@@ -27,41 +27,64 @@ For asynchronous communication, we use our [anti-censorship-team](https://lis
## Priorities for 2022
- make Tor accessible in China
- detect and categorize attempts to censor Tor
- improve the design and reliability of our software
- release our data and software for use by the broader anti-censorship community
- improve the performance of Snowflake so that Tor bootstraps reliably on a mobile phone in China
- deploy TapDance and Conjure as high collateral damage PT
- commit to a design for a reputation-based bridge distribution system
- streamline our private bridge setup and distribution process
- deploy probes in areas that are likely to censor Tor and collect pack captures and probe results for storage and analysis
- provide OONI with suggestions for improving the accuracy of OONI's Tor tests
- summarize the details of Tor blocking events with data from our probes and volunteers
- add more user metrics based monitoring and alert rules using prometheus
- deploy rdsys ad the new backend of bridgedb
- ensure that key infrastructure can survive machine outages and restarts
- remove hacky shims necessary for moat
- future improve the snowflake library api to allow easy integration of snowflake with other tools
- sanitize, publish and archive the results of our Tor reachability probes
- complete our documentation for each of our tools so that other organizations can run their own anti-censorship infrastructure
## Roadmap
[Priorities for previous years](previous-priorities)
### Q1
## Roadmap
* s30 O2.3.1 - Develop new and/or improve existing bridge selection and distribution strategies based on data collected about successful, effective methods per evaluation during O1.1.
* s30 conjure (2 months) - cecylia starting in March
* server side : the university
* start talking again with eric and his team
* define server side for them to setup bridge (documentation)
* client side : around 2 weeks
* write a conjure client based on their specification (tor pt part with their client side library)
* staging/testing
* deployment
* add it to alpha version of TB
* metrics (discuss with people that maintain conjure bridge)
* how many clients are connected
## Q3
* sponsor 30
* O2.3.1 - Develop new and/or improve existing bridge selection and distribution strategies based on data collected about successful, effective methods per evaluation during O1.1.
* conjure (2 months) (cecylia) <-- maintenance in Q3.
* s96 O1.1.1 Prepare the Snowflake system for a surge in operators and users. <-- scale in Q1/Q2 2022 <-- shell
* s96 O1.2: Increase the number of Snowflake bridges.
* s96 O1.2.2: Scale Tor reachability through mobile Snowflakes. <-- support to GP
* s96 O1.4: Increase the number of active obfs4 and HTTPT bridges.
* s96 O1.4.3: Monitor bridge health. <-- Meskio/Shel with Gus
* s96 O2.1: Make it easier for humans & harder for censors to get bridges from moat distributor. <-- Q1
* s96 O2.2: Deploy improved bridge distribution systems.
* s96 O2.2.2: Deploy next generation bridge distribution system (rdsys) <-- meskio -
* s96 O2.3: React and steer our response to censorship. <-- shel
* s96 O3.1: Improve automatic censorship detection during bootstrapping in Tor Browser (desktop and Android). <-- meskio
* sponsor 28 & extension: we will have the new developer full time into this project.
* sponsor 28 - Improve the performance of Snowflake for users in Asia (cecylia)
## Q2
* sponsor 30 O2.3.1 - Develop new and/or improve existing bridge selection and distribution strategies based on data collected about successful, effective methods per evaluation during O1.1. - Implement and deploy conjure - Cecylia
* sponsor 96 O1.1.1 Prepare the Snowflake system for a surge in operators and users. - shel
* sponsor 96 O1.2: Increase the number of Snowflake bridges. - shel
...
...
@@ -79,44 +102,35 @@ For asynchronous communication, we use our [anti-censorship-team](https://lis
* s96 O1.1.1 Prepare the Snowflake system for a surge in operators and users. <-- scale in Q1/Q2 2022 (shell)
### Q1
* s30 O2.3.1 - Develop new and/or improve existing bridge selection and distribution strategies based on data collected about successful, effective methods per evaluation during O1.1.
* s30 conjure (2 months) - cecylia starting in March
* server side : the university
* start talking again with eric and his team
* define server side for them to setup bridge (documentation)
* client side : around 2 weeks
* write a conjure client based on their specification (tor pt part with their client side library)
* staging/testing
* deployment
* add it to alpha version of TB
* metrics (discuss with people that maintain conjure bridge)
* how many clients are connected
* s96 O1.1.1 Prepare the Snowflake system for a surge in operators and users. <-- scale in Q1/Q2 2022 <-- shell
* s96 O1.2: Increase the number of Snowflake bridges.
* s96 O1.2.2: Scale Tor reachability through mobile Snowflakes. <-- support to GP
* s96 O1.3: Implement bridges with pluggable transport HTTPT support. <-- Q2/Q3 (Shell)
* s96 O1.4: Increase the number of active obfs4 and HTTPT bridges.
* s96 O1.4.3: Monitor bridge health. <-- Meskio/Shel with Gus
* s96 O2.1: Make it easier for humans & harder for censors to get bridges from moat distributor. <-- Q1
* s96 O2.2: Deploy improved bridge distribution systems.
* s96 O2.2.1.: Deploy the Salmon bridge distribution system. <-- Start as soon as possible (Meskio & Shell - support from Cecylia)
* s96 O2.3: React and steer our response to censorship. (Shel)
* s96 O4.3: Modify GetTor so that it can distribute Tor Browser via messaging apps <--- we are late - any time during 2022 (meskio)
* s96 O3.1 plan for deprecation of bridgedb (Q4)
* s96 O2.2.2: Deploy next generation bridge distribution system (rdsys) <-- meskio -
* s96 O2.3: React and steer our response to censorship. <-- shel
* s96 O3.1: Improve automatic censorship detection during bootstrapping in Tor Browser (desktop and Android). <-- meskio
* sponsor 28 & extension: we will have the new developer full time into this project.
## OKRs for Q1 and Q2 2022
- make Tor accessible in China
- detect and categorize attempts to censor Tor
- improve the design and reliability of our software
- release our data and software for use by the broader anti-censorship community
- improve the performance of Snowflake so that Tor bootstraps reliably on a mobile phone in China
- deploy TapDance and Conjure as high collateral damage PT
- commit to a design for a reputation-based bridge distribution system
- streamline our private bridge setup and distribution process
- deploy probes in areas that are likely to censor Tor and collect pack captures and probe results for storage and analysis
- provide OONI with suggestions for improving the accuracy of OONI's Tor tests
- summarize the details of Tor blocking events with data from our probes and volunteers
- add more user metrics based monitoring and alert rules using prometheus
- deploy rdsys ad the new backend of bridgedb
- ensure that key infrastructure can survive machine outages and restarts
- remove hacky shims necessary for moat
- future improve the snowflake library api to allow easy integration of snowflake with other tools
- sanitize, publish and archive the results of our Tor reachability probes
- complete our documentation for each of our tools so that other organizations can run their own anti-censorship infrastructure
[Priorities for previous years](previous-priorities)
* sponsor 28 - Improve the performance of Snowflake for users in Asia (cecylia)
You can follow up what we are working on in this [kanban board](http://eweiibe6tdjsdprb4px6rqrzzcsi22m4koia44kc5pcjr7nec2rlxyad.onion/groups/tpo/anti-censorship/-/boards).
