See also http://lists.wikimedia.org/pipermail/ee/2014-May/000938.html

See also https://bugzilla.wikimedia.org/show_bug.cgi?id=58034

And I was not notified when a user signed their posts without using four tildes.

I am changing the priority from enhancement to normal, because the user who mentioned me in the most recent case seems to have intended to ping me and I wanted to know about the pings. Since user intentions weren't followed I think this should be categorized as a bug. I am changing importance from normal to high because I think the impact of this unexpected behavior is significant for users.

If we do this, we need to somehow make sure there's virtually no chance of false positives, since false positives will lead users to disable mention notifications, and that'd be bad for everyone involved.

In T66778#795605, @Jackmcbarn wrote:

If we do this, we need to somehow make sure there's virtually no chance of false positives, since false positives will lead users to disable mention notifications, and that'd be bad for everyone involved.

Agreed. Before considering this, we would need to carefully specify the behavior. Bartosz's idea (see description) seems workable if it can be done performantly.

Bartosz:

Good point. Luckily I think we could easily avoid that issue by just counting the number of mentions for each user in the old and new wikitext and only sending a notification is the number increased.

I don't think this would handle manual archiving. When comments are moved to the archive page, the edit expanding the archive page will increase the number of mentions on that page. (There is an offsetting edit removing mentions on the other page, but I don't think it's feasible to link those.).

This could be addressed by having a way to mark archive pages so mentions there were ignored. Flow has some ways of keeping track of archive page title conventions, but it's not complete.

Related: T128535: Consider ping parser function instead of link-based mention.

We discussed in triage this week, we don't have time to prioritize working on this in the short-to-medium term.

Idea: __PING__ and __NOPING__ magic words. Talk pages would be __PING__ by default. Content pages would be __NOPING__ by default. Wikis would typically add __NOPING__ to all archive templates. There could be safeguards against false positives, like "Ignore edits with multiple section headings" or "Ignore edits that start and end on different indentation levels." When an edit is ignored for one of these reasons, notify the user that no ping was sent (rather than the status quo, where you only get a "failed notification" message if the notification was sendable but not delivered.) That way false negatives aren't that big of a deal. And maybe have a way for users to intentionally disable pings for an edit, like by including "NOPING!" in their edit summary.

This would have the added benefit of obviating the need for templates like {{noping}} in countervandalism work. A page like [[en:WP:AIV]] could just have __NOPING__ on it and that would take care of that.

Alternately, treat the existing magic words __NEWSECTIONLINK__ and __NONEWSECTIONLINK__ the same way.