Sean Quah

Faster joins: non-`lazy_load_members` `/sync`s should not block

4

Instead they should pretend that the room has not been joined yet. `/sync`s with `lazy_load_members` turned on should continue to return the room.

Faster joins: Handle leaving a room while re-syncing state

3

https://github.com/matrix-org/synapse/blob/7c6b2204d143550d81e5bf9612c4e69fe0866b4c/synapse/handlers/federation.py#L1512-L1515

Faster joins: Handle a second join while syncing state for the first one

3

We want to avoid launching a second partial state syncing job for the same room.

Faster joins: improve handling of race where we can fail to persist a client event with partial state after `_sync_partial_state_room` clears the partial state flag for a room

2

Follow on from #12988. Currently, when we encounter the race in #12988, we send a 429 Too Many Requests back to clients with the expectation that a retry happens. [@reivilibre...

When the `plumb-irc` command fails for a subset of rooms, an admin may want to run it a second time. However, on a second run, it emits spurious errors for...

Also figure out how this feature would interact with message edits.

When an invite fails, such as when using the !inviteme command, the bot spits out a very unhelpful message like "There was an error processing your command: undefined". It's not...

The bot sets up a video widget in auditoriums and jitsi+scoreboard widgets in talk rooms. If a moderator were to delete these by accident, they would have no way to...

When there is a gap between talks, the orange "Q&A has started" banner remains on the scoreboard widget until the next talk starts.

The command to build a single talk room looks like `!conference build talk K.fosdem 13564`. But if you accidentally submit `!conference build K.fosdem 13564` instead, the bot will go ahead...