[Standards] Message Retractions

JC Brand lists at opkode.com
Wed Sep 4 09:21:40 UTC 2019


Hi folks

I'm going to implement message retractions for Converse.js and while
researching what's available XEP-wise I came across this proposed
XEP from Lance Stout:

https://xmpp.org/extensions/inbox/message-retraction.html

It's from 2016 and was never accepted (i.e. assigned a XEP number).

I searched the standards list archives and found a thread where it was
discussed: https://mail.jabber.org/pipermail/standards/2016-October/031506.html

I also found a later thread (2018) about message corrections where people were
discussing putting message retractions in XEP-308 (Last message correction).
https://mail.jabber.org/pipermail/standards/2018-June/035154.html

I don't see any specific reason in the archives why the XEP wasn't advanced,
except that apparently enthusiasm for it fizzled out.

I'm not the author of the proposed XEP, but I'd like to see whether this can be
moved forward and I hereby offer to make any changes necessary to get it
accepted (unless Lance would like to do so himself).

Given that the proposed XEP is fairly old, there are a few things I'd like to
add to it to bring it up to date with latest practices.

These are:

* Mandate support for XEP-0359  Unique and stable stanza ids)
* Mention XEP-0421 (Anonymous unique occupant IDs) as an alternative to
  including the user's JID in the tombstone (section 4 example 5).
* Allow admins to see the original (now retracted) message when they receive
  the tombstone from MAM.
* Allow for supplying a reason why the message was deleted.

Concerning the question of putting this XEP inside an updated XEP-0308, I
propose keeping message retractions in a separate XEP, for these reasons:

* We want MUC/MIX admins to be able to retract other occupants' messages
  (but not "correct" them).
* A message can be corrected multiple times, but retracted only once.
* Message retraction has different implications for MAM than corrections.

I look forward to hearing your thoughts on this.

Regards
JC


More information about the Standards mailing list