Recovered from websocket._exceptions.WebSocketConnectionClosedException: Connection is already closed.

yeah, additional debugging that !!/errorlogs doesn't provide...

that one too ^^

ideally additional debugging that !!/errorlogs will then provide

tbh, if !!/errorlogs actually provided the debugging itself and not just the log, we would have won trillions of $$$...

(patent :P)

Recovered from BrokenPipeError: [Errno 32] Broken pipe

[ SmokeDetector | MS ] Potentially problematic ns configuration in body (27): Как сделать запрос на api сервер через java by Вадим Орловский on ru.stackoverflow.com

fp- by Erik the Outgolfer

@SmokeDetector Help request.

Recovered from BrokenPipeError: [Errno 32] Broken pipe

[ SmokeDetector | MS ] Potentially bad keyword in body (1): License error using Benchmark.NET + DevArt dotConnect for PostgreSQL by Dmitry Senin on stackoverflow.com

fp- by bertieb

[ SmokeDetector | MS ] Potentially bad ns for domain in answer, potentially bad keyword in answer (2): Recebendo Email Java by Wilson Filho on pt.stackoverflow.com

Recovered from BrokenPipeError: [Errno 32] Broken pipe

iBug/EC2 received failover signal.

Restart: API quota is 19998.

!!/report mathoverflow.net/questions/315233

[ SmokeDetector | MS ] Bad keyword in body, potentially bad keyword in body (98): Why is spiritual treatment (Roohani Elaj) necessary? by Roohani Elaj on mathoverflow.net

tpu- by Erik the Outgolfer

[ SmokeDetector | MS ] Potentially bad keyword in body (1): Is an email answer for an apartment rental legally binding? by german_oder on law.SE

fp- by Glorfindel

[ SmokeDetector | MS ] Bad keyword in answer (86): Is this true, the use of fish and milk at a time cause a disease named vitiligo? by MacKenzie McFadden on cooking.SE

fp- by Glorfindel

[ SmokeDetector | MS ] Potentially bad keyword in body (1): Which VPS supports registration and management using the Tor Browser? by tosx on webmasters.SE

[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +7 more (986): www.usafitnessguide.com/androdna-testo-boost/ by nawty102 on apple.SE

tpu- by Glorfindel

[ SmokeDetector | MS ] Bad keyword in link text in body (65): DMX Control - 50W COB LED by BaseThree on electronics.SE (@TetsuyaYamamoto)

fp- by Glorfindel

!!/report english.stackexchange.com/a/472743

[ SmokeDetector | MS ] Manually reported answer (94): Is there a collective name for pieces of metal that hold wood together? by asseater on english.SE

as seater

[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body, pattern-matching product name in body, pattern-matching website in body (492): anywhere beginners happen upon the best by Levushina on askubuntu.com

tpu- by suraj

[ SmokeDetector | MS ] Potentially bad asn for hostname in body (1): MegaMenu Tablist collapse on click aid required please ✏️ by HugsVille1e6 on stackoverflow.com

fp- by Filnor

[ SmokeDetector | MS ] Pattern-matching website in body (98): Compiling modules for cross-compiled arm64 kernel fails with Syntax error by Akito on raspberrypi.SE

fp- by Glorfindel

tpu- by Glorfindel on Is there a collective name for pieces of metal that hold wood together? [MS]

[ SmokeDetector | MS ] Potentially bad keyword in body (1): My girlfriend's dog gets hurt by Escanor on pets.SE

fp- by Glorfindel

> and if not the eggs

...autocucumber?

[ SmokeDetector | MS ] Potentially bad keyword in body, potentially bad keyword in title, toxic body detected (71): Is ejaculating without masturbating still haram for Muslims? by Asef Bodlox on islam.SE

[ SmokeDetector | MS ] Link at end of body (99): What is the work of Bekausa? by Beka USA on superuser.com

tpu- by Erik the Outgolfer

!!/watch- bekausa\.com

CI on 96d1cb6 succeeded. Message contains 'autopull', pulling...

[ SmokeDetector ] SmokeDetector started at rev 96d1cb6 (SmokeDetector: Auto watch of bekausa\.com by Erik the Outgolfer --autopull) (running on iBug/EC2, Python 3.6.6)

Restart: API quota is 19083.

fp- by ByteCommander on Is ejaculating without masturbating still haram for Muslims? [MS]

fp feedback on autoflagged post: Is ejaculating without masturbating still haram for Muslims? MS

@ByteCommander Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@quartata Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@CalvT Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

wha...

anyway, @quartata @WELZ you've got an autoflag on that fp above.

@SmokeDetector I figured this would happen one day... -_-

Your flag conditions were manually disabled by @SmokeDetector if its done by a bot is it done manually?

that's automatic

botually

Why does it disable all of my conditions and not just the problem ones.

no idea

also all of them show at least 99.84% accuracy in the dashboard...

many times the problem isn't caused by any one condition

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

Ok, ok calm down. I'm working on it.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

o_o

he doesn't like you...

@WELZ Your flag conditions can be set with more calmness if you click on @SmokeDetector's avatar and then the "hide posts" button

(fail...)

Lmao

My current conditions are 280/1/3

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

wait what

I've got 250/1/1 and 100/30/3 now.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

how can it complain on yours, if that's way higher than mine?

I even tried creating an entirely new condition -_-

actually, those numbers don't make much sense unless you also specify the sites (yeah, good luck), but, seriously?!

btw, I have 280/1/1 (*)

@EriktheOutgolfer all of them.

maybe you should ask Undo to give MS a good smack...

I think @Art set up this system, maybe he'll know.

fp- by Glorfindel on Which VPS supports registration and management using the Tor Browser? [MS]

@CalvT Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

@SmokeDetector It is over querido…

[ SmokeDetector | MS ] Bad keyword in answer, blacklisted username, toxic answer detected (260): Did Stan Lee finish his cameo shooting for the upcoming Avengers 4 movie? by Penis Man on scifi.SE

2 conditions, one at 99.84% and the other 99.88% (cc @ArtOfCode)

tpu- by Glorfindel

[ SmokeDetector | MS ] Potentially bad ns for domain in answer (1): ReactJS web font by TigranK on stackoverflow.com

@SmokeDetector Come on, respect the dead. :-(

inane whistling

@CalvT the combination of them might still be under 99.75%

How does that work?

like an OR applied between the conditions, I think

I have one, with accuracy at 100% (rounded)

@EriktheOutgolfer basically, yeah

@ArtOfCode Is there somewhere I can check this?

yep

looking like a bug

@ArtOfCode seems to have been this fp that set it off

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): stop spammer attack over postfix/dovecot by Marilyn D. Hornung on unix.SE

fp- by Glorfindel

theres definitely a bug, i only have one condition

and it let me re-enable it just fine

99.83%

[ SmokeDetector | MS ] Potentially bad ns for domain in body (1): HTTP403: PROHIBIDO by Beto Javi on magento.SE

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Received an email saying someone has hacked my email account by Marilyn D. Hornung on apple.SE

fp- by Glorfindel

@SmokeDetector Famous email...

Hmmm ...

2 cleantalk posts today, 3 a couple of days ago.

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): How to find a spam link? by Marilyn D. Hornung on wordpress.SE

^ cleantalk

@Glorfindel 3 and counting

You know what they say, "Third time's the charm spam"

Three cleantalk in 9 minutes, all from the same user, who also has 11 rep on SO 🤔

^ I was about to say that.

[ SmokeDetector | MS ] Bad keyword in answer, bad keyword with email in answer, email in answer, pattern-matching email in answer (346): What is the meaning of the Mega-Million's Cash Option - Avoiding lottery's winning taxes by lisa on money.SE

tpu- by Glorfindel

!!/watch-force- almightspellhome@yahoo\.com

@EriktheOutgolfer That pattern looks like it's already caught by Pattern-matching email in body and Pattern-matching email in answer; append -force if you really want to do that.

@CalvT that post is plagiarized. I'm calling this spam.

tpu- by Glorfindel on How to find a spam link? [MS]

[ SmokeDetector ] Blacklists reloaded at rev a6b69cb (SmokeDetector: Auto watch of almightspellhome@yahoo\.com by Erik the Outgolfer --autopull) (running on iBug/EC2)

@Glorfindel Looks like the one upvote came from either FP, LQP or LA review queues

tpu- by Glorfindel on stop spammer attack over postfix/dovecot [MS]

[ SmokeDetector | MS ] Bad keyword in link text in body (65): Linking error Google Colab with Google Drive by Bhaskar Ghosh on stackoverflow.com

tpu- by Glorfindel on Received an email saying someone has hacked my email account [MS]

fp- by K.Dᴀᴠɪs

@SmokeDetector VVVVLQ

@iBug good luck!

[ SmokeDetector | MS ] Potentially bad keyword in body (1): Need to identify this chip on camera lens mainboard ✏️ by Factulas on electronics.SE (@TetsuyaYamamoto)

fp- by Glorfindel

Ironic that she's spamming posts asking how to get rid of spam.

[ SmokeDetector | MS ] Potentially bad keyword in body (1): Margin Calls Short Seller Lifespans Losing Cash by Bruce Roberson on aviation.SE (@TetsuyaYamamoto)

[ SmokeDetector | MS ] Bad keyword in link text in body (65): Помогите отредактировать скрипт на JS by Алексей on ru.stackoverflow.com

tpu- by CalvT

[ SmokeDetector | MS ] Bad keyword in title, phone number detected in title, potentially bad keyword in body, potentially bad keyword in username (168): Quickbooks support number +1 877-756-9341 by quickBooks tollfree on meta.SE

tpu- by Erik the Outgolfer

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): How to completely remove spam bots from Telegram groups? by ColinShark on android.SE

!!/watch-number- +1 877-756-9341

[ SmokeDetector ] Blacklists reloaded at rev 2f5af07 (SmokeDetector: Auto watch of +1 877-756-9341 by Erik the Outgolfer --autopull) (running on iBug/EC2)

fp- by Glorfindel

!!/watch-number- 877-756-9341

[ SmokeDetector ] Blacklists reloaded at rev f240688 (SmokeDetector: Auto watch of 877-756-9341 by Erik the Outgolfer --autopull) (running on iBug/EC2)

hm, what if I used that ? quantifier...

@SmokeDetector Don't worry, the [1] doesn't point anywhere.

@WELZ Your flag conditions were manually disabled by @SmokeDetector: Total accuracy must be >= 99.75%.

yuck

It happens if I edit my condition, if I then manually (unlike SD) enable it... it stays until.... I'm not sure what

[ SmokeDetector | MS ] Bad keyword in body, potentially bad keyword in body (98): 2018 Beste männliche Verbesserungslösung für Männer by sanskritmohli on english.SE

tpu- by Glorfindel

@EriktheOutgolfer The -number checks are strictly strings, not regular expressions.

[ SmokeDetector | MS ] Bad keyword with email in answer (86): http request within Observable happens even without subscription by farooq on stackoverflow.com

@iBug Good luck.

[ SmokeDetector | MS ] Bad keyword with email in answer (86): can i get my BTC back from scam site by michael roland on bitcoin.SE

tpu- by K.Dᴀᴠɪs

!!/watch- michaelroland727\W?@\W?gmail\W?\.\W?com

[ SmokeDetector ] Blacklists reloaded at rev 8924b7b (SmokeDetector: Auto watch of michaelroland727\W?@\W?gmail\W?\.\W?com by Erik the Outgolfer --autopull) (running on iBug/EC2)

[ SmokeDetector | MS ] Toxic body detected (69): How far is too far by S.Huzovic on christianity.SE

@iBug That's going to depend on the specific detection. For many of them, my opinion is that the blacklisted-website detections are already too broad, due to not terminating the regex on both ends. That results in bar\.com detecting the site foobar.com, which is a completely different domain that may have no association with the domain which we intend to detect. Some, few, of the entries are intended to take advantage of this, but, when it occurs, it often results in FP.

fp- by Erik the Outgolfer

[ SmokeDetector | MS ] Bad keyword in title, phone number detected in title, potentially bad keyword in body, potentially bad keyword in title, potentially bad keyword in username (169): Quickbooks support number +1 877-756-9341 by quickBooks tollfree on superuser.com

tpu- by K.Dᴀᴠɪs

Hey all, can someone explain why this is spam as opposed to VLQ? It's short and link-only, but the link is relevant to the question and there's nothing that looks like undisclosed affiliation to me:

I would've deleted it because it's VLQ, but I think the spam penalty is a bit harsh for what looks like a genuine attempt to be helpful unless there's something I'm missing?

someone failed to QA that website

[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, pattern-matching product name in body, pattern-matching product name in title, potentially bad keyword in body (395): How to use weight loss diet pills? by Paulzampas on communitybuilding.SE

@thesecretmaster lemme see if I can find out if it's UDA, otherwise it's an FP

tpu- by K.Dᴀᴠɪs

@ArtOfCode warning, possibly slow loading with VPN

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): 0 confirmations after 3 days. by Bloodir Wiredancer on bitcoin.SE

@thesecretmaster I don't have 100% proof, but I suspect it's undisclosed affiliation. I'd suggest leaving a comment to ask the author if they're affiliated and disclose it if so. TP for Smokey purposes and technically spam-flaggable in its current state, but could be fixed if it was disclosed.

@ArtOfCode I understand that it feels spammy because it's a link-only answer and the link isn't that high quality, but unless it's either obvious spam or clear undisclosed affiliation, it doesn't fit the definition of spam. At least, as far as I understand our definitions of spam.

[ SmokeDetector | MS ] Bad keyword with email in body (66): $500 for product feedback for qualified candidates by New Relic Research on stackoverflow.com

@thesecretmaster Your call on spam flags, obviously, since it's on your site. I would suggest a comment to clarify, though.

tpu- by ArtOfCode

@ArtOfCode I don't like the answer, and I think it should be deleted either way, but I don't think it's spam-flaggable or tp-able for charcoal purposes.

hm

I actually tend to agree with thesecretmaster here

as long as the spam isn't proven, the user shouldn't be punished...

And, IMO, y'all should try to give the benefit of the doubt when possible rather than flag whenever possible.

When you click the flag post link, what does the use case state for spam?

> Exists only to promote a product or service

Which pretty much fits what happened in that post

I said technically spam-flaggable because if you interpret the policy literally and take for granted that it's affiliated, then it is flaggable, but there's enough uncertainty that I personally wouldn't flag it.

@K.Dᴀᴠɪs No, the link seems to be intended to help the asker

It is relevant, but affiliation does still need to be disclosed if it's there

Agreed, and I think you're right, but I think it's borderline enough that Charcoal shouldn't have flagged it. I also doubt that the 6 users who spam flagged it did the same level of research that @ArtOfCode is doing to confirm affiliation.

well... "rsg2003" and "Radhika Grover" don't return even one similar result when googled

I didn't feedback either way, nor did I cast a flag (Mainly because I was rate-limited from the entire stack exchange network due to a userscript that went rouge). But it's certainly borderline.

Is there any way to revisit MS reviews that I have skipped and that are not yet completed?

review history

@K.Dᴀᴠɪs Maybe I'm wrong, but I thought that CHQ only deals with stuff that's clearly spam? If it's borderline, it's maybe mod-flaggable or vlq, but shouldn't be spam-flagged by charcoal?

although, if you've skipped a review, you've skipped it

@thesecretmaster so, the feedback guidance states this:

> If it’s not obvious which type of feedback was appropriate for a post, it’s recommended that you leave a comment on the post - this is generally useful for others looking at the post, and makes invalidating any conflicting feedbacks in the future much easier.

@EriktheOutgolfer Pretty sure that the comment guidance is obsolete.

and, yeah, pretty sure that unsure feedback isn't as good as sure feedback... although, there is another saying: "don't leave comments if you're not familiar with the site"

Didn't we change the policy to be "Don't comment unless you know the site"?

yes, yes we did

well, the link actually refers to the MS commenting facility

@thesecretmaster Let me rephrase, it's borderline to me - I personally wouldn't have provided feedback (tp or fp) in this case. What is spam is a matter of personal opinion, you have yours, everyone else has theirs. Regardless if you are from CHQ or not, opinions will not be the same from one person to another. The people who flagged that post would have flagged it regardless if they saw it pop up in this room or if they saw it on the front page of the site.

@thesecretmaster Take a look at the IP the post was authored from. If it's a California IP, +1 to undisclosed affiliation.

Also, at this point I'm pretty convinced by @ArtOfCode that it's spam, but, as a matter of policy, I don't think it should be spam flagged or tp-ed unless someone's done the research Art did... which doesn't appear to have happened at the time of the report.

[ charcoal-se.github.io ] continuous-integration/travis-ci/push success on 757a7ed: The Travis CI build passed

Someone who spam's "Go check out ClearTalk" 20 times in an hour may not think it's spam, but that doesn't mean it's not

It would be clear spam if it was posted 20 times in an hour, but this link was only posted once, and it's reasonably relevant to the question.

Confirmed middle initial is S.

alright, tpu- is tpu- :-D

@thesecretmaster I was just referring to matter of opinion on why one person may feel it's not spam and another would. I wasn't being specific to your case.

@ArtOfCode You can stop, I agree it's spam :P But as a policy, I think it shouldn't be spam-flagged by charcoal users.

[ charcoal-se.github.io ] continuous-integration/travis-ci/push success on 757a7ed: The Travis CI build passed

@K.Dᴀᴠɪs If it comes down to a matter of opinion, CHQ shouldn't be flagging it, right?

[ SmokeDetector | MS ] Luncheon meat detected (46): File name length by user9983979 on stackoverflow.com

@thesecretmaster In general it's down to each individual whether or not they want to flag it, but without the research and confirmation that a post is spam, agreed.

Everything that is flagged is an opinion. 100% clear spam is an opinion that it's 100% clear spam.

fp- by ArtOfCode

100% to one may be 90% to another.

@ArtOfCode Sure, it's down to the individual, but Charcoal also has policies

I think Art means that, even though there are policies, the final say goes to the feedback owner

(well, the admins, actually :P)

Aye. In this case it's more of a wider "don't flag things unless you're pretty sure they're spam".

^^

^^ = basic Charcoal philosophy, in other words

Then there's a separate question, whether a flag should translate to a positive feedback

I think that the flaggers here were too flag-happy and I'm kinda asking if there's a policy around that/any policy changes that should be made.

@ArtOfCode Again, "Pretty sure" is a matter of opinion. Perhaps the users who flagged that post was "Pretty Sure" it was spam, just as others may be "pretty sure" it's not spam.

A spam flag translates to a positive feedback; that's almost 100% correlated. A positive feedback does not necessarily translate to a spam flag.

@thesecretmaster About flags, nah.

[ SmokeDetector | MS ] Potentially bad asn for hostname in answer (1): What does "burst" mean in this context? ✏️ by a deleted user on ell.SE

@thesecretmaster Is there a policy about that? Not beyond what I said above, "be reasonably certain it's spam" and beyond "it's your account on the line if you're being dumb"

@K.Dᴀᴠɪs I'm sorry, but "everything is an opinion" isn't a good argument here. Charcoal is based around flagging stuff that's objectively spam or r/a. There are objective judgements that can be made here.

That's your opinion

No, it's not. It's network policy on what spam or r/a is.

These things are defined.

Then follow the network policy

And IMO it is undisclosed affiliation

which is part of network policy you're pointing out

my opinion = undisclosed affiliation... Yours = not

Given what Art found, it's clearly spam. There is clear undisclosed affiliation. But I think that at the time of the report, people didn't spend the 15 minutes researching that Art did. So I think that the original flaggers should not have flagged.

Charcoal doesn't - and shouldn't - restrict their members' flags outside of Charcoal. Now would you like to discuss what feedback should have been sent to Metasmoke?

Obvious spam is obvious. Obvious FPs are obvious. The territory in between can be a matter of opinion, but it's usually reducible to one or t'other with some research - if you're not doing that research before you cast a flag, then it's your account on the line for a flag-ban if you're overzealous.

That is the Charcoal flag policy.

yep, Clifford the big red button warns you

fp- by Erik the Outgolfer

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Can I place a comment inside a tag in ASP.NET? by user10647980 on stackoverflow.com

@SmokeDetector University of Jordan (official page)...

@ArtOfCode I don't think that "it's your account on the line" is a good policy here. Especially because once something is spam-deleted, there's a tiny tiny tiny chance a mod will look at it again and invalidate your flags. The chance of a flag ban is much much smaller than the chance that you're flagging incorrectly.

not that you want to flag incorrectly

@EriktheOutgolfer Sure, but with little/no risk of consequences, you don't need to give it as much thought before flagging. That's a bad thing.

[ SmokeDetector | MS ] Potentially problematic ns configuration in answer (46): Exploiting HTTP redirect function via the Host header by Chetan Soni on security.SE

OK, how about flag policy: do good and do it well

@thesecretmaster What would you suggest? Bear in mind that we need to be careful around policy for casting flags - we don't want to make it look like we're a flagging ring or to restrict what people do outside of Charcoal (and that line can be pretty blurry).

Define "flagging ring"

@JohnDvorak That policy doesn't really prevent what happened here. People flagging without much/any evidence of spam. This wouldn't be a problem if just 1 user did it, but charcoal is a space where it's easy for a group of 6+ users to all make this same mistake. That's why I think there should be a bit more policy around it.

@thesecretmaster "only flag if you're reasonaly sure it's spam" is a policy

[ SmokeDetector | MS ] Potentially problematic ns configuration in answer (46): Host header attacks: Possible exploits by Chetan Soni on security.SE

Yet we are still discussing why this post was flagged for spam when we've already determined it was spam. It was flagged correctly, the primary purpose of that post was to promote a product or service - didn't even explain why they would recommend the product or service - it's just clickbait and click at your own risk. And on top of the obvious product promotion, the person was even affiliated with the website, which came secondary.

If you're basing your certainty on that five other people flagged, not great. If you're basing it on what they said in chat, I'd say it's a plus.

naa- by Glorfindel on Exploiting HTTP redirect function via the Host header [MS]

@ArtOfCode What about "only flag smokey-reported posts if you have concrete evidence that its spam"?

naa- by Glorfindel

@thesecretmaster no need for "smokey-reported" there, and the rest is common sense

^

Gotta agree there. That's what we already have, just stated a different way - "do your research" == "have concrete evidence"

@JohnDvorak But in this case, the second half didn't happen. So maybe that part of common sense should be codified somewhere.

@ArtOfCode actually, it doesn't need to be "your research", just "make sure there is concrete evidence"

[ SmokeDetector | MS ] Potentially bad ns for domain in answer, potentially bad keyword in answer (2): Visa for transit through Oman? by Devkant Tiwari on travel.SE

(not just "have concrete evidence")

tpu- by Glorfindel

@thesecretmaster In general, codification only helps if the policy is unclear. That's not the case here - the policy is clear enough, it just plain wasn't followed.

!!/blacklist-website- musafirevisa\.com

[ SmokeDetector ] Blacklists reloaded at rev 05d962d (SmokeDetector: Auto blacklist of musafirevisa\.com by Glorfindel --autopull) (running on iBug/EC2)

@ArtOfCode Ok, so how can we make sure that the policy is followed better?

pile of bricks?

@thesecretmaster That is the question. It's usually followed pretty well, expected differences of opinion excepted. I'm not sure what we can do about edge cases like this.

@EriktheOutgolfer No, unless I should send the flaggers usernames to Undo/Art and they should get out the bricks?

alright, not so large pile

more like what Art said, if you are caught doing stuff like that repeatedly, you risk losing your privs, simple as that

(^ contains addendum)

@ArtOfCode This is prolly too much of a pain in the neck for y'all, but what about a policy that you should leave a MS comment with your reason when it isn't very very obvious?

or our observations, as I have done quite a few times

That could be useful, yes

Defining "very very obvious" is the hard bit there.

@thesecretmaster That would be a breach of the mod agreement :)

[ SmokeDetector | MS ] Blacklisted website in answer, potentially bad ns for domain in answer, blacklisted user (161): Visa to Oman travelling with a car from UAE by Devkant Tiwari on travel.SE

tpu- by Glorfindel

I'd also tone it down from "very very obvious".

@ArtOfCode Sorry, I was being snarky, not serious

@thesecretmaster Aye, that's the ideal situation. Not sure making that a hard and fast rule is the way to do it, though.

[ SmokeDetector | MS ] Bad keyword in link text in body, bad pattern in url body, pattern-matching website in body, potentially bad keyword in body (259): How to Activate Roku Screensavers by hannery lion on stackoverflow.com

tpu- by K.Dᴀᴠɪs

Agreed on it being a soft rule

@thesecretmaster common sense

What it needs is for people to be more circumspect about when they're flagging and resort to other tools (mod flags, MS comments, etc etc) when flagging isn't obviously the right answer.

@ArtOfCode Agreed, but I'm not sure how to accomplish that

aye

hm, maybe some parts of the wiki need a little bit of emphasis if we really want to achieve that

@thesecretmaster Just talking to them should do the trick

sure, talking to them should almost always do the trick, but what if we want to also help the situation not come up in first place? that's a concern too

@ArtOfCode What do you think of pinning a message about this now and maybe kicking off a discussion on the team about possible ways to make people more thoughtful about their flags and more aware of the other options?

prolly easiest, aye

^^ might be a good idea, this is really a core principle of our operation

Please make sure you're doing your research before flagging posts as spam. If you're not sure, leave an MS comment or a mod flag, or discuss with people here first.

My usual strategy when I'm not sure is to ignore the report; is that still OK?

you can always ignore if you're not in the mood :D

even if you take a long hiatus, none of your privs will go away ;-)

Thanks :)

Generally if you're not sure, it seems like not doing anything is the best choice. Maybe also ask the room what they think/ping someone who marked it as a tp?

(Disclaimer: it's been a while since I was feedbacking here)

if I'm not sure, I often post a MS comment

your insight doesn't have to be lost :P

By the way, thanks y'all for being receptive to my concerns. I think that sometimes y'all can be a bit aggressive to people coming in from the outside and requesting changes to what you do, so thanks for being nice to me :)

thanks for raising your concerns too :-) you can see how we actually revealed a very basic issue that we didn't even suspect might be there!

I need to find some more time to work on MS and just be around more

be calm; nobody's pushing you into it

[ SmokeDetector | MS ] Mostly punctuation marks in answer, repeating characters in answer (145): Ошибка Warning: mysqli_fetch_array() expects parameter 1 to be mysqli_result, boolean given by й3уйцуцй on ru.stackoverflow.com

@SmokeDetector "RIGHT!!!!!!!..."

[ SmokeDetector | MS ] Offensive answer detected, toxic answer detected (161): What does 软文 and 小号 mean in internet slang? by hhghh on chinese.SE

tpu- by Glorfindel

naa- by Glorfindel on Ошибка Warning: mysqli_fetch_array() expects parameter 1 to be mysqli_result, boolean given [MS]

@ArtOfCode just to know: are you working on a post over Teams?

not at the moment

feel free to, I'll probably get around to it later if no one else does

ninja :P

Yep

I wanted to ask, do MS announcements really work?

Yes

yes they do

I mean, I just saw an announcement from October, but don't recall getting a notification... maybe it's tucked in my email?

announcements work, but emails for them don't

yeah, especially if your email is geez@se-oauth.metasmoke... :-P

(geez hides actual number)

posted!

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Can I access a raspberry pi from a unity android app? by Jack on stackoverflow.com

[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Why would a country persecute people with autism? by Robus on worldbuilding.SE

fp- by NobodyNada on Can I access a raspberry pi from a unity android app? [MS]

fp- by NobodyNada

@SmokeDetector ig

[ SmokeDetector | MS ] Potentially bad keyword in body (1): how do I install Bitcoin Private electrum wallet on ubuntu 16.04 LTS? ✏️ by Dr.Joe on askubuntu.com

fp- by Erik the Outgolfer

@SmokeDetector Come on, don't describe yourself as "retarded"...

So I missed some stuff.

Is this a "something happened once and now we need a system for it" thing, or is there more than one case?

(actually, more looking at stackoverflow.com/c/charcoal/questions/470)

@WaiHaLee Post 1: Already recently reported [ MS ]

@Undo see the linked discussion over the post, it's essentially its flow that produced it :P (should've mentioned you, eh?)

yeah probably

Okay. If there's a pattern of false positives being nuked wrongly, that's a thing to address aggressively.

If there's not a pattern of false positives being nuked wrongly, then we're doing stuff right.

It looks like what happened here is neither of those; rather, @thesecretmaster is noticing a case where it looks like the system barely worked.

And it's natural and right to see a case like that and go "whoa, that was close, why?"

I haven't really developed an opinion on this, just typing stuff.

well... it was agreed by 3 people including me to post it :P I dunno any further

What's "it"? The Channels post?

yeah

Figured.

yeah, if you had been into the discussion, you would've realized the thing a bit better

I think the Channels post is missing the point: You're not worried about feedback, you should be worried about bad flags

Feedback is just an abstraction for making machines happy. Flags are what count.

flags may deserve a separate post

So your question probably should be "How do we encourage wise use of flags?"

afk

[feedback-wanted] on last part of stackoverflow.com/c/charcoal/a/472/9

um

:)

@Undo suggestion provided

Gut response with no further thought than what happened while typing: This seems like an over reaction based on one post. If there are others like this that we've been called on, then it'd be a bit more useful, but I still think just adding a comment to the MS post is better in terms of development time and effort. We have handled close issues before and I don't like the idea of changing from a binary yes/no on "is this spam" to yes/maybe/no. Maybe becomes the default and makes us less effective

Should have been clearer. Not saying "this is a thing we should build right now to stop an emergency", more "maybe this is a thing we could build if it turns out there's a big problem here"

feedback provided from here too :-)

Our system didn't do that great in this instance. People, including myself, were a bit too willing to flag/feedback. This does this happen from-time-to-time. A significant number of those times we do self-correct (at least of those I've observed). People in here do ask others why they provided the feedback they did, ask them to explain why they flagged, and/or leave MS comments with information from their research. I agree we should do so more often, when we see things that don't quite add up.

I agree a social change would be desirable, where people are a bit more willing to question.

Completely unrelated to this post: Responses in Teams don't trigger the notifications on the SE app. That's annoying.