[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, body starts with title and ends in url, pattern-matching website in body, potentially bad ip for hostname in body, +1 more (492): www.goodforfitn ess.com/zep hrofel-se/ by pagilituanoki on 3dprinting.SE
@Makyen That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer; append -force if you really want to do that.
If a user edits their post is there any way to force a rescan without doing a !!/report-force, or is it by design that there is one entry for each version in MetaSmoke?
@WaiHaLee If the user edits their post, and it's not a grace period edit, then SD should automatically detect that there was an edit and re-scan automatically.
@WaiHaLee There is not currently a restriction that there's only one MS report per version of an SE post. There are many SE posts that have been reported multiple times, but have never been edited. Obviously, this happens mostly for FP posts, as TP posts tend to be deleted.
Lately there seems to have been a pattern of grace-period edits, e.g. this post 1, 2 or this post 1, 2. Metasmoke doesn't appear to monitor grace-period edits: if there are edits, should we !!/report-force them?
@WaiHaLee SE provides no way of detecting grace period edits other than getting the current post body, title and tags and comparing them against what they originally were. To do this would effectively mean we would have to re-check every post after 6–8 minutes (5 minutes for the grace period plus time for the SE API to update, which SE claims can be ~1 minute).
Doing this could double the number of SE API accesses we'd need to make (this could be optimized a bit, but would cause further delay in detection). We definitely don't have enough SE API quota to support doubling our accesses.
[ SmokeDetector | MS ] Few unique characters in answer, mostly punctuation marks in answer, repeating characters in answer (232): How to limit div width? by pipo on stackoverflow.com
@EffectiveRobot Thanks for reporting it. Here is a reasonable place to do so. Obviously, if you feel it's spam or rude/abusive, then you should also flag it. That specific post had already been reported a couple of messages up from your first post here.
Rebooting MS.
However, the last time Redis was OOM, the reboot button was not sufficient. @Undo
@EffectiveRobot There should be a row of buttons (which look just like text) immediately below the post on the left (e.g. "share" "edit" "flag"). If you click on "flag" a modal dialog should pop up allowing you to select a type of flag to raise. In this case, I wouldn't raise one, as the post has been edited.
@EffectiveRobot Yeah, there's quite a bit. You can take the tour, take a look at the help center, and read the dialogs for actions as they are displayed. There's also a lot on Stack Overflow Meta, including a FAQ.
It's... hard... to argue in favor of writing an algorithm and making a bot spend time calculating which of the two flag types to use when they are, in actuality, the exact same flag. It's not even Coke vs. Pepsi, it's Pepsi vs. a second bottle of Pepsi that's using a different but still valid Pepsi logo.
@PaulWhite FYI, on metasmoke.erwaysoftware.com/post/159775, we count nonsense as R/A, which is TP. "NAA" feedback is just glorified "FP" feedback for much of the system.
@CodyGray thanks, we're pretty liberal with those flags by design. Fixed now.
@CodyGray Had the discussion a bunch with mods too. I eventually set a "this isn't worth debating" policy; in the few cases where mods decline spam flags on R/A we just ignore it and move on.
All our systems can flag R/A, and do when told to by a human (i.e. from a userscript), but it's simply not worth the investment to make the distinction between spam and R/A automatically.
@CodyGray yeah, it's an internal tool with a steep learning curve. If you've got a specific thing, might be able to help.
@CodyGray It's a tool meant for internal consumption that's developed by a limited number of people with a limited amount of time. If it works, it works. UX is secondary, frankly.
@CodyGray It would help quite a bit if you could provide more detail as to what you find confusing or have a problem with. Additional detail would allow us to have a decent chance at resolving the issue, rather than us just taking stabs in the dark as to what might be the issue you are having.
You can't just wave your hands and say "it's an internal tool" when people who are not part of your group have to deal with it. That makes it explicitly not internal.
And makes it very obvious why you've had a lot of friction when this is your attitude about it.
@CodyGray That auto other flag? That links to the Charcoal website. Not to metasmoke. If you have an issue with what that linked page says, happy to look at it. If you're complaining because you decided to go look at the nuts and bolts without taking the time to understand first, that's not helpful.
Obviously I am missing something here. Nuts and bolts? I didn't look at any nuts and bolts. I looked at the output that was dumped into my lap as a moderator.
The flag message makes very little sense. When I tried to investigate, I became even more confused.