Confirm before deleting multiple emails

[burninleo]

This feature would strongly decrease the risk of data-loss: An option to ask of more then X emails are about to be deleted (with a default of 10 emails). So deleting emails to the trash would normally not trigger a confirmation, but deleting a large number of emails would.

Background: In my organization I have multiple times observed someone losing their inbox (or another directory) that was abused as todo list, and a lot of important information was lost along. Finding the relevant emails in the trash along thousands of other emails was no actual option. They use IMAP and access the account from multiple devices. As IMAP is very poor on keeping track of the changes, no one exactly knows what happend ... howevery, most likely they pressed something like Ctrl+A and Del/Backspace on one of their devices, thinking they were typing in another window.

[Jon]

Thanks for submitting an idea to the Mozilla Connect community! Your idea is now open to votes (aka kudos) and comments.

[MattAuSupport]

Perhaps a corporate policy of enforced deleted mail folder emptying is what is needed so only a week or a months worth of mail is even in the trash. 

I find it rather interesting you think IMAP is not all that reliable.  It is in my experience very reliable, and the crux of your comments here indicate reliability in the face of user error.  If they delete everything, there is an undo (ctrl+Z) that works for most of those things., and when used multiple times can undo a number of steps.

Personally, If such an option were to be implemented, I would require an ability to turn the nanny message off entirely.  Especially in IMAP servers that do a delete and an add of every move between folders.

My observations from support are that loss of data is more often the result of impatience and repeated pushing of the same button than anything else.  So anything that speeds up the programs reactions are a bonus.  Things like email scanning and encrypted traffic scanning and saturated networks are all contributing factors here that are rarely looked at in the context of users problems.