It said that matched because it said "nologyadvice.co", but the actual URL was "technologyadvice.com — that seems like a problem to report something that contains another URL that's bad
@gparyani I am aware of 39 code admins. Currently in this room: bertieb, Nisse Engström, Makyen, Olivia, Suraj Rao, John Dvorak. Not currently in this room: iBug, doppelgreener, WELZ, CalvT, A J, a spaghetto, NobodyNada, Thomas Ward, Ferrybig, angussidney, K.Dᴀᴠɪs, tripleee, Mithrandir, Ashish Ahuja, Magisch, ArtOfCode, Shree, Erik the Outgolfer, micsthepick, Byte Commander, ProgramFOX, paper1111, Henders, DavidPostill, double-beep, Tetsuya Yamamoto, Andy, bwDraco, Undo, Glorfindel, thesecretmaster, Machavity, Paul Stenne.
@Stormblessed It means that the website matched one or more patterns which are defined in findspam.py. There are quite a number of patterns used for that detection reason. The patterns used are based on things previously seen or expected to be seen for website/domain names.
The regular expression (regex) pattern which matched that post is (\w{11}(idea|income|sale)|\w{6}(?<!notebook)(advice|problog|review))s?\.(co|net|in(\W|fo)|us), which can currently be found on line 1883 of findspam.py. The pattern is actually applied as a rule on line 2061. You can see a regex101 test of that regex here.
"Pattern-matching website in {}" is currently: in answers: 78.17%TP, 13.69% FP, 6.58% NAA/both (4,893 detections); in questions: 97.12%TP, 2.77% FP, 0.0% NAA/both (51,790 detections); in titles: 99.78%TP, 0.22% FP, 0.0% NAA/both (16,264 detections)
It is not directly related to the watchlist (cc @Machavity). The watchlist is exclusively used for the detection reason "Potentially bad keyword in {}", which as a "Potentially" reason is considered experimental and given a weight locked at 1.
[ SmokeDetector | MS ] Url in title, blacklisted website in body, pattern-matching product name in body, pattern-matching product name in title, pattern-matching website in body, +9 more (599): healthlinesplus.com/natural-burn-keto/ by flexuline on android.SE
@Stormblessed Probably because medium.com is not a spam site, but allos hosting of third party content that is a mix of spam and not spam. Also see the search research on MS for medium.com metasmoke.erwaysoftware.com/…
@Stormblessed For basically the reason that @Ferrybig said. The domain itself isn't a spam site, but it is commonly used by spammers to host a spam post, due to the ability to host things there with little over-site, or repercussions/responsibility. The number of posts detected containing medium.com is 534, of which 385 have been marked as TP. That equates to 72.1% TP, which is a reasonable indicator that the post may be spam.
@Stormblessed Because there was a large string of spam that was posted by someone with that name two to three months ago. In a case like the above post, it's reasonable to give fpu feedback. Doing so puts the user on the username whitelist, preventing that user from being detected by reasons looking at their username.
@Stormblessed Because it was chosen that metasmoke should not facilitate using mod-powers in that way.
@Makyen More explicitly, metasmoke doesn't permit either automatic or manual flags (e.g. through the MS API) to be raised by moderators on sites on which that user is a moderator.
@Stormblessed Of course. What MS and SD do in no way restricts what moderators can do on the sites they moderate. The only thing that MS and SD do that affects what users can do on any site is that MS may consume up to the maximum number of flags the user has specified on the sites they have specified.
However, we do have a policy in place (i.e. not a technological restriction) that people who are watching SD reports should not comment on posts which SD brought to their attention, unless you are familiar with the site through other interaction (e.g. have asked/answered on the site; etc.).
@Stormblessed The choice was made to not permit moderator flags to pass through the code which MS uses to raise flags. I was not involved in the project when that choice was made. I'd expect that it was, at least in part, to limit the possibility of the project causing harm. It may also have to do with wanting moderators to use the normal SE interface (i.e. see the extra information available to moderators on their SE site), but I'm not a moderator, so I don't know how much that matters.
As to our "special way of flagging", there are two ways that I'm thinking of off the top of my head to use MS to manually cast a flag: there's a "Spam flag" button on MS post pages; the FIRE userscript raises flags through the MS API. It's consistent for the flags cast from MS pages to be restricted. As to FIRE, it's something that could be added as an option, but the primary/original author declines to have that as part of the functionality.
@Makyen That pattern looks like it's already caught by Mostly punctuation marks in body and Mostly punctuation marks in answer; append -force if you really want to do that.
How can I check what roles I have and what they mean? I tried to visit a MS page and was told I don't have access (which is fine) and that I have the "Reviewer" and "Flagger" roles. Where can I find that out and where can I look up what those mean?
@Makyen That pattern looks like it's already caught by Blacklisted website in body and Blacklisted website in answer; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Mostly punctuation marks in body and Mostly punctuation marks in answer; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Mostly punctuation marks in body and Mostly punctuation marks in answer; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body; append -force if you really want to do that.
@Makyen That pattern looks like it's already caught by Mostly punctuation marks in body, Potentially bad keyword in body, Mostly punctuation marks in answer, and Potentially bad keyword in answer; append -force if you really want to do that.
@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.
