MSC1693: Specify how to handle rejected events in new state res

[erikjohnston]

It's possible for events in an an event's auth chain to be rejected due
to not having passed auth (based on the state at the time), so we need
to be explicit about how to handle that case.

Updates MSC #1442

Rendered full doc

[erikjohnston]

(Note that you can ask github to show a rich diff, alas it doesn't appear to have a stable link though)

[erikjohnston]

I also need to think a bit about how to deal with "reverse topological power ordering", in particular:

1. x's sender has a greater power level than y (calculated by looking at
   their respective auth events), or if

[richvdh]

I'm a bit confused about some things here.

In terms of process: I think we need to get some clarity on how this is supposed to work (hence #1694), but I don't think we should be changing existing proposal docs once they are accepted and merged into the spec.

[ara4n]

MSC processwise we really should have closed this once it was merged into the spec, and started a new MSC to fix the rejections stuff. But in the interest of minimising bureaucracy let's just wrap it up here and then close this (and merge the result again into the spec).

[erikjohnston]

MSC processwise we really should have closed this once it was merged into the spec, and started a new MSC to fix the rejections stuff. But in the interest of minimising bureaucracy let's just wrap it up here and then close this (and merge the result again into the spec).

TBH I thought it had been agreed that updating an existing proposal doc was fine in certain circumstances, like adding clarifications or whatever, so long as it is documented that it has been updated.

[KitsuneRal]

@mscbot resolve Fix typos

I didn't expect FCP to be broken by petty typo fixing request - sorry for the mess.

[mscbot]

🔔 This is now entering its final comment period, as per the review above. 🔔

[mscbot]

The final comment period, with a disposition to merge, as per the review above, is now complete.

[richvdh]

I believe this is implemented in synapse, so it should be spec-pr-missing, not finished-final-comment-period

[anoadragon453]

Good point, those old labels will confuse things. I do need to get the bot to switch to our custom labels still.

[turt2live]

Merged via #1773 🎉