Something different, but related: https://gerrit.wikimedia.org/r/80710

Few thoughts...

There are two options, "forwarding", in which the target user gets the notifications (with some indication that it has been forwarded) and the actual user does not, or "stalking" where both users get notifications and mark them as read individually.

I think stalking is a better option here, and opens up more uses.

To implement stalking we need:

• An interface where users can link accounts together
  • We could re-use the token system, but I think that would be icky.
  • Way to store linked accounts...?
• An interface to unlink accounts.
• Internal changes to detect if a user is being stalked, and create additional notifications for that
• Some way to annotate that a notification is a "stalked" one.

mano.queixo wrote:

It would help when an user is renamed, and in "historic pages" is needed to link to the old name.

Prioritization and scheduling of this bug is tracked on Mingle card https://mingle.corp.wikimedia.org/projects/flow/cards/215

pinkampersand.wikimedia wrote:

Just giving this a more clear title.

Feature requested recently onwiki at https://www.mediawiki.org/wiki/Topic:Syia5np4gmj123s6

Rather than forwarding this could also be done with a 'stalk' option where you can choose to stalk a users notifications (all of which link to publicly viewable info as far as I know)

They are not all public info.

E.g. which edit you are thanked for is not public, we just added excerpts of Special:EmailUser emails you received (even the *fact* that you received an email is not public).

However, the forwarding (with approval from both source and destination users) could potentially work.

If people want to receive notifications for accounts that they don't log into often, they can just set the notifications for those accounts to be emailed. I don't see any reason why we would need to build a new feature to address this case.

The Growth-Team doesn't plan to work on this in the short-to-medium term but patches are welcome. Before a patch is submitted it would be a good idea to seek consensus on the approach, though (see T50892#525096).