Posts that are repeated targets of spam edits should be flagged to moderators

Question

I just rejected a spam edit on a post today and out of curiosity decided to check the history of that post. It turns out that same post has been the target of repeated spam edit attempts in the past. Looking back at some other rejected edits in the history for other posts I see similar things have happened too. Some posts are the target for repeated spam edit attempts.

Thankfully all these edits were reviewed and rejected, but it strikes me as something that would be useful for moderators to know about. Perhaps a flag stating "{this post} has been repeatedly targeted for spam edits".

However this also brings up the issue of what a Mod can do about it. Aside from protecting the whole question (not ideal) or deleting the answer (again, not ideal) there's very little we can do. I think that by editing the answer so that it is longer and more substantial reduces the risk of it being targetted again, but that might not be appropriate.

So I suggest the following (and am open to alternative ideas):

If a post has had 2 or more different suggested edits reviewed and rejected as spam then it should automatically raise a flag and be blocked from being edited by <50 rep users.

Possibly the option to Protect the post itself could be applied, but I'm not 100% sure this is ideal because I don't want to prevent people from genuinely editing posts. Therefore I'm open to suggestions as to what actions a mod should be able to do on such posts.

Answer 1

There's some plumbing under those floorboards that you might not know about. We track these seemingly innocuous, generally terse posts that tend to attract a lot of spam edits, because spammers erroneously believe that spam on short posts has a better chance of going undetected. Yeah, I know, I made the same noise when someone told me that too.

If you suggest an edit to a post that has a certain number of edits rejected as spam recently and then that edit is rejected as spam - we deal you a blocking penalty in the spam system. In other words, known targets dynamically become honeypots. Cool, huh?

Where this breaks down is when large 'snow shoe' spam rings come into play, and start doing it from literally 50 or more distinct networks. When this happens, human intervention is often needed (I go run a query, find out who has been naughty and then block the whole steaming bunch).

I'm going to get with Marc and see if we can detect honeypots that don't seem to be effectively blocking, and possibly then raise a flag. If it actually gets to that point, it means a very unfortunate user has quite a lot of noise in their global inbox, and we like to keep that out.

To be fair, we do an excellent job of scaling up to meet large, coordinated gelatinous meat-like substance waves:

You'll note, the same amount of effort that everyone puts in doesn't move much - the system is already very well-trained on what's been trying to get in. It's just these occasional rogue bot nets that spring out of nowhere that need a human to intercede.

I don't yet know if I want to put that in as a flag, as mods can't yet do anything except tell us that it's there. Therefore it comes down to two possibilities:

• Funnel alerts like this directly to the community team (specifically Me, Shog or Jon)
• Find a way to give moderators more access to fix these situations without involving us.

No status yet, but yeah - something needs to go in.

Answer 2

This goes beyond spam, in my opinion. Sometimes clueless users will try to edit their own question inside existing answers, or put comments like "thanks" inside answers. Seen it all, and while no spam, those should be handled too.

So what I suggest is more "generic": if two edits from the same user (can also be anonymous) on the same post have been rejected in a row (within specific time range e.g. 1 day) then block this user from suggesting further edits on this post for X days, kind of "specific post edit ban".