Comment 2 for bug 106162

I think our code for what to notify to subscribers to the primary bug has evolved since this was filed. That said, bugs with hundreds of duplicates will generate terrible noise if we notify when a new duplicate is found. Its probably only interesting to the duplicate's subscribers that it has been folded into the primary bug.

That said, if we do 'notify subscribers to the primary bug of changes to duplicate bugs', then its entirely appropriate that these comments be forwarded: with ajax bug updates the ordering of changes like 'mark as duplicate' will often be not strictly semantic. E.g. someone may mark as duplicate then explain why, or vice versa.

That said, if two separate people comment within the timeout period for bug notifications, the only events that really makes sense to forward by this logic are the ones from the person that marked it as a duplicate. Which we don't do today.

So I'm going to downgrade this to low: most of the time that this happens it will be the person making the bug a duplicate; the noise from duplicates is a real issue but orthogonal to this fairly precise bug - and we probably need a LEP + re-review of the stories around notifications to change that.