[ SmokeDetector | MS ] Url in title, body starts with title and ends in url, link at end of body, link following arrow in body, pattern-matching website in body, +2 more (688): healthytalkz.com/igr-plus/ by user432094 on meta.SE
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +3 more (589): getmagnetiques.com/genodrive/ by yigoza on wordpress.SE
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +7 more (888): noomdietplan.com/ by cwoy on android.SE
The first link in Denaton's answer to "Best Open Source Tools to Build an AR Application" is: unity3d.com/partners/… --- That means that they get paid for click-throughs. Could someone write something to catch that sort of thing. Thanks.
One of the problems we'll have I suppose is that unless you know the site, there's nothing to tell you why it's been caught. So you'll get people marking it as fp
@CalvT Feedback should be based, primarily, on the version detected by SD. IMO, this is tp. It's reasonable to have it be tp and not tpu due to the user editing to make it a real question/answer, but as detected by SD this is something that should be blocked, IMO.
[ SmokeDetector | MS ] Bad keyword in answer, link at end of answer, pattern-matching website in answer, potentially bad ns for domain in answer, potentially bad keyword in answer (228): Format XML in table row by MaribelADay on stackoverflow.com
np. It seemed like a good idea to detect those. That should work at least for the basic form. If they use something more complex in the URL (e.g. additional parameters prior to the _ga), we'd need to adapt the regex.
In my spare time I've been searching for a list of popular affiliate marketing links that you might consider adding, similar to the watched_.text and blacklist_.text files that currently exist. An abbreviated version of the most popular ones would catch the majority. A full list would be too big.
If I spot another I'll examine our webpage to see if they are "no follow" because if they aren't our search rankings will sink. That user has a three year history here so it wasn't obvious as spamming. Indeed my flag ended up being disputed (with 5 downvotes!?) but shortly after the entire thing was deleted and it's 404 to me.
A list like that, even if just added to watched keywords, sounds like a good thing to have. I know we've put a couple such affiliate codes in the watchlist, but having something that's generalized, at least for more popular specific sites' codes, would enable us to proactively catch those, rather than reactively.
@SmokeDetector Has disclosure; may be NAA, but the last paragraph appears to have the very basics. OTOH, it does appear to be primarily there to drive traffic to their blog for a "a step-by-step how-to".
@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.
@Makyen That pattern looks like it's already caught by Pattern-matching website in body and Pattern-matching website 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.
(1) experiment running on TSM's infra using newly-created tokens, (2) now-not-experiment migrated to my infra using metasmoke's existing tokens, (3) TSM invalidated the tokens he had
Not a "holy crap we lost stuff" thing, just good hygiene.
