[ SmokeDetector | MS ] Bad keyword with email in answer, pattern-matching email in answer, potentially bad keyword in answer (184): NASM printf link by Anita Davenson on askubuntu.com (@Ollie)
@Lyxal I went over your recent feedbacks and invalidated a couple; on the whole, looking good, but perhaps abstain from providing feedback on stuff which isn't completely cut and dried (like obviously repeating spam, Quickbooks errors and CBD gummy bears etc) until you see how others here are reacting
thanks -- it obviously takes some time to learn what is typical and what is not; there are a few corner cases where you really have to know the context (like R-HUB and some returning trolls)
@Undo A couple of users posted questions written entirely in Chinese, including the titles, on [mathoverflow.net]. Having the post make no attempt to be in English and using Chinese characters in the title are usually strong indicators that the post is spam (hits three reasons, each in the low 80's for weight), but, obviously, these are not spam. Unfortunately, you got unlucky and were selected as a flagger on both questions, even though there are many eligible flaggers.
metasmoke.erwaysoftware.com/data/sql/queries/… shows worst ever is 25%, top 5 are over 10% but then the tail rather quickly approaches 0.5% (not sure I understand the query's parameters exactly; I guess the 10 to 60 bracket is to exclude all-out abusers as well as users with very few invalidated feedbacks)
... but I don't see you in there even though I think you should be in that bracket (also the results are anonymized to only show the numeric chat ID which makes it a bit tricky to map back to individual users)
some well-respected users with high invalidated rate are around 7% metasmoke.erwaysoftware.com/data/sql/queries/… but by quick glance the average is probably below 1% or perhaps 1.5% which seems to be the mode
flagging a lot of obvious spam obviously makes you look good on average (-; (my accuracy is 0.61%)
ah, okay. was just wondering if my 4.93% could be found under that ID
i've definitely self-modified a fair amount of stuff. especially since sometimes i tp something and then naa flag it on site and it changes my feedback
you mean you click TP here, then go to the site and click as NAA, and something changes your feedback in metasmoke too? There's a userscript which does that, and if you have it installed I guess it's not entirely trivial to turn it off when you want to
but nothing in the system itself which would do that on your behalf
when I was using Iridium as my browser it prevented that from working, but it had some other unattractive consequences too (-:
@hyper-neutrino I've been thinking the same thing, but I guess the functionality at this point would probably be delegated to a userscript for those who want it
metasmoke only stores the TP so in that respect no; sometimes it would be useful to see the difference when you try to find actual spam and not just all TPs with some search
@RyanM it doesn't seem to actually ask a question though. maybe i am reading into it wrong; it seems to be making a (positive) statement about online gambling
@hyper-neutrino The FIRE userscript is the one which will also flag from chat. FIRE doesn't raise NAA flags because, currently, it only flags through MS and MS doesn't raise NAA flags (changes to have MS be able to raise NAA flags were declined).
It's possible for FIRE to raise NAA flags through the SE API. I actually wrote the code to do so about 3 years ago, IIRC, but the original author of FIRE wouldn't permit the functionality of flagging through the SE API to be merged into the released version. That decision has since been overridden, so FIRE will, at some point in the not toooo distant future, have that capability.
pro tip: the naa button is disabled if it's a question :p
if you haven't installed them the userscripts are pretty helpful (i have fire, aim, and fdsc), though they might be a bit confusing at first so just going through ms review queue is probably better just to get the hang of things
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in url, potentially bad ip for hostname in body (390): Sexpuppen-Hersteller RealDoll by coydoll1997 on ja.stackoverflow.com
@tripleee I'm of the same mind. I don't think SO should tolerate people who aren't willing to make the slightest effort to meet the question asking rules
@Spevacus 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.
