« first day (1854 days earlier)      last day (185 days later) » 
00:00 - 10:0010:00 - 00:00

12:09 AM
[ SmokeDetector | MS ] Bad keyword in answer, potentially bad keyword in answer (87): Ketogenic diet , side effects and precautions by Hollow on fitness.SE
fp- feedback received
 
1:05 AM
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Learning to write a compiler ✏️ by a deleted user on stackoverflow.com
fp- feedback received
 
@SmokeDetector ignore-
 
fp- feedback received
 
1:25 AM
[ metasmoke ] ci/circleci: rails-tests failure on f755a1f: Your tests failed on CircleCI
 
1:41 AM
[ SmokeDetector | MS ] Bad keyword with email in body (65): Tic Tac Toe game crash Java ✏️ by Openwell on stackoverflow.com
 
Rob
With the exception our websites https://medicalsciences.stackexchange.com/ and https://chemistry.stackexchange.com/ could we add more rejection weight to the word keto* expect for the word ketoconazole and possibly a very short list of additional words starting with "keto"? --- There's extremely few words starting with those four letters.
 
rejection weight?
 
Rob
The number in the parentheses (the score) is formed by totalling filters?
 
yeah
each reason has a weight
which is just the percentage of TPs that reason catches rounded to a whole number
sum those together and you have the total weight for the post
 
Rob
For example, this, above mentions "ketone", and only scores a 2, but it's on Chemistry.SE - so that's OK. BUT mention that elsewhere and it's probably the diet, pills for sale. We need more weight for keto* mentioned almost everywhere else.
 
1:51 AM
[ SmokeDetector | MS ] Potentially problematic ns configuration in body (1): Xcode10.1 RestKit error _$s7RestKit11DynamicKeysVN by Shin on stackoverflow.com
 
that's what blacklisting and other reason-improving operations are there to do
 
Rob
Just a sec., finding low scored example ...
 
blacklisting a term takes it from 1 weight (using "potentially bad keyword", which is pinned to 1) to the high 90s (using the various blacklist reasons)
if there's a pattern, we'll also edit an existing or create a new check to hit it with another reason as well
 
[ SmokeDetector | MS ] Potentially bad keyword in body (1): pppd (VPN Server) Works in windows, but not in routers? by David Martins on networkengineering.SE
 
Rob
This scored under 200, twice that score would have been better.
I've seen others under 400. It's just that the word seems to be a fad, and usage without a doctor's advice is a poor idea. We have little need for keto-questions. IMO.
If you don't like the idea, and everything is OK as-is that's OK too. I'm just asking/suggesting.
Back in 1/2 hour.
 
2:02 AM
[ SmokeDetector | MS ] Potentially bad keyword in body (1): Reading data from UMFT200XD or UMFT201XB by JeffreyH on electronics.SE (@TetsuyaYamamoto)
 
@Rob It's... it's not that I don't like it, it's that it's not possible how you're suggesting it
There isn't a convenient button to press that says "give this reason more weight"
it's a calculated value, deliberately so that it can't be tampered with
so to increase it, we need to get the low-score posts to be caught by more reasons or make existing reasons more accurate
 
!!/blacklist-keyword keto\W?ultimate\W?diet
 
blacklisting is a cheap way to do that, but there's a point where we need to work with other reasons too
!!/test keto ultimate diet
 
> Bad keyword in body, bad keyword in title, bad keyword in username, potentially bad keyword in body, potentially bad keyword in title, potentially bad keyword in username
----------
Title - Position 0-18: keto ultimate diet
Title - Position 0-4: keto
Username - Position 0-18: keto ultimate diet
Username - Position 0-4: keto
Body - Position 0-18: keto ultimate diet
Body - Position 0-4: keto
[ SmokeDetector ] FindSpam module reloaded at rev bc428c2 (𝐢𝐁𝐮𝐠 ♦: catch "ultimate") (running on iBug/EC2)
 
2:10 AM
!!/test keto ultimate diet
 
> Bad keyword in body, bad keyword in title, bad keyword in username, pattern-matching product name in body, pattern-matching product name in title, potentially bad keyword in body, potentially bad keyword in title, potentially bad keyword in username
----------
Title - Pattern-matching product name: Position 0-18: keto ultimate diet
Title - Position 0-18: keto ultimate diet
Title - Position 0-4: keto
Username - Position 0-18: keto ultimate diet
Username - Position 0-4: keto
Body - Pattern-matching product name: Position 0-18: keto ultimate diet
 
now there should be enough weight :)
 
fp- feedback received on Tic Tac Toe game crash Java [MS]
 
!!/blacklist-keyword keto\W*\w*\W*diet
maybe try if this generates FPs
 
2:13 AM
I'm fairly confident it won't generate too many
 
@iBug blacklisting all of them is probably not a great idea...
 
@ArtOfCode Moved to "pattern-matching product name"
 
that'll do :)
 
CI on e8ada73 succeeded.
 
2:22 AM
!!/pull
 
[ SmokeDetector ] FindSpam module reloaded at rev e8ada73 (𝐢𝐁𝐮𝐠 ♦: flake8) (running on iBug/EC2)
 
Rob
Thanks for looking into that. If I spot FPs I'll be certain to mention that.
 
!!/test keto asdfghjkl diet
 
> Pattern-matching product name in body, pattern-matching product name in title, potentially bad keyword in body, potentially bad keyword in title, potentially bad keyword in username
----------
Title - Position 0-4: keto
Title - Position 0-19: keto asdfghjkl diet
Username - Position 0-4: keto
Body - Position 0-4: keto
Body - Position 0-19: keto asdfghjkl diet
 
Rob
Would it be a bit better if at the bottom of the answer it said: That would score: XXX ? (or is it expected to know what that adds up to).
!!/test Yesterday my mother's friend's neighbor Tony asked me to babysit his daughter Mary's dog Keto.
 
2:30 AM
> Potentially bad keyword in body, potentially bad keyword in title, potentially bad keyword in username
----------
Title - Position 89-93: Keto
Username - Position 89-93: Keto
Body - Position 89-93: Keto
 
Rob
If that scores ~300 that's OK - or not?
 
@Rob I think you have a misconception about what weight does. Weight, beyond enough to trigger autoflagging, particularly 4 autoflags, makes little or no difference to the post. The reason to have more detections for other things that happen to be in a post is in order to detect posts where those other things are the one ones there, not, necessarily, to give it more weight.
At the low end of weight and reasons, a post being autoflagged depends on people having set their autoflagging conditions such that a post qualifies. It's required that the each condition puts the post in the 99.75% is spam range, and a user must maintain that % for the combination of all their autoflagging conditions.
 
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): What does “my six ounce” mean? ✏️ by jdoe on english.SE
 
Rob
OK, I should study the code. Thanks for adding the filters shown above.
 
There are really only 4 different bins into which posts are placed: 1) detected with only "potentially bad" (i.e. experimental) reasons and no other reasons. These are only reported into Charcoal HQ, Charcoal Test, and The Fire Department. 2) Detected with 1+ non-experimental reasons, but not autoflagged. These are reported in to all applicable chat rooms.
3) Detected with enough weight/reasons to be autoflagged by at least 1 person, but below 99.9% chance of being spam, based on historical data for weight, reasons, user rep. Up to 3 autoflaggs will be placed on these, if there are enough people who have autoflagging conditions that qualify. 4) > 99.9% chance of being spam. Up to 4 autoflaggs are placed, assuming there are enough people with autoflagging conditions that qualify.
The default autoflagging condition is weight=280, IIRC. So, most people will have that weight as their point where they will autoflag.
At the lower end, it's going to depend significantly. However, there are some people that have set an autoflagging condition at 3 reasons, no minimum weight. That means that if there are 3 reasons, even three experimental reasons, then those people's autoflagging conditions will be satisfied and the post will get autoflaggs.
 
2:40 AM
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Lid Keys on ThinkPad X41t doesn't work when device is 'transformed' by SuperPrower on unix.SE
 
Rob
So (in reference to your last paragraph) those people risk a false flag - which is preferable to avoid.
 
@Makyen I've been gently debating a minimum weight as well as minimum accuracy, just to avoid this. I haven't run the data but I suspect a lot of our FPs come from this.
 
Rob
@SmokeDetector fp-
@SmokeDetector fp-
 
So, basically, there really isn't any functional benefit (currently) for a post to have a weight > 280. The primary benefit for of additional detections is to have those detections active if the spammer chooses to change their spam sufficiently such that the original detections don't catch the new post. And, frankly, there are enough people with custom autoflag conditions such that it's very rare for a post with a weight above ~200, or so, to not have enough people to max the autoflaggs.
@ArtOfCode This is something I've grown more concerned about over time. Personally, I've chosen to specify a weight for my 3 reason conditions (except for some that I enable during spam rushes, or that I've kept active on specific sites due to specific activity). I think that some (low) minimum would be reasonable, to prevent the 3 experimental reason issue. We'd probably want to look at the data to see what makes sense.
 
Rob
Weights of ~300-~400 are often still there for me to read and make a decision. When the number in parentheses is over ~600 it's very difficult to read, think, and flag without the rug being pulled out from under you. -- So weights between 280 and 500 have a purpose, you don't have to be speedy Gonzales.
 
2:49 AM
@Makyen aye, likewise. My 3-reason has a weight on it as well.
 
Rob
In the search bar (upper right corner) I simply typed keto to find that example that scored under 200.
 
Or perhaps we could exclude experimental reasons from counting towards reason count
 
@Rob Yes, but that example got 4 autoflaggs.
 
Rob
Divine experiment reasons by two to calculate the weight, but list them all?
 
@Rob experimental reasons all have their weight pinned to 1
 
2:53 AM
@ArtOfCode I'd prefer not to do that. At least in the past, we've found it useful to (temporarily) add both additional blacklist and watchlist detections to get specific spam waves to trigger autoflagging.
 
Rob
@Makyen - This is the one I was referring to, that scored under 200: metasmoke.erwaysoftware.com/post/148090
 
@Makyen aye, point
 
@Rob Yes, that's the one. The link I provided was to a record of the 4 autoflags that were placed on it.
 
Which brings up the question of whether a min weight also defeats that
 
At least for what we've done in the past, it wouldn't, as long as it was a reasonably low minimum.
 
Rob
2:56 AM
It got 5 hits but they added up to less than 200. No human would think that text should appear on SuperUser.SE.
 
Which also raises the question of a spam wave tool. I've wondered if it would be useful to allow admins to define a search for which, for 24 hours, every matching post gets autoflagged regardless of conditions
Suitable safeguards, of course
Time limited, minimum % TP, etc
But if, say, AU suddenly gets an influx of spam about ProductX that's too big for the community to handle... admin enables a spam wave for it, every post matching body ~= "ProductX", site=AU gets flagged for 24 hours.
 
That might be useful. Obviously, it could be abused/manipulated, but we're not designing to prevent intentional abuse, just accidental. Admittedly, once we were aware of the last wave (as far as I know) we got the posts to be consistently above autoflagging in a reasonable amount of time, just by adding black/watch-list entries.
afk
 
Rob
An expiry time field for the rule, with a message to the rule maker that their rule will expire in 10% * Duration - Example: rule expires in 20 hours gives 2 hours notice, and set minimum notification to half hour regardless, unless duration is less than one hour. ?
 
Aye, we have ways of hacking the system to be reasonably sure that the wave gets flagged - but I can't help feeling that a tool meant for the purpose would be better than abusing other mechanisms
 
[ SmokeDetector | MS ] Potentially bad asn for hostname in body (1): generate/generatetoaddress what are this two rpc methods used for? by huahuayu on bitcoin.SE
 
3:03 AM
@Rob Aye, I like that. Let the admin set the expiry time, max 24 hours but renewable, and drop a message in here 10% before it expires. Nice.
 
[ SmokeDetector | MS ] Potentially bad keyword in username (1): Open Ethereum mining pool by Krispycream on ethereum.SE
[ SmokeDetector | MS ] Potentially bad ns for domain in body (1): A complicated use of NDSolve ✏️ by JaJ on mathematica.SE
 
Rob
There must be "Trend Analysis Library" that sees the wave coming and adjusts, much like YouTube had trending videos (but their algorithm isn't very good).
 
Need to remind myself to type that up as a GH issue in the morning
@Rob spam waves aren't predictable
 
I agree that a tool that does that could have use, but it starts to get into the range of sockpuppetry, which we've been strongly trying to avoid.
 
fp- feedback received on Open Ethereum mining pool [MS]
 
3:06 AM
we had this problem once with baba­support.org, and we ended up stacking 3 reasons onto the domain
 
@Makyen it does, which has been my major concern about it. The initial idea was an autoflag button on every post for admins, which... I shot that down fairly quickly :)
 
@SmokeDetector ignore-
 
I feel like defining a search for posts rather than specifying individual posts is far enough removed that I'm comfortable it's not-socky enough
 
fp- feedback received
tpu- feedback received
 
3:08 AM
These are all things we should discuss, which is why I want to throw it into a GH issue
I'll do that tomorrow unless anyone else wants to kick it off sooner
 
Rob
@ArtOfCode Yes, the waves (chart) are very spikey.
 
Discussing it in a GH issue sounds like a good idea. I think it's something that needs a more consideration, and a broader base of input than we get by just bringing it up in chat.
 
Rob
 
@Rob heh, that's not even waves, that's just normal daily fluctuations. Waves look huge on that chart
 
Rob
We need an error chart that reads close to zero (maybe Log scaling so we can see the small values and squish the spikes). Is there a free cloud service that does 'AI Pattern Analysis' for spam, we could get a second opinion, test another method, and chart that. Maybe let it add It's 2¢ if it turns out to be a reliable 2nd set of eyes.
 
3:17 AM
Heh
AI and machine learning on spam is the one problem that we have never solved
 
Rob
Back in 2 minutes, Googling.
 
@TetsuyaYamamoto FYI: MS has made quite a few attempts to flag using your account on sites where you don't have an account. This isn't something that's an actual problem, per se. But, it indicates that things may not be as you are intending them. I just happened to be looking at the failed flag logs and noticed that many of them were yours.
If you want accounts on all sites, you can use this SEDE query to find all the sites on which you don't have an account.
 
Our best have had many attempts at it and... nada. Nothing's come close to what we've built with detailed heuristics and regex.
That's the problem. Creating a classifier isn't hard. Making it accurate enough to be worthwhile sustaining it is.
 
@SmokeDetector ignore-
 
[ SmokeDetector | MS ] Few unique characters in answer, repeating characters in answer (168): Forgot Gmail password, how do I get back into my account? by Jayonda Nelson on webapps.SE
tpu- feedback received
fp- feedback received on Reading data from UMFT200XD or UMFT201XB [MS]
[ SmokeDetector | MS ] Potentially problematic ns configuration in answer (1): Is "to worry about their children being lead astray" correct? by M. Bal on ell.SE
fp- feedback received
 
Rob
3:27 AM
@ArtOfCode Suggestion 1: codeproject.com/Articles/1229597/… - off to look again.
 
@ArtOfCode Did I get this one yet?
 
Anyone who has the time and inclination is more than welcome to give it a go
 
 => [["10.3.9-MariaDB-1:10.3.9+maria~trusty"]]
 
@Undo not yet, but that particular question is answered :) there's also a manual deploy waiting for you when you have time
@Undo sweet, confirmed theory
 
@Makyen And then get a ton of hats for it.
 
That'll be proprietary
 
Rob
True, but they might offer it on the cloud, and harvest our input (which we give away anyways CC) in return for an opinion on what is fed to them. We don'tneed t code, just the result.
brb, making coffee
 
[ SmokeDetector | MS ] Potentially bad ns for domain in body (1): The meaning of "thanks chat" by saitouena on english.SE
 
@Catija :-)
 
all the same hat... but lots of them. :D I got 66... I'm sure someone got more than that, though.
 
3:53 AM
@ArtOfCode Just config/blazer.yml?
 
[ SmokeDetector | MS ] Potentially bad keyword in body (1): pppd (VPN Server) Works in windows, but not in routers? by David Martins on serverfault.com
fp- feedback received
 
Rob
A nice collection @Catija.
 
[ metasmoke-deploy ] deploy started by Undo
 
@Rob You mean the 66 or the general collection? Thanks, either way. I've been too busy to actually put much effort in but I got three from my knitting design alone, so that helps.
 
@ArtOfCode Done, looks like it's not-broken
 
Rob
4:08 AM
The general collection. Joe Friend, Jon Skeet have zero, Shog9 has 5, a lot of high Rep (and active users) have fewer than half a dozen.
Makes me feel better about my 5, one of which are the 'giveaway glasses' and the Doughnut Hole I'm wearing now - not a hat but better than the default box with one flap opened.
 
Ah. :) Yeah. They're kinda hard this year... but we're also really busy. I haven't actually had any time to try, what with a baby and working half-time.
 
Rob
I scored 200 Rep yesterday for knitting, along with 3 badges; and Nog put up a bounty to reward my 'next level knitting'.
 
... well, I spent a stupid amount of time on the knitting...
Is yours the GIF?
 
Rob
Me too, two hours before and another three today searching for a way to beat my last effort.
Yes.
You need to stand back. You can't be too close to the screen, that makes it smoother.
 
You did the whole thing by hand?
 
Rob
4:19 AM
I took the video, found 0.5 seconds that I could loop. Extracted .pngs, uploaded them, downloaded them, and then combined the frames into an animation - but it was 'by hand?' on my phone.
 
Ah. That's still cool... and a fun idea.
 
Rob
It's a famous Bugs Bunny classic, I'm not the original animator.
 
Sure, I recognize that.
I'm not a programmer, so I have no clue what it's possible to automate. I did my logos with a touchpad. :D
 
Rob
If I had a computer it would be much better. I must wait, they've literally just taped out my CPU and engineering samples are being tested. One motherboard for PCIe 4 has been leaked, probably the one used for tested - so buying a computer is 6-7 months away - like they say: It's gonna be great.
 
Can we put ChatExchange on PyPI? Maybe under a different name (e.g. ChatExchange-Charcoal) so it doesn't conflict with Manish's
 
Rob
4:25 AM
Working on my phone means holding a magnifying glass and typing one finger at a time.
 
Maybe @ArtOfCode can transfer his fork to our GH organization
 
Rob
Thanks for the chat @Catija, I'm off to make a coffee and poison myself with a smoke.
 
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, potentially bad ns for domain in body, +2 more (496): Power Boost works naturally to optimize your sexual health by jeanksuh on arduino.SE
tpu- feedback received
 
4:43 AM
@Rob Bye! Thanks! Good luck with your new computer, whenever it arrives. :D
 
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Is there a way to pretty format or beautify NSIS script source code? by Ed Sheehan on stackoverflow.com
fp- feedback received
 
sd f
 
tpu- feedback received
[ SmokeDetector | MS ] Offensive answer detected, potentially bad keyword in answer, toxic answer detected (161): Multiple breaches of employer confidentiality by Gak Head on workplace.SE
 
sd rude
 
4:59 AM
@SmokeDetector That got deleted with only three flags... am I missing something?
 
[ charcoal-se.github.io ] continuous-integration/travis-ci/push success on e3977f7: The Travis CI build passed
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Enabling MacType for Chrome 52+? by Сергей Шевчук on superuser.com
 
Rob
@Catija - Grossly offensive. The letters "MS" provide an answer that you don't want to see.
 
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword in username, bad pattern in url body, link at end of body, +3 more (681): Ace CBD Oil Trial Information? by acecbdoil on askubuntu.com
 
@Catija A few sites are 3-flag nuke sites
 
@K.Dᴀᴠɪs AH... that's it. I guess TWP is one of them.
@Rob Oh, I can see it. It's not even hidden behind a warning for me the way it is for most people. :)
 
5:06 AM
Ahh that's right, aren't you in TX or something?
 
I am... what does that have to do with it?
 
tpu- feedback received
 
For some reason I was thinking of a completely unrelated privacy law... meh nvm
Short answer, cause I am a moron xD
 
? Now I'm confused. Were you talking about how I said that it's not behind a warning?
 
Rob
There was no warning for me either, we must accept that we see things; so we can decide that others ought not see them.
 
5:12 AM
10k users see a warning on posts deleted as r/a... if it's not deleted, you don't see the warning... but mods don't see it ever.
 
Rob
I'd say to someone else file at Meta but in your case just ask directly - if you want a warning it seems fair - up to your co-workers.
 
We were actually talking about it a week or two ago.... both in terms of staff and mods.
 
@SmokeDetector The link in this answer is to some site/blog in Japanese. The download links on that site appear to be to non-official versions of the software. However, I'm completely unfamiliar with this software, so it could be that this is some secondary or tertiary distribution, but it doesn't appear so from information on the official sites.
 
@Makyen I see this on the main site though
 
5:23 AM
@paper1111 OK. Thanks. That didn't show up in my primary browser (higher security settings disabled it).
 
fp- feedback received on Enabling MacType for Chrome 52+? [MS]
 
Rob
That filter: Position 0-12: <p><a href=" seems to catch a few innocent links from time to time. Example: h:/site/read.pdf is a great article on fixing what you wrote about, page 2: says: "short quote from page 2". That nails you for starting with an URL and saying great article.
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +6 more (885): nutritionless.com/curafen/ by griffith on drupal.SE
tpu- feedback received
[ SmokeDetector | MS ] Url-only title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +8 more (989): nutritionless.com/curafen/ by lavernearlon on askubuntu.com
tpu- feedback received
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword in username, blacklisted website in body, pattern-matching website in body, +4 more (487): Does It Really Perform ? Teal Farms Keto by Teal Farms Keto on askubuntu.com
tpu- feedback received
[ SmokeDetector | MS ] Url in title, bad pattern in url body, blacklisted website in body, blacklisted website in title, body starts with title and ends in url, +4 more (783): www.smore.com/39268-one-beast-test-reviews by sandraeyes on drupal.SE
[ SmokeDetector | MS ] Url in title, bad keyword in body, body starts with title and ends in url, potentially bad ns for domain in body, potentially bad ns for domain in title (292): 99newsdaily.com/ by spilorcio on wordpress.SE
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad keyword in username (1): Network automation module error on GNS3 by Doncream on stackoverflow.com
 
5:40 AM
@Rob Yes, it catches a significant number of FP. The feeling that it catches a lot of FP is exacerbated by the fact that many of those same posts are detected multiple times (for some, it's many times). As of the point I last updated the detection, the updated version had, from the time the original version was added, 96 TP, 43 FP, and 22 NAA, which is 60% TP. The original version of the watch was 98 TP, 132 FP and 31 NAA at that point (38% TP). The update reduced the non-TP by 60%.
 
naa- feedback received
 
Run against all the data in MS (the data is quite biased towards TP), it is 14,363 TP, 173 FP and 85 NAA.
There is also a second detection, which is youtube.com and youtu.be only, which has a significantly higher FP rate, but with also was the only detection that caught a few TP in the time it was up, so I left it as a separate watch. At this time, it will appear to be the same detection from what's reported in the "why" information, but it's not, currently, intended to migrate into a separate detection.
 
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, potentially bad ns for domain in body, +2 more (564): Regulates the biological functioning of males by jeanksuh on arduino.SE
tpu- feedback received
[ SmokeDetector ] SmokeDetector started at rev e8ada73 (𝐢𝐁𝐮𝐠 ♦: flake8) (running on iBug/EC2, Python 3.6.7)
Restart: API quota is 15089.
[ SmokeDetector | MS ] Potentially bad keyword in body (1): How can I process Hot Pepper without killing the spice (capsaicin)? by Irodoku on cooking.SE
fp- feedback received
[ SmokeDetector | MS ] Url in title, bad keyword in body, blacklisted website in body, blacklisted website in title, pattern-matching product name in body, +6 more (692): supplementfordiet.com/super-keto-australia/ by user130710 on graphicdesign.SE
tpu- feedback received
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, pattern-matching website in body, potentially bad keyword in body, +1 more (393): The additives of Privy Farms Keto are biotic by aneerjudd on rpg.SE
tpu- feedback received
fp- feedback received on Google play store error code 910 [MS]
fp- feedback received on My IK hand doesn't work right [MS]
fp- feedback received on The meaning of "thanks chat" [MS]
 
Rob
6:10 AM
Seems like a good (enough) answer, user for two+ years. Really the only thing wrong is the reassurance that it should work and the "Have a nice day!" is chatty (but that ought to have been edited out long ago) - we only revisit it because the Community bumped it to get more/newer answers (which is helpful) and to bring it to the attention of those more knowledgeable about it if it remains the best answer - which might have been confirmed by paper1111 - so it 'site business' not our
business.
afk
 
@SmokeDetector It may not be obvious on it's face, but this is spam. The user has a history of spamming for the site they have linked here. Most of their detected posts are short blurbs that look like they might be coming close to maybe answering, but don't actually do so.
 
fp- feedback received
 
@Rob If you make comments like this an actual reply to a particular post, it helps other people understand what you're talking about, and includes what you write as a comment on the MS post, so others reviewing the post can see it.
 
6:28 AM
[ SmokeDetector | MS ] Url in title, blacklisted website in body, blacklisted website in title, potentially bad ns for domain in body, potentially bad ns for domain in title (297): Eretron Aktiv : www.dealhitch.com/eretron-aktiv/ by eretronaktiv on astronomy.SE
tpu- feedback received
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, body starts with title and ends in url, link at end of body, potentially bad ns for domain in body (393): Endeavor to incorporate action today day by day propensity by user973240 on superuser.com
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad asn for hostname in body (1): Not able to open a website by using selenium webdriver with chrome in java program by Rakesh on stackoverflow.com
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching product name in title, pattern-matching website in body, potentially bad keyword in body, +1 more (397): to Australia (AU) Where to Buy? What's more, Shark Tank Reviews by ramjasv on meta.SE
tpu- feedback received
[ SmokeDetector | MS ] Blacklisted website in body, blacklisted website in title, link at end of body, link following arrow in body, link following arrow in title, +2 more (498): user @> amazonhealthstore.com/super-keto-australia/ by HuiopLeshs on superuser.com
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad keyword in body (1): Alat Musik Bali Tradisional List Daftar & Penjelasanya by Aris Rahman on workplace.SE
tpu- feedback received
[ SmokeDetector | MS ] Blacklisted website in answer (92): Convert EML to MSG - Java by Shane will on stackoverflow.com
 
Rob
6:51 AM
@Makyen Understood. I was in the middle of something when your reply came and my reply would have been consecutive were it not for the intervention of Smokey's text. It's a bad habit I'll remember to break. I have to zoom in all the way and side-scroll to confirm that I have the correct reply arrow, then zoom back out to access the reply window. I was running ten minutes late by the time I got to typing "afk". I'll be certain not to trade expediency for threading convenience. Sometimes
being online here makes me late for things. I'm fighting to avoid that too.
 
fp- feedback received
tpu- feedback received on Convert EML to MSG - Java [MS]
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, pattern-matching product name in title, pattern-matching website in body, +1 more (492): AmbroSina Skin Cream Reviews: There is no side-effect of this product by Walkeroscar on webapps.SE
 
@SmokeDetector tpu-
 
[ SmokeDetector | MS ] Url in title, blacklisted website in body, blacklisted website in title, pattern-matching website in body, pattern-matching website in title, +4 more (496): supplementfordiet.com/super-keto-australia/ by aplozjmajklss0 on graphicdesign.SE
 
7:06 AM
@SmokeDetector tpu-
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, pattern-matching website in body, pattern-matching website in title, +2 more (490): Loriax Cream AVIS Today everybody is sorted by their looks, a gorgeous www.supplement4fitness.org/loriax-cream-avis/ by jhbmji on graphicdesign.SE
 
@SmokeDetector tpu-
 
[ SmokeDetector | MS ] Blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body, potentially bad ns for domain in body, +1 more (495): Boosts the endurance and sexual surge by jeanksuh on askubuntu.com
 
@SmokeDetector tpu-
 
fp feedback received
[ SmokeDetector | MS ] Potentially bad keyword in body (1): Best wedding photographer in noida/gurgaon by Thewedding file on meta.SE
[ 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 ns for domain in body, +2 more (492): It stimulates strip room regeneration to rejuvenate your scorched by mariarojas on android.SE
 
7:32 AM
@SmokeDetector tpu-
@SmokeDetector tpu-
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +7 more (790): Order Now:-market4supplement.com/forskolin-keto-cycle/ by gbhszrttgnbserk on drupal.SE
 
@SmokeDetector tpu-
 
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Good wind-screen for a bicycle? ✏️ by Morgan on bicycles.SE
 
@Rob I'm sorry to hear that. Stuff in Real Life usually is more important. As to having to zoom in and out, there is a mobile view. I haven't really used it, but it might be something that can help with how you use chat.
 
fp- feedback received
[ SmokeDetector | MS ] Bad keyword in body (97): How thinking about sex triggers sexual drive? by Amir Khan on psychology.SE
fp- feedback received
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, bad pattern in url body, body starts with title and ends in url, +8 more (883): excelgarciniafacts.net/alpha-titan-testo-reviews/ by rayohem on wordpress.SE
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Error: L6218E: Undefined symbol main (referred from __rtentry2.o) by Timur on stackoverflow.com
[ SmokeDetector | MS ] Bad keyword in link text in answer, potentially bad keyword in answer (81): What are the possible causes of vibrating steering wheel? by user44089 on mechanics.SE
tpu- feedback received
 
8:11 AM
!!/watch- cartrends\.in
 
[ SmokeDetector | MS ] Potentially problematic ns configuration in body (1): FATAL ERROR: attr/error_context.h does not exist by sivarajang on stackoverflow.com
[ SmokeDetector | MS ] Link at end of answer, username similar to website in answer (112): Google Plus Share Button is not sharing on Business Page by iamhja on stackoverflow.com
tpu- feedback received
 
!!/watch- iamhja\.com
 
[ SmokeDetector | MS ] Manually reported answer (94): Is there a tool or a way to convert a c program to c# by Full Time Freelancer on stackoverflow.com
tp feedback received
 
8:40 AM
[ SmokeDetector | MS ] Blacklisted website in body, pattern-matching website in body, potentially bad ns for domain in body, potentially bad keyword in body, potentially bad keyword in title (199): Super Keto Australia by TaauRfff on superuser.com
tpu- feedback received
tpu- feedback received
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted username, blacklisted website in body, blacklisted website in title, +4 more (685): UroGenX :pickwellnessdeal.com/urogenx/ by jameasldson on puzzling.SE (@TetsuyaYamamoto)
tpu- feedback received
CI on 4a83564 succeeded.
CI on 3ff7ccb succeeded.
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +5 more (788): amazonhealthstore.com/super-keto-diet/ by LeosBullock on graphicdesign.SE
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad keyword in body, potentially bad keyword in title (2): Oxidation of hexan-3-one (unsymmetrical ketone) ✏️ by Manmeet on chemistry.SE
fp- feedback received
[ SmokeDetector | MS ] Bad keyword in body (97): We could not find out by user156015 on wordpress.SE
tpu- feedback received
 
8:59 AM
Why is there no link in the MS version? ^^
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, blacklisted website in body, blacklisted website in title, body starts with title and ends in url, +3 more (590): hbmbzone.com/urogen-x/ by Katujizzz on graphicdesign.SE
tpu- feedback received
[ SmokeDetector | MS ] Potentially bad keyword in body (1): Elementary OS hangs when using libreoffice by zjv on elementaryos.SE
[ SmokeDetector | MS ] Link at end of body, potentially problematic ns configuration in body (100): What improvements should i make on my website? by Shefali Singh on superuser.com
tpu- feedback received
 
Proposal: a spam-wave tool - food for discussion
4
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, blacklisted website in body, blacklisted website in title, potentially bad keyword in body, +1 more (394): amazonhealthstore.com/super-keto-australia/ by Mariaudwig on meta.SE
tpu- feedback received
 
cc @Rob @Makyen
 
[ SmokeDetector | MS ] Url in title, blacklisted website in body, blacklisted website in title, body starts with title and ends in url, link at end of body, +3 more (786): healthytalkz.com/igr-plus/ by user438490 on meta.SE
[ SmokeDetector | MS ] Bad keyword with email in body, potentially bad keyword in body (66): Get an official IELTS, TOEFL, All English Language certificates by berelyce02 on english.SE
tpu- feedback received
 
9:16 AM
@SmokeDetector tpu-
 
!!/watch registraboard@protonmail\.com
 
tpu- feedback received
 
!!/watch aptoidedownload\.in
 
[ SmokeDetector | MS ] Potentially bad ns for domain in body (1): Describe in detail the furniture, artwork, and feature by Dissertation Help Now on math.SE
tpu- feedback received
 
!!/watch allassignmenthelp\.org
 
!!/watch valentinesdayromantic\.com
 
[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +6 more (787): mummydiet.org/ketolean/ by fazozuhe on wordpress.SE
tpu- feedback received
 
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in url, link at end of body, +3 more (589): Vital progenix australia by Nicole James on meta.SE
@PeterJ Of course
tp feedback received
 
9:48 AM
NobodyNada/Raspberry Pi received failover signal.
Restart: API quota is 19996.
[ SmokeDetector | MS ] Bad keyword in link text in body (66): Why Dr. Rahul Gupta Is the Best Brain Tumor Surgeon in Delhi? by Dr Rahul Gupta Best Surgeon on medicalsciences.SE
tpu- feedback received
 
!!/watch rgbrainspinesurgeon\.com
 
@Shree Added rgbrainspinesurgeon\.com to watchlist
Restart: API quota is 13803.
@Shree Already watched on line 7248 of watched_keywords.txt
[ SmokeDetector ] FindSpam module reloaded at rev 3127760 (SmokeDetector: Auto watch of rgbrainspinesurgeon\.com by Shree) (running on NobodyNada/Raspberry Pi)
 
!!/location
 
@Shree NobodyNada/Raspberry Pi
@Shree iBug/EC2
 
!!/status
 
9:53 AM
@Shree Running since 17:43:55 UTC (16 hours, 9 minutes, 6 seconds)
@Shree Running since 09:50:17 UTC (2 minutes, 45 seconds)
 
!!/standby NobodyNada/Raspberry Pi
 
[ SmokeDetector ] SmokeDetector started in standby mode at rev 3127760 (SmokeDetector: Auto watch of rgbrainspinesurgeon\.com by Shree) (running on NobodyNada/Raspberry Pi)
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted username, blacklisted website in body, pattern-matching product name in body, +5 more (685): Regal Keto Diet Reviews best health care products by ner juriyam on webapps.SE
tpu- feedback received
 
00:00 - 10:0010:00 - 00:00

« first day (1854 days earlier)      last day (185 days later) »