * Questions to ask and goals to have: * What aspects of the paper are questionable? * Are there immediate actions we can take based on this work? * Are there long-term actions we can take based on this work? * Is there future work that we want to call out in hopes that others will pick it up? == Updates == Name: This week: - What you worked on this week. Next week: - What you are planning to work on next week. Help with: - Something you need help with. cecylia (cohosh): last updated 2022-01-12 (away this week) Last week: - updated default STUN servers in the proxy and client - https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - submitted MR for conjure on desktop TB - https://gitlab.torproject.org/tpo/applications/tor-browser-build/-/merge_re… - worked on debugging domian fronting issues with conjure - https://gitlab.torproject.org/tpo/tpa/team/-/issues/41023 - wrote up an analysis of BridgeDB anti-enumeration methods - https://gitlab.torproject.org/tpo/anti-censorship/team/-/issues/84 This week: - continue work on conjure client - more work on lox integration Needs help with: dcf: 2023-01-12 Last week: - snowflake CDN bookkeeping https://gitlab.torproject.org/tpo/anti-censorship/team/-/wikis/Snowflake-co… - experimented with increasing conntrack limits on snowflake-01 and documented how to do it in the installation guide https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - reviewed merge request to abbreviate stun server specifications https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - made a public repo with code to generate snowflake graphs https://gitlab.torproject.org/dcf/snowflake-graphs, used for the graph at https://opencollective.com/censorship-circumvention/projects/snowflake-dail… - restarted snowflake-01 with 4 KCP state machines https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/snow… - made a public repo for obfs4 remote testing scripts https://gitlab.torproject.org/dcf/obfs4-checks and made a graph of obfs4 bridge upgrade progress https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/obfs… Next week: - migrate goptlib to gitlab https://gitlab.torproject.org/tpo/anti-censorship/team/-/issues/86#note_282… - try Conjure PT development version https://forum.torproject.net/t/tor-dev-introducing-a-conjure-pt-for-tor/4429 Help with: meskio: 2023-01-12 Last week: - publish obfs4 deobfuscation bugs - start working on localization support for rdsys (rdsys#11) - review dummy whatsapp bot (rdsys!66) - add microsoft edge installation to snowflake website (snowflake-webext!62) Next week/after holidays: - add localization support to rdsys (rdsys#11) - support bridgescanner (rdsys#143) Shelikhoo: 2023-01-12 Last Week: - [Merge Request Awaiting] Add SOCKS5 forward proxy support to snowflake (snowflake!64) - [Discussion & Deployment] Rollout of Distributed Snowflake Support - [Coding & Deployment] Proposal: Centralized Probe Result Collector (anti-censorship/team#54) - [Research] HTTPT Planning https://gitlab.torproject.org/tpo/anti-censorship/pluggable-transports/http… - [Merge Request Done] Update to support chunked upload of probe log(https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurem… - [Merge Request Done] Add Resumable Upload to Log Collector (https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/…) - [Merge Request Done] cumulative updates on providing more options: loopback traffic collection, raw log collection, remove obfs4 bridge line input for snowflake (https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/…) Next Week: - [Research] WebTunnel planning (Continue) - Fixing the vantage point in Russia - push the chunked upload raw data upload change to vantage points Itchy Onion: 2022-1-12 Last week: - worked on RACE performance issue This week: - Made progress on the RACE performance improvement -- the plugin now is able to reuse snowflake connections to send multiple messages between the same hosts. I'm not entirely convinced the improvements are substantial yet. Will run some tests and see if it can be further improved hackerncoder: 2022-01-12 last week: - figure out what makes ooni-exporter put all reports from a country in either success or failure (I still don't know why. But I got it to work) Next week: - getting ooni-exporter to work with torsf (snowflake) - work on monitoring bridges health cece: 2022-12-22 This week: - working on creating a dummy WhatsApp bot Next week: - My bot is not yet working as expected s still trying to figure that out Help with: - resources

1 0

OONI Monthly Report: December 2022
by Maria Xynou 11 Jan '23

11 Jan '23

Hello, Below I share OONI's status report for December 2022. *# OONI Monthly Report: December 2022* Throughout December 2022, the OONI team worked on the following sprints: * Sprint 79 (1st-4th December 2022) * Sprint 80 (5th-18th December 2022) * Sprint 81 (19th-31st December 2022) Our work can be tracked through the various OONI GitHub repositories: https://github.com/ooni Highlights are shared in this report below. *## Year in Review: OONI in 2022* We published our annual blog post, "Year in Review: OONI in 2022", which shares: * OONI highlights from 2022 * Highlights from the OONI community (including many of our partners) * Our plans for 2023 You can read this post here: https://ooni.org/post/ooni-in-2022/ *## OONI’s 10th Anniversary* On 5th December 2022, OONI celebrated its 10th anniversary! To mark this milestone, we organized and hosted 2 live-streamed events, published a timeline animation, community video, and blog post. Further details are shared below. *### Live-streamed events* In celebration of OONI’s 10th anniversary, we organized and hosted 2 live-streamed events (https://ooni.org/post/10th-ooniversary-events/) on our YouTube channel: * 10th Ooniversary: OONI Highlights (5th December 2022) * 10th Ooniversary: OONI Community (6th December 2022) As part of the “OONI Highlights” live-streamed event, Arturo (OONI’s founder) shared OONI’s history and highlights from the past 10 years, as well as our plans for the future. This presentation can be viewed here: https://www.youtube.com/watch?v=f4D4gq9TBMg As part of the “OONI Community” live-streamed event, we invited 5 community members who shared how they’ve used OONI tools and data as part of their research and advocacy efforts ( https://ooni.org/post/10th-ooniversary-events/#10th-ooniversary-ooni-commun…). Their presentations can be viewed here: https://www.youtube.com/watch?v=Co_f8KPBIZM Both events included presentations followed by discussion addressing questions shared by participants through the live chat and shared pads. On 5th December 2022, we also hosted an online Distance Disco ( https://distancedisco.nl/dd/ooni-s-10th-birthday-party) to dance with community members in celebration of OONI’s 10th anniversary. *### New animation of OONI highlights* On 5th December 2022, we published an animation that provides a timeline of OONI highlights from the past 10 years. This animation was produced in collaboration with Robotina (https://www.robotina.it/). The animation is available here: https://www.youtube.com/watch?v=AjpfF2kQ9jA *### New OONI Community Video* In celebration of OONI’s 10th Anniversary, we published a new video with OONI community members. You can watch this video here: https://www.youtube.com/watch?v=zrclQ2QZjVo The video features 11 community members discussing how OONI has been useful to their work, and what they would like to see OONI do in the future. You can learn about the featured community members through our blog post: https://ooni.org/post/ooni-community-video/ *### Highlights: 10 Years of OONI* To mark OONI’s 10-year anniversary, we published a blog post where we document the key OONI highlights from the past 10 years, as well as some of our future plans. This blog post is available here: https://ooni.org/post/highlights-10-years-of-ooni/ *## OONI Probe Mobile* We released OONI Probe Android 3.7.3 ( https://github.com/ooni/probe-android/releases/tag/v3.7.3) and OONI Probe iOS 3.73 (https://apps.apple.com/us/app/id1199566366) with bug fixes and improvements. We worked towards ensuring that OONI Probe Mobile uses the check-in API instead of the test list API ( https://github.com/ooni/probe/issues/2390), we fixed broken Android tests ( https://github.com/ooni/probe/issues/2386), and we worked towards ensuring that the testing of OONI Run links does not stop if a device is locked ( https://github.com/ooni/probe/issues/2361). *## Published a new OONI Run user guide* To enable community members to use the OONI Run (https://run.ooni.io/) platform for custom website testing (and for coordinating censorship measurement campaigns around the world), we created and published a new user guide for OONI Run. Our OONI Run user guide can be accessed here: https://ooni.org/support/ooni-run/ Through the OONI Run user guide, you can learn how to: * Create OONI Run links for testing websites (of your choice) for censorship; * Use OONI Run links with the OONI Probe mobile app; * Share OONI Run links with other OONI Probe mobile app users; * Generate OONI Run widget code. Upon reading this guide, we hope you will feel empowered to share your OONI Run knowledge and skills with others! We also hope this guide helps with coordinating OONI Probe website testing around the world. *## OONI Probe CLI* We released OONI Probe CLI 3.16.6 ( https://github.com/ooni/probe-cli/releases/tag/v3.16.6) and OONI Probe 3.16.7 (https://github.com/ooni/probe-cli/releases/tag/v3.16.7). These two releases include improvements to the experimental Web Connectivity v0.5 nettest, as well as the following important fixes: * In OONI Probe CLI 3.16.6, we included the reliability fix (developed in November) that makes OONI Probe try all the available test helpers, rather than just the first one in the list; * In OONI Probe CLI 3.16.7, we backported fixes for the WhatsApp and Telegram experiments that stop measuring unencrypted web endpoints to avoid possible false positives (an issue discussed in detail here: https://github.com/ooni/ooni.org/issues/1317). We continued the process of merging ooniprobe (the official CLI client) and miniooni (the experimental CLI client) by introducing a common database abstraction that could be used by both in the following pull request: https://github.com/ooni/probe-cli/pull/988 *## OONI Probe Desktop* We released OONI Probe Desktop 3.8.3 to include fixes from OONI Probe CLI 3.16.7 https://github.com/ooni/probe-desktop/releases/tag/v3.8.3 *## Expanding OONI’s testing model to support richer testing input* To improve the quality of OONI measurements (and reduce the risk of false positives), we stopped measuring unencrypted web endpoints for WhatsApp and Telegram (https://github.com/ooni/ooni.org/issues/1317). We continued our refactoring work to support providing richer testing input ( https://github.com/ooni/ooni.org/issues/1291, https://github.com/ooni/ooni.org/issues/1292, https://github.com/ooni/probe/issues/2362). During December 2022, we focused on cleaning up “probeservices” (the Go package that communicates with the OONI API) to set the stage for implementing richer input. We removed support for deprecated APIs ( https://github.com/ooni/probe/issues/2380), we started to sync up the data format with the backend (https://github.com/ooni/probe-cli/pull/996), and we added support in the backend and in ooniprobe for receiving compressed responses (https://github.com/ooni/probe/issues/2379). The latter change, in particular, is required because adding more testing options to each URL significantly increases the size of the response message. We then annotated code to communicate with the backend with the request and response message types (https://github.com/ooni/probe-cli/pull/1010, https://github.com/ooni/probe-cli/pull/1011). This change allows us to generate a swagger describing ooniprobe’s view of the API, which we can compare to the one generated by the server, to be sure the two implementations are (and stay) in sync. Regarding richer input, we started sketching out a design document ( https://github.com/ooni/probe/issues/2381) and a prototype ( https://github.com/ooni/probe-cli/pull/1005) implementing this design document. The general idea is to give experiments freedom to call the specific backend API they require for getting their inputs, which removes the need to maintain middleware code that delivers inputs to the experiments. With more input diversity, such a middleware code becomes increasingly the bottleneck to innovation, so we want to experiment with removing it and adding additional functions that experiments can call to implement common functionality instead. We also added support to the Web Connectivity Test Helper (TH) for measuring HTTP/3 hosts (https://github.com/ooni/probe-cli/pull/985). Support for HTTP/3 in the TH is crucial to start measuring websites using HTTP/3 in the Web Connectivity experiment. Reflecting on our own experience writing experiments in the step-by-step style ( https://github.com/ooni/probe-cli/blob/master/docs/design/dd-003-step-by-st…), as well as by learning from users’ contributions ( https://github.com/ooni/probe-cli/pull/989#discussion_r1032544878), we determined that the step-by-step API provides the basic building blocks for writing all the experiments we need, but there is also a need to have more user-friendly and intuitive wrappers on top of it. To this end, we started experimenting with a simple domain specific language for expressing and composing low-level operations that should simplify the writing of experiments (https://github.com/bassosimone/oonidsl/). *## OONI Explorer* We started working towards integrating relevant third-party data into OONI Explorer. Specifically, we worked on third-party data visualizations on country and network pages using the internet outage data from IODA and internet traffic data from Cloudflare. We also continued conducting UX research to improve the OONI Explorer measurement pages ( https://6q8h6s64.optimalworkshop.com/optimalsort/avnrhnv5/sort). We added support for choosing the time granularity on the Measurement Aggregation Toolkit (MAT), beyond day granularity (which has been the only option so far). Users can now also display the data per hour, week and month (https://github.com/ooni/explorer/pull/818). *## Published an OONI Explorer demo* To enable community members to use OONI Explorer (https://explorer.ooni.org/) for investigating internet censorship based on OONI data, we published an OONI Explorer demo on our YouTube channel. The OONI Explorer demo is available here: https://www.youtube.com/watch?v=6Rce-xshLac *## OONI backend* In response to a change in the testing of WhatsApp that made our tests fail, we implemented a temporary workaround that restores the functionality of the test (https://github.com/ooni/pipeline/pull/403). We added support for HTTPS to the deb.ooni.org Debian package repository after it was requested by the community to circumvent blocking of the HTTP version. We also moved the service to the new backend host. We enabled gzip support in Nginx to compress responses from the check-in API and other JSON content and created a simple dashboard to monitor the improvement. This makes the responses from the Measurement Aggregation Toolkit (MAT) quicker for the browser to download. *## Automating censorship detection and characterization based on OONI measurements* As part of our ongoing effort to automate and improve our characterization of censorship events through the analysis of OONI measurements, we made significant progress on the ooni/data tool (https://github.com/ooni/data). Following internal technical discussions, as well as conversations with members of the OONI community, we implemented a new iteration of the data model used for representing the results of experiments ( https://github.com/ooni/data/pull/19). Moreover, we started the process of evaluating the new analysis engine against the analysis that currently exists and is mostly performed as part of the measurement workflow (https://github.com/ooni/data/issues/21). Some very preliminary findings suggest that it’s quite an important improvement both in terms of accuracy and recall. We also experimented with producing plots that breakdown the means through which blocking is implemented (https://github.com/ooni/data/issues/22). Smaller incremental improvements include: Handling the deduplication of already processed data to support the easier reprocessing of data ( https://github.com/ooni/data/pull/19/commits/eea26f7138b14fc4595d55b432d287… ); Expanding cloud provider detection to reduce DNS false positives ( https://github.com/ooni/data/pull/19/commits/01dc691d695dfc103d604846ad7568… ); Refactoring of the DNS anomaly analysis subsystem ( https://github.com/ooni/data/pull/20). We also started implementing the extraction of the blocking type in the fastpath pipeline. *## Creating a Social Media Censorship Alert System* We continued to work on the development of the Social Media Blocking Event detector. We started running a prototypal version, created a dashboard and a jupyter notebook. We ran the prototype for weeks to review the initial results and improve the algorithm over time. *## Published Test Lists Editor screencast* To enable community members to contribute websites for censorship testing, we created and published a new screencast for our Test Lists Editor ( https://test-lists.ooni.org/). The Test Lists Editor screencast can be viewed here: https://www.youtube.com/watch?v=6i2OVHUQEpE Through our Test Lists Editor, you can review and contribute websites for censorship testing by OONI Probe users around the world. *## TikTok blocked in Jordan* On 16th December 2022, Jordan blocked access to TikTok amid fuel protests. We rapidly responded by sharing relevant OONI data, charts, and information about the block on Twitter: https://twitter.com/OpenObservatory/status/1603845292343889924 *## Test list updates* Following the blocking of TikTok in Jordan, we added TikTok endpoints to the test list for Jordan (https://github.com/citizenlab/test-lists/pull/1201). Thanks to our URL prioritization system, these newly added URLs were immediately prioritized for testing in Jordan as soon as the pull request was merged (enabling the immediate collection of relevant measurements and supporting rapid response efforts). Throughout December 2022, we also reviewed and merged many other updates to test lists contributed by community members: https://github.com/citizenlab/test-lists/pulls?q=is%3Apr+is%3Aclosed *## Creating a new OONI Outreach Kit* Based on review and feedback, we continued to make improvements to the materials and resources of the upcoming OONI Outreach Kit. Specifically, we made extensive edits and improvements to the set of OONI workshop slides included in the OONI Outreach Kit. *## Updated OONI Partner pages* On our website, we have an “OONI Partners” section which features our partners, their important work, and the specific projects and research reports that we’ve collaborated on (https://ooni.org/partners). We updated many OONI Partner pages to include our latest collaborations: https://github.com/ooni/ooni.org/pull/1328 *## Cloudflare blog post* Cloudflare published a blog post describing how civil society organizations use Cloudflare Radar (https://radar.cloudflare.com/) to track internet shutdowns around the world: https://blog.cloudflare.com/partnering-with-civil-society-to-track-shutdown… This blog post highlights OONI, where we share how Cloudflare Radar has been useful to our work. *## Community use of OONI data### iMAP 2022: Research reports on internet censorship in 8 Asian countries* On 21st December 2022, our long-term partner, Sinar Project, published 8 new research reports on internet censorship in Southeast Asia and Hong Kong (China) in collaboration with their Internet Monitoring and Action Project (iMAP) partners (https://imap.sinarproject.org/). Based on the analysis of OONI data, these research reports investigate internet censorship in Malaysia, Thailand, Myanmar, Indonesia, Cambodia, Vietnam, Hong Kong, and the Philippines. Their research reports can be found here: https://imap.sinarproject.org/reports/2022 Our iMAP partners launched and presented these 8 research reports through the following live-streamed presentation: https://www.youtube.com/watch?v=u9JsCqXebPI To highlight and help disseminate their important work, we published a blog post which briefly discusses and links to their 8 research reports: https://ooni.org/post/2022-imap-8-research-reports-southeast-asia/ *## Community activities### Internet Governance Forum (IGF) 2022* On 1st December 2022, OONI’s Maria presented OONI on the panel “Help! The Kill switch is taking away my limited agency” at the Internet Governance Forum (IGF) 2022 ( https://intgovforum.org/en/content/igf-2022-town-hall-54-help-the-kill-swit… ). *### OONI workshop for journalists in Kyrgyzstan* On 20th December 2022, OONI’s Elizaveta facilitated an online OONI workshop for journalists in Kyrgyzstan as part of a training organized by Internews. During this workshop, Elizaveta explained how journalists can use OONI Probe and OONI data to investigate internet censorship. *## Userbase* In December 2022, 57,009,977 OONI Probe measurements were collected from 2,815 AS networks in 161 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/ ~ OONI team.

1 0

OONI Monthly Report: November 2022
by Maria Xynou 11 Jan '23

11 Jan '23

Hello, Below I share OONI's status report for November 2022. *# OONI Monthly Report: November 2022* Throughout November 2022, the OONI team worked on the following sprints: * Sprint 77 (1st-6th November 2022) * Sprint 78 (7th-20th November 2022) * Sprint 79 (21st-30th November 2022) Our work can be tracked through the various OONI GitHub repositories: https://github.com/ooni Highlights are shared in this report below. *## Technical multi-stakeholder research report on Internet shutdowns in Iran* In collaboration with IODA, M-Lab, Cloudflare, Kentik, Censored Planet, ISOC, and Article19, OONI published a new research report: "Technical multi-stakeholder report on Internet shutdowns: The case of Iran amid autumn 2022 protests". Read the report here: https://ooni.org/post/2022-iran-technical-multistakeholder-report/ This report combines multiple measurement datasets in an attempt to investigate the internet shutdown events in Iran over the past months amid protests. We consider this a particularly important research report, as it’s likely the first ever report that combines so many internet measurement datasets in investigating a case study. Specifically, we provide data and analysis on the following in Iran: * Mobile network outages * Regional outages * HTTP/3 and QUIC traffic drop * IPv6 disruption on Irancell * Increased blocking of encrypted DNS * Impact on internet speed tests * Blocking of WhatsApp, Instagram, Skype, Linkedin, and Viber * Blocking of Google Play Store and Apple App Store * Blocking of browser extension repositories ISOC also shared their analysis on examining the economic impact of the shutdowns. We provide a summary of the key findings here: https://ooni.org/post/2022-iran-technical-multistakeholder-report/#summary-… ISOC also published a blog post about the report: https://pulse.internetsociety.org/blog/technical-multistakeholder-report-on… *## OONI Team Meeting* Between 2nd-4th November 2022, we organized and hosted a 3-day OONI Team Meeting in Rome, Italy. This was the first in-person OONI Team Meeting since the start of the COVID-19 pandemic. Unfortunately not everyone was able to join us in person in Rome, and so we facilitated a hybrid meeting, with some participants in person, and others joining us remotely online. In addition to OONI’s 9 team members, the meeting also included the 2 OTF Information Controls Research Fellows (Ain Ghazal and Gurshabad Grover) that we’re currently hosting. During this 3-day meeting, we improved our roadmaps and discussed strategic priorities and implementation details for many new systems we’ll be building over the next few years. Specifically, the 3-day OONI Team Meeting included the following sessions: * Time for reflection * Community needs and priorities * Expanding OONI’s testing model to support richer testing input * Incrementally integrating Websteps into Web Connectivity LTE * Measuring the blocking of VPNs * Research on Internet censorship in South Asia and analysis of “failed” measurements * Creating a methodology for measuring throttling: Next steps * Automating the detection and characterization of censorship * Creating a Social Media Censorship Alert System * Presenting thematic censorship findings on OONI Explorer * Creating a Censorship Incident Reporting Platform * Aligning the anomaly calculations between the pipeline, Explorer, and probe * Handling of HTTP response bodies in measurements * Triage and prioritization of issues * Roadmapping sessions * Fundraising priorities During the OONI Team Meeting, we also had an OONI-themed cake to celebrate OONI’s 10th Anniversary together ( https://ooni.org/post/ooni-in-2022/images/01.png). After the 3-day OONI Team Meeting, we had a “team bonding day” together in Rome. *## Organizing OONI’s 10th Anniversary* In preparation for OONI’s 10th Anniversary (on 5th December 2022), we worked towards organizing live-streamed events and creating multimedia content to mark the important milestone. In November 2022, we published a blog post which invites the internet freedom community to attend 2 live-streamed events for OONI’s 10th Anniversary: https://ooni.org/post/10th-ooniversary-events/ Throughout November 2022, we interviewed 11 OONI community members and worked with a videographer on creating a video that highlights how OONI has been useful to the internet freedom community. We also wrote a script for an animation that provides a timeline of key OONI highlights from the past 10 years, and we worked with animators on producing an animation. Both the community video and animation were scheduled for publication around OONI’s 10th Anniversary in early December 2022. *## OONI Probe Mobile* In November 2022, we released OONI Probe Android 3.7.2 ( https://github.com/ooni/probe-android/releases/tag/v3.7.2) with bug fixes and improvements. We worked towards ensuring that flutter implementations are updated properly on all platforms and we started making progress on unsupported platforms (https://github.com/ooni/probe-cli/pull/880). We also worked on triaging issues reported in Sentry. *## OONI Run* We edited the copy on the OONI Run platform: https://github.com/ooni/run/pull/118 *## OONI Probe CLI* We released OONI Probe CLI 3.16.4 ( https://github.com/ooni/probe-cli/releases/tag/v3.16.4) and v3.16.5 ( https://github.com/ooni/probe-cli/releases/tag/v3.16.5) to fix issues with the Signal nettest caused by Signal changing the certification authority it uses to authenticate its endpoints. Additionally, we started merging ooniprobe (the official command line client) and miniooni (the research command line client) by abstracting common functionality into shared Go packages to be used by both tools. This work is documented through the following pull requests: https://github.com/ooni/probe-cli/pull/979 and https://github.com/ooni/probe-cli/pull/982. Notably, we added support to ooniprobe for a `--proxy` command line argument, thus facilitating the use of circumvention tools (such as tor): https://github.com/ooni/probe-cli/pull/981 We also made the code that communicates with the Web Connectivity Test Helper more reliable by trying all the available Test Helpers, rather than just using the first one returned by the backend API. This work is available through the following pull request: https://github.com/ooni/probe-cli/pull/980 *## OONI Probe Desktop* We released OONI Probe Desktop 3.8.2 which includes Signal nettest fixes from OONI Probe CLI: https://github.com/ooni/probe-desktop/releases/tag/v3.8.2 *## Published an OONI Probe Desktop screencast* To enable community members worldwide to use OONI Probe, we created an OONI Probe Desktop screencast which we published on our YouTube channel. Our OONI Probe Desktop screencast is available here: https://www.youtube.com/watch?v=tLDVpyHFsW0 *## Russian translation of the OONI Probe Mobile and Desktop User Guides* We translated the OONI Probe user guides to Russian, and published the translated versions: * OONI Probe Mobile: https://ooni.org/ru/support/ooni-probe-mobile * OONI Probe Desktop: https://ooni.org/ru/support/ooni-probe-desktop *## Expanding OONI’s testing model to support richer testing input* We started refactoring work to support providing richer input to OONI Probe tests (https://github.com/ooni/ooni.org/issues/1291). This work stems from the observation that OONI experiments require input that is more complex than a URL in several cases. For example, to correctly test HTTP/3 websites we would need to know whether a website supports HTTP/3. We started off by refactoring the signature of OONI experiment’s “Run” function to simplify the changing of arguments passed to each experiment ( https://github.com/ooni/probe-cli/pull/983), without having to refactor ~30 Go packages each time. *## OONI Explorer* We continued to work on the user feedback reporting mechanism ( https://github.com/ooni/explorer/pull/790), focusing on usability aspects with the help of a UI/UX designer. In order to improve the current OONI Explorer interface and plan new features, we started conducting UX research. We added support for RTL languages and localized date formatting across OONI Explorer as part of localization work ( https://github.com/ooni/explorer/pull/805). We also fixed the back navigation on pages that include filter params in the URL ( https://github.com/ooni/explorer/pull/815) and we added support for filtering by anomalies (https://github.com/ooni/explorer/pull/816) on the Measurement Aggregation Toolkit (MAT). *## OONI backend* We started reviewing the security report from Cure53 (who conducted an independent security audit of our software) and we deployed different mitigations based on the feedback we received. We implemented Bearer Token in the API as an improvement over the previous security cookie ( https://github.com/ooni/api/pull/313). We also implemented a mitigation to a minor security issue affecting parameter validation in the API documentation path "/apidocs". We deployed a fix to hide Clickhouse error traces from the API output for improved security. We also carried out some tests to compress ooniprobe binaries using UPX. We updated the fastpath pipeline ( https://github.com/ooni/pipeline/pull/401) to support the updated version of the Signal test. *## OONI Test Lists Editor* Following the backend changes to authentication, the cookie based authentication was replaced by a Bearer Token one in the Test Lists Editor: https://github.com/ooni/test-lists-ui/pull/68 *## Automating censorship detection and characterization based on OONI measurements* We created a new data analysis tool (https://github.com/ooni/data) for exposing anomaly details of website measurements and characterizing website blocking. Some of the building blocks for designing this tool were previously applied in practice as part of our investigation into emergent website blocks in Russia ( https://ooni.org/post/2022-russia-blocks-amid-ru-ua-conflict/), where we were able to characterize the blocking methods across ISPs ( https://ooni.org/post/2022-russia-blocks-amid-ru-ua-conflict/#blocked-websi… ). The high level architecture ( https://github.com/ooni/data#architecture-overview) of this new data analysis tool is that of transforming the raw network measurements coming from OONI Probes into normalized and post-processed “Observations” ( https://github.com/ooni/data#observation-generation). These Observations are “time stamped” statements about some network condition that was observed by a particular vantage point. These observations are then in turn post-processed to generate what we call an “Experiment result” ( https://github.com/ooni/data#experimentresult-generation), which attributes to a set of observations from an individual test run a series of outcomes with an associated level of confidence. In order to reach a design that would be flexible enough to be expanded, but also performant enough to run in a reasonable amount of time, we implemented several prototype iterations of this design ( https://github.com/ooni/data/pull/15). Throughout the process we carried out benchmarks of key components to identify any performance bottlenecks, working to resolve them (https://github.com/ooni/data/issues/4). The last iteration (https://github.com/ooni/data/pull/17), based on some cursory benchmarks (https://github.com/ooni/data/pull/17#issue-1452100512), should be able to generate observations from raw OONI measurements in less than a week and re-generate all experiment results in about a day, running on a single machine. Through this data analysis tool (which anyone can run on their own computer pointing it at the raw public OONI dataset: https://github.com/ooni/data#developer-setup), it’s possible to gain a much deeper understanding about the root cause of a blocking event, and to carry out more advanced research that might not be captured by the existing analysis. The new design is an important shift in how we view results from OONI measurements and we are confident that it will allow us to characterize blocks in a much richer way. It also opens the door for empowering researchers to carry out more advanced censorship investigations using our datasets. *## Creating a Social Media Censorship Alert System* To enable human rights defenders to rapidly respond to social media blocks, we aim to create a Social Media Censorship Alert System which allows access to real-time OONI data on the blocking of major social media platforms around the world. In November 2022, we started the initial research process for designing this new platform, which involved experimenting with implementations for social media blocking event detection. Specifically, we experimented with different designs to provide lightweight event detection without impacting the backend infrastructure. As part of this initial experimentation, we applied analysis logic for issuing an “alert” when anomalies are present for selected social media targets (the initial experiment involved Twitter and Facebook). We set the thresholds for issuing “alerts” based on moving averages, as the volume and frequency of relevant measurements varies significantly from ASN to ASN, and country to country over time. Based on our experimentation, we created Jupyter Notebook charts to investigate ways to classify groups of measurements as blocking or unblocking events. *## Social media blocked in Turkey* On 13th November 2022, following the bomb explosion in Istanbul, some ISPs in Turkey blocked access to Twitter, Instagram, Facebook, and YouTube. We responded by sharing real-time OONI data collected from Turkey (along with a MAT chart), documenting the blocks: https://twitter.com/OpenObservatory/status/1591871076677701635 Our tweet was cited in articles published by: * Global Voices: https://globalvoices.org/2022/11/13/turkey-throttles-internet-access-follow… * Coda Story: https://www.codastory.com/newsletters/apple-beijing-turkey-istanbul-blast/#… *## Test list updates* In collaboration with community members, we updated the test list for Kazakhstan (https://github.com/citizenlab/test-lists/pull/1182). We also reviewed and merged updates to the test list for Malaysia ( https://github.com/citizenlab/test-lists/pull/1175) contributed by our partner, Sinar Project. *## Community use of OONI tools and data### OONI censorship measurement campaign during Malaysia’s 2022 general elections* Leading up to and during Malaysia’s 2022 elections, our partner, Sinar Project, coordinated an OONI Probe measurement campaign (using OONI Run widgets for custom testing) to monitor potential censorship events. Information about the censorship measurement campaign, along with OONI Run widgets for relevant testing (including election monitoring and political party websites, among others), was published on their website: https://sinarproject.org/digital-rights/measuring-and-detecting-network-int… Sinar Project also shared relevant OONI Probe testing instructions through this Twitter thread: https://twitter.com/sinarproject/status/1592879196610920448 *### OONI censorship measurement testing during Kazakhstan’s 2022 elections* Leading up to Kazakhstan’s 2022 elections, we collaborated with local groups on creating a custom list of websites that are most relevant to test in Kazakhstan for censorship. To enable the testing of such sites during the elections, we embedded an OONI Run widget (https://github.com/ooni/ooni.org/pull/1304) into our website: https://ooni.org/get-involved/run (the “Kazakhstan (38 URLs)” widget) We shared relevant OONI Probe testing instructions on Twitter: https://twitter.com/OpenObservatory/status/1593988313932828675 We also wrote documentation providing relevant OONI Probe testing instructions for community members in Kazakhstan. *### Global Voices blog post on the blocking of a website that digitizes books in Mexico* Global Voices published a blog post documenting the blocking of a website run by a Mexican collective that digitizes books as a political position. This post cites OONI data on the block. Their blog post is available here: https://globalvoices.org/2022/11/30/mexican-collective-pirating-books-to-ma… *## Community activities### OONI workshop for civil society in Kazakhstan* On 3rd November 2022, OONI’s Elizaveta facilitated an online OONI workshop for civil society groups in Kazakhstan. Information about the workshop is available here: https://www.facebook.com/medianetkaz/posts/pfbid02nhnfrTotir9yaeZA3EJtG64oi… *### OONI presentations at IETF 115* On 8th November 2022, OONI’s Simone presented our censorship measurements in Iran as part of the Internet Architecture Board (IAB) open meeting during IETF 115 (https://www.ietf.org/live/ietf115-iab-open/). His IAB Open presentation can be viewed here: https://www.youtube.com/watch?v=7_G6XVjlun8&t=1730s On 9th November 2022, OONI’s Simone presented our strategies for measuring the blocking and throttling of encrypted protocols (including HTTPS, DNS over HTTPS, and HTTP/3) as part of the Privacy Enhancements and Assessments Research Group (PEARG) meeting during IETF 115 ( https://datatracker.ietf.org/meeting/115/session/pearg). His PEARG presentation can be viewed here: https://www.youtube.com/watch?v=NO3pw3_hees&t=3592s *### OONI workshop for journalists in Central Asia* On 10th November 2022, OONI’s Maria facilitated an online OONI workshop for journalists in Central Asia. *### OONI presentation at State of the Onion 2022* On 16th November 2022, OONI’s Maria presented OONI highlights from 2022, as well as upcoming OONI projects for 2023 as part of the Tor Project’s annual State of the Onion event ( https://blog.torproject.org/state-of-the-onion-2022/). The live-streamed event can be viewed here: https://www.youtube.com/watch?v=O-7k0PjnBbk *### OONI presentation at DataFest 2022* On 17th November 2022, OONI’s Elizaveta presented OONI at DataFest 2022 in Tbilisi (https://www.datafest.ge/schedule). As part of this presentation, Elizaveta explained how journalists can use OONI’s Measurement Aggregation Toolkit (MAT) to investigate internet censorship around the world. *### OONI datathon at DataFest 2022* On 19th November 2022, we facilitated an OONI datathon as part of DataFest 2022 in Tbilisi ((https://www.datafest.ge/datathons). During the OONI datathon, participants worked on analyzing OONI data collected from Russia, Kazakhstan and Uzbekistan to investigate local internet censorship. *### OONI presentation at Internet Without Borders conference* On 18th November 2022, OONI’s Arturo presented censorship measurement findings from Russia as part of the Internet Without Borders conference in Paris ( https://web.archive.org/web/20221101023014/https://internetborders.net/conf… ). *### OONI hackathon at Internet Without Borders conference* On 19th November 2022, we facilitated an OONI hackathon as part of the Internet Without Borders conference in Paris ( https://internetborders.net/ooni-challenge/). During the hackathon, a participant wrote new SMTP(s), IMAP(s) ( https://github.com/ooni/probe-cli/pull/989) and bittorrent experiments ( https://github.com/ooni/probe-cli/pull/986) for OONI Probe! *### OONI workshops in Kyrgyzstan* Between 24th–26th November 2022, OONI’s Elizaveta traveled to Kyrgyzstan to facilitate 5 in-person OONI workshops for local journalists, students and civil society organizations. As part of these workshops, Elizaveta explained how they can use OONI Probe to collect internet measurements and how they can use OONI data to investigate internet censorship. *### OONI workshop in Kazakhstan by Access Now* On 18th November 2022, Access Now organized an offline event for elections observers in Kazakhstan. As part of this event, Anastasia Zhyrmont from Access Now facilitated an OONI workshop and explained how election observers can contribute to collecting internet measurements during the elections. *### OONI Community Meeting* On 29th November 2022, we hosted the monthly OONI Community Meeting on our Slack channel (https://slack.ooni.org/), during which we discussed the following topics: 1) Strategies for monitoring potential social media blocks during Indonesia's 2024 elections. Discussion of how blocks are implemented in Indonesia and improving the data quality of OONI measurements. 2) Requesting feedback on the type of materials that community members would find most useful and would like to see in the upcoming OONI Outreach Kit. *## Userbase* In November 2022, 61,907,164 OONI Probe measurements were collected from 2,851 AS networks in 166 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/ ~ OONI team.

1 0

OONI Monthly Report: October 2022
by Maria Xynou 11 Jan '23

11 Jan '23

Hello, Happy New Year! We'd like to share updates from the OONI team over the last 3 months. Below I share our October 2022 report, and I'll follow-up to share our November and December 2022 reports. *# OONI Monthly Report: October 2022* Throughout October 2022, the OONI team worked on the following sprints: * Sprint 75 (1st-9th October 2022) * Sprint 76 (10th-23rd October 2022) * Sprint 77 (24th-31st October 2022) Our work can be tracked through the various OONI GitHub repositories: https://github.com/ooni Highlights are shared in this report below. *## New OONI Team member* In October 2022, Kathrin Elmenhorst joined the OONI team! Kathrin will work on improving the data quality of OONI measurements and adding support to OONI Probe for new experiments. *## Launch of Advocacy Assembly OONI training course* On 30th September 2022, Small Media’s Advocacy Assembly launched our online OONI training course ( https://advocacyassembly.org/en/courses/63/#/chapter/1/lesson/1). Following the launch, we published a blog post to share the course with our community: https://ooni.org/post/2022-ooni-training-course-advocacy-assembly/ This 90-minute free, online training course provides a deep-dive into measuring internet censorship with OONI Probe, and using OONI Explorer to access and interpret real-time OONI data collected from around the world. Designed for human rights defenders, activists, journalists, and researchers, the course includes a mix of videos, screencasts, slides, quizzes and hands-on exercises. It also features case study videos from OONI community members, and the course is available in English, Arabic, Spanish, and Farsi. *## OONI Probe Desktop* In October 2022, we released OONI Probe Desktop 3.8.1 ( https://github.com/ooni/probe-desktop/releases/tag/v3.8.1) which was supported by OONI Probe CLI 3.16.4, and which fixed an issue affecting the OONI Probe Signal test. Specifically, Signal changed their root CA, resulting in our Signal test leading to false positives globally. We documented this issue and the fix we implemented in the following ticket: https://github.com/ooni/probe/issues/2344 *## Farsi translation of OONI Probe Mobile and Desktop User Guides* Thanks to our community’s localization efforts, we published a Farsi translation of: * OONI Probe Mobile User Guide: https://ooni.org/fa/support/ooni-probe-mobile * OONI Probe Desktop User Guide: https://ooni.org/fa/support/ooni-probe-desktop *## OONI Run* As part of our work to release an improved version of OONI Run (“OONI Run v2”), we have written a specification describing the new functionality. In October 2022, we updated the specification to include more details about the descriptor format ( https://github.com/ooni/spec/pull/249#issuecomment-1286974194). *## Expanding censorship measurement methodologies* We started moving traffic generated by OONI Probe to a remote host at the TCP/IP level to flexibly implement integration testing in a censored environment. The overall idea has been documented in the following ticket: https://github.com/ooni/probe/issues/2340 The main pull request (https://github.com/ooni/probe-cli/pull/969) includes a fully working prototype. We also merged pull requests that re-introduce the TProxy functionality (https://github.com/ooni/probe-cli/pull/975) and other functionalities (https://github.com/ooni/probe-cli/pull/976). *## OONI Explorer* In October 2022, we continued working on OONI Explorer localization efforts, which involved moving hardcoded text to translation files and uploading them to Transifex in order for translators to fill in missing strings (https://github.com/ooni/explorer/pull/805). We also fixed the text alignment for languages that require RTL support ( https://github.com/ooni/ooni.org/issues/1264, https://github.com/ooni/ooni.org/pull/1269). We continued working on OONI Explorer user accounts and on the user feedback reporting mechanism, which will enable users to login and contribute feedback about each measurement ( https://github.com/ooni/explorer/pull/790). We also worked on the following OONI Explorer improvements: * Build process optimisations that result in smaller Docker image ( https://github.com/ooni/explorer/pull/807); * Replaced Babel with SWC for faster code transpilation ( https://github.com/ooni/explorer/pull/804); * Refactored the MAT filters to virtualize rows which improves the performance and memory usage when we’re showing a big list of rows ( https://github.com/ooni/explorer/pull/809); * Removed unused environment variables to reduce ambiguity ( https://github.com/ooni/explorer/pull/807); * Refactored the OONI Explorer Search page to move the API request from server side to client and improve rendering speed ( https://github.com/ooni/explorer/pull/799). Moreover, we contracted a designer to help us improve the user interface of OONI Explorer pages over the next months. *## OONI Test Lists Editor* Due to the API changes on the backend, code was refactored to only use one endpoint instead of three to retrieve data. These changes contribute to simpler code and improved performance of the Test Lists Editor: https://github.com/ooni/test-lists-ui/pull/67/ *## OONI backend* In October 2022, we worked on the following backend activities: * Made backend adjustments to annotate OONI Probe Signal measurements as “failed” as a temporary solution to an issue ( https://github.com/ooni/pipeline/pull/400), and updated the measurements in the database; * Looking into legacy-probe entry points and worked towards disabling the onion service; * Disabled the MySQL and PostgreSQL interfaces ( https://github.com/ClickHouse/ClickHouse/issues/40994); * Test list API: Added measurement summary statistics in the test list UI/API to enable review (https://github.com/ooni/api/pull/309) and added statistics from measurement feedback; * Backend changes to handle unsupported countries in test lists ( https://github.com/ooni/api/pull/310); * Wrote an internal design document to look into moving away from orchestrate for login/registration and documented the next steps; * Added a probe login metric and implemented other fixes ( https://github.com/ooni/api/pull/311); * Fixed the API aggregation time range (https://github.com/ooni/api/pull/309 ). *# Advancing OONI data analysis* We made progress on our new data analysis tool (https://github.com/ooni/data), through which we aim to automate the detection and characterization of more forms of internet censorship. Specifically, we added support for generating blocking events based on OONI Probe Signal measurements ( https://github.com/ooni/data/pull/10), we introduced a shared data directory for storing assets needed for performing analysis tasks (eg. GeoIP databases and blockpage fingerprints) ( https://github.com/ooni/data/pull/8), and we improved testing and the CLI ( https://github.com/ooni/data/pull/9). *## Published new OONI Code of Conduct* In October 2022, we published a new version of the OONI Code of Conduct (CoC): https://ooni.org/get-involved/code-of-conduct/ To help ensure a safe and inclusive environment for the global OONI community, we created a new Incident Response Committee with the goal of improving OONI’s Code of Conduct (CoC) and addressing any CoC violations when they occur. In collaboration with the Committee and based on community feedback collected from several meetings, we re-wrote and published a new version of OONI’s Code of Conduct (CoC). *## Published a survey to collect community feedback on OONI’s strategic priorities* Leading up to OONI’s 10th Anniversary (5th December 2022), we wanted to collect community feedback to help inform the development of our strategic priorities for the future. As the OONI community has always been at the heart of our work, we consider it essential that community needs continue to inform OONI’s future goals and priorities. We therefore circulated a survey to collect community feedback that can help shape OONI’s strategic priorities for the future. On 20th October 2022, we published a blog post which links to the survey: https://ooni.org/post/2022-survey-ooni-strategic-priorities/ We thank all community members for their valuable feedback! *## Test list updates* In October 2022, we discussed with Netalitica how we can improve our policies for updating test lists, with the goal of improving the quality of measurements and simplifying the workflow of researchers. Specifically, we discussed cases that involve updating URLs to HTTPS, how to handle parked domains or domains that otherwise no longer exist, and about involving researchers in the process of determining which URLs to configure for automated prioritized testing. We reviewed and merged multiple pull requests that involve updates to the Thai test list, contributed by our partner, Thai Netizen Network ( https://ooni.org/partners/thai-netizen-network/). These pull requests are the following: https://github.com/citizenlab/test-lists/pull/1152, https://github.com/citizenlab/test-lists/pull/1169, https://github.com/citizenlab/test-lists/pull/1168, https://github.com/citizenlab/test-lists/pull/1167, https://github.com/citizenlab/test-lists/pull/1166, https://github.com/citizenlab/test-lists/pull/1165. We also reviewed the pull requests contributed (via our Test Lists Editor) by community members for the test lists of France ( https://github.com/citizenlab/test-lists/pull/1125), the Global test list ( https://github.com/citizenlab/test-lists/pull/1162, https://github.com/citizenlab/test-lists/pull/1163), and several other test lists (https://github.com/citizenlab/test-lists/pull/1159). *## OONI citations### Freedom on the Net Report 2022* OONI data (and OONI research reports) are cited in the following Freedom on the Net 2022 ( https://freedomhouse.org/report/freedom-net/2022/countering-authoritarian-o…) country reports: * Iran: https://freedomhouse.org/country/iran/freedom-net/2022 * Myanmar: https://freedomhouse.org/country/myanmar/freedom-net/2022 * Malaysia: https://freedomhouse.org/country/malaysia/freedom-net/2022 * Cuba: https://freedomhouse.org/country/cuba/freedom-net/2022 * Azerbaijan: https://freedomhouse.org/country/azerbaijan/freedom-net/2022 * Colombia: https://freedomhouse.org/country/colombia/freedom-net/2022 * Ethiopia: https://freedomhouse.org/country/ethiopia/freedom-net/2022 * Italy: https://freedomhouse.org/country/italy/freedom-net/2022 * Jordan: https://freedomhouse.org/country/jordan/freedom-net/2022 * Kenya: https://freedomhouse.org/country/kenya/freedom-net/2022 * Saudi Arabia: https://freedomhouse.org/country/saudi-arabia/freedom-net/2022 * Tunisia: https://freedomhouse.org/country/tunisia/freedom-net/2022 * UAE: https://freedomhouse.org/country/united-arab-emirates/freedom-net/2022 * Uzbekistan: https://freedomhouse.org/country/uzbekistan/freedom-net/2022 * Zambia: https://freedomhouse.org/country/zambia/freedom-net/2022 *### Ranking Digital Rights Research Lab* Ranking Digital Rights (RDR) launched a new Research Lab which discusses how researchers can make use of OONI tools and data (among other resources). The RDR Research Lab is available here: https://rankingdigitalrights.org/research-lab/preparing-your-research/ *### Team CommUNITY’s publication* Team CommUNITY published a blog post (“Internet Freedom Tools, Resources & Actions to Support Iran’s Feminist Uprising”) which cites our research report on new blocks that emerged in Iran amid protests ( https://ooni.org/post/2022-iran-blocks-social-media-mahsa-amini-protests/). Team CommUNITY’s blog post is available here: https://www.digitalrights.community/blog/tools-resources-actions-to-support… *## Community activities### OTF Shutdown Getdown Event* Between 12th-14th October 2022, OONI’s Maria and Arturo traveled to Oxford to attend the OTF Shutdown Getdown event. This 3-day event brought internet measurement projects together to share skills and knowledge around measuring internet censorship and shutdowns. As part of our participation at this event, we: * Presented OONI as part of a lightning talk; * Facilitated a workshop to collect community feedback on OONI’s Measurement Aggregation Toolkit (MAT); * Facilitated a workshop to brainstorm on methods for measuring throttling; * Co-facilitated a workshop with M-Lab which aimed to introduce participants to the various measurement datasets available for investigating different aspects of internet shutdowns. *### OONI Community Meeting* On 25th October 2022, we hosted the monthly OONI Community Meeting on our Slack channel (https://slack.ooni.org/), during which we discussed the following topics: 1) Feature request: Adding support to the MAT for filtering measurements based on multiple domains 2) Upcoming 10th OONI Anniversary: Community survey and feedback for related activities 3) Test lists updates and re-thinking the test lists *### RIPE85* On 27th October 2022, OONI’s Arturo presented our research on measuring the blocking of encrypted DNS services (and how this experiment is now part of OONI Probe) at RIPE85: https://ripe85.ripe.net/ *## Userbase* In October 2022, 66,272,107 OONI Probe measurements were collected from 2,797 AS networks in 169 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/ ~ OONI team.

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

tor-project January 2023