Submitted as an answer since I couldn't find a way to match below into neither of comment / chat message / meta question.
Original revision of the proposed FAQ contains a statement that makes me uncomfortable:
If the post is fixable, and the poster is not given time to fix it before the flag is issued, the flag will most likely be declined.
I fail to see how above approach can scale. It works fine when site gets 1 to 3, maybe 5 bad answers a day, but now traffic stats say we are getting 16 questions/day, with about 4x as much answers meaning 60-80 answers a day.
From what I observe today, it looks quite realistic for site to eventually receive 200-300 answers a day, 20-30 of which would be of the low quality.
If above approach allows to handle 20-30 low quality answers a day, I would like to see an explanation on how that could be.
In particular, I would appreciate explanation on how this would be better compared to an alternative flow, when answer gets deleted (with mod comment - there is a feature that guarantees such a comment to reach the author), author edits it and flags for undeletion.
Only benefits that I can see in theory of letting low quality posts stay visible are:
- Less burden for moderators to deal with undelete flags from post authors
- Opportunity for non-10K editors to salvage these posts
- Opportunity for community regulars to indicate troublesome posts by voting it down, thus explicitly teaching other readers about site quality norms
- Having moderators handling every immediately submitted flag won't scale
Let's take a closer look at above...
Less burden to deal with undelete flags - per my observations, there were no complaints about this at larger sites (SO and Programmers) so it's unlikely to happen to start with. And, even if it happens, there is a known, tried and true way to deal with: flags can be off-loaded to a dedicated review queue for users with sufficient privileges (10K in this case) to handle.
Opportunity for others to do salvaging edits - it's unrealistic that active editors will be able to cope with substantial loads of stuff like that. Again, looking at larger sites like SO and Programmers, this doesn't happen there. Also, which is probably even more important, it doesn't feel fair to off-load this to editors, if you consider that salvaging bad answers would essentially block them from improving those that already have reasonably good content to start with ("optimize for pearls not sand").
Indicate troublesome posts by voting it down - given the observed frequency of the voting pattern that doesn't have politically correct name yet, it is more likely that opposite will happen, and low quality answers voted up by outsiders for popularity reasons will make broken windows, teaching site visitors the wrong way. How many upvotes were there on that just-quit answer in an agony aunt question? What did it teach readers until it was deleted?
Having moderators handling every immediately submitted flag won't scale - right, this won't scale. Thing is though, experience of larger sites (where delayed flagging is used) indicates that alternative won't scale either. Here I am specifically referring to changes introduced in low quality review queue a while back when six 2K users were allowed to delete answers. You see, delayed flagging doesn't help. Sites practicing it experience scaling issues and are forced to delegate deletion of low quality posts to wider community anyway.
Summing up, mentioned statement seems to do more harm than good for the site.
Consider changing approach, so that troublesome post is quickly deleted with mod comment, without keeping it visible (expecting that author willing to salvage would find and edit it while deleted, then flag for moderator to review and undelete).
For the sake of completeness, I can see how site like SO can afford such an approach in their site-specific guidelines: having 2,000 (that's two thousands) trusted users capable of quickly, directly removing bad stuff at their own discretion, without any flags at all makes it not that important. But Workplace is not like that and is unlikely to become like that anytime soon.