MSC4294: Ignore and mass ignore invites

[grinapo]

Rendered

[grinapo]

There is a long history of ignored open ignore invites proposals. There is also a summary MSC to list all the MSCs about the topic.

• MSC2270: Proposal for Ignoring Invites #2270
• MSC3840: Ignore invites #3840
• MSC4155: Invite filtering #4155
• MSC3847: Ignoring invites with policy rooms #3847
• MSC3659: Invite Rules #3659
• MSC4264: Tokens for Contacting Accounts or Joining Semi-Public Rooms #4264

All of these proposals are open and have been for more than 5 years. They are fairly complex, and handle various complex combinations of ignores and allowlists, for users, servers, and patterns, as well as some preauthorisation.

This current proposal is deliberately very simple, using the already established ignore machanism, which should be easy to implement using parts of the same code. It is hoped that it will not take years to be accepted, since the problem is acute.

cc: @Johennes

[turt2live]

Implementation requirements:

• Client
• Server

[turt2live]

Your MSC currently uses the master branch of your fork, which has already shown to create compatibility issues with another MSC of yours: #2063 (comment)

Please consider switching the branch to a feature branch instead of master.

[grinapo]

@turt2live
Please consider switching the branch to a feature branch instead of master.

(I hate git)
What magical commands shall be casted upon the repo which would move it onto a feature branch while keeping this PR intact?

[turt2live]

The internet is telling me it's not possible to change the source branch after the PR is opened, but I'm pretty sure we've done that before. For all the hassle, a new PR is probably easiest and causes no harm.

[turt2live]

Overall, I feel this MSC is in direct competition with #4155 which has a number of implementations and experimentation. If there's parts that 4155 doesn't cover, this MSC would be best positioned to propose extensions to 4155 so that it may protect more users.