Replace Pidgin

For now, we’re doing all easy things we can to make Pidgin safer in Tails (AppArmor confinement, shipping support for only a couple protocol), so the current situation is not that scary, but people I trust say it’s riddled with security issues.

It would be good to have a long term plan to replace Pidgin with something safer, without losing the functionality we need.

Blueprint: https://gitlab.tails.boum.org/tails/blueprints/-/wikis/replace_Pidgin

Subtasks

• Can we stop including an IRC client by default? (#15816 - closed)
• Replace Pidgin: refine blueprint (#11686 - closed)
• Check if/how candidate IM clients match our requirements
• Evaluate candidate clients and end-to-end encryption:
  • Check Dino and Gajim wrt. security [intrigeri]
  • Test clients
    • Dino (on Buster: dino-im/buster-backports; on Bullseye: dino-im/bullseye-backports)
    • Gajim
    • What to test
      • OMEMO
      • OMEMO-encrypted group chat (not a requirement, but would be nice to have)
      • Optionally, interoperability with mobile XMPP clients
• Choose a new IM client
• Integrate this new IM client
  • Persistence
  • Automated tests
  • Documentation
  • Consider confining the new client with AppArmor
• Document how users can add IRC support back the... (#18052)
• A few months later (to give time for communities to migrate from OTR to OMEMO): remove Pidgin

Related issues

• Blocks Wayland (#12213 - closed) (the only X.Org specific bits remaining in our test suite are about Pidgin)