« first day (2491 days earlier)      last day (1345 days later) » 

10:00 PM
[ SmokeDetector | MS ] Potentially bad keyword in answer (1): Can't remove a folder - "Element not found" by cyfrost on superuser.com
[ SmokeDetector | MS ] Link at end of answer (61): How to manage many static html pages with similar content by gridlocdev on stackoverflow.com
 
10:16 PM
morning @Rubiksmoose ... :P
 
user435118
@Makyen Will we implement SmokeDetector implementing pingable /superping support since it's actually easier than I thought?
 
Hmmm let's see if I get this right.
!!/scan www.businessapac.com
 
@Rubiksmoose Post 1: That does not look like a valid post URL.
 
10:20 PM
nope
!!/bisect www.businessapac.com
 
@Rubiksmoose Matched by businessapac\.com on line 13850 of watched_keywords.txt
 
@Daniil what exactly are you proposing? If SD can't ping someone, post a warning message so a mod can superping?
 
10:26 PM
tp feedback received on [MS] Computer does not detect usb stick
 
@Daniil I'm aware of that API route (I don't recall if I found out about it before or after out last conversation on SD pinging). I'm not sure what you're asking, or what exactly you want to implement or how you are wanting to implement it.
 
API quota rolled over with 9853 requests remaining. Current quota: 19999.
math: 655
stackoverflow: 324
askubuntu: 112
physics: 105
superuser: 86
unix: 83
electronics: 76
mathoverflow.net: 59
english: 53
diy: 45
mathematica: 36
serverfault: 34
ell: 34
academia: 32
worldbuilding: 29
meta: 29
apple: 28
salesforce: 25
scifi: 25
money: 22
rpg: 22
chemistry: 21
hermeneutics: 21
engineering: 20
datascience: 20
space: 19
ai: 18
emacs: 18
law: 17
drupal: 17
chess: 16
wordpress: 16
raspberrypi: 15
gaming: 15
bicycles: 15
es.stackoverflow: 14
dba: 14
security: 14
android: 14
gamedev: 14
[ SmokeDetector | MS ] Few unique characters in answer (87): Testing Question (API Test, Not A Real Question) by User-92 on stackoverflow.com
 
tpu- feedback received
 
user435118
@NobodyNada Yes, for autoglagged fp's, flagged conditions disabled and notify messages.
 
user435118
@SmokeDetector Shouldn't that be @metasmoke?
 
10:30 PM
@Daniil Nope, because that message comes from the SmokeDetector side
 
fp feedback received on [MS] Whirlybird part
tp feedback received on [MS] Using Google Maps Data
 
user435118
@NobodyNada The distribution of messages seem quite random to me but never mind.
 
API is per-IP, so per-instance (so Smokey side)
 
user435118
@Mithical There's other messages as well when it's confusing and inconsistent.
 
10:34 PM
@Daniil Yeah, it kinda can be sometimes :) Essentially, SmokeDetector handles fetching, scanning, and reporting posts; blacklist/watchlist operations; and chat commands, while Metasmoke handles cataloguing past reports, interacting with userscripts/other API apps, and monitoring GitHub activity
 
API quota falls into "fetching, scanning, and reporting posts" so it's managed by SmokeDetector
 
In the end, whether a message comes from SD or MS is really determined by where the relevant data is
 
10:39 PM
@Daniil That seems reasonable to me, although that I'm not sure that we really need it for notify messages
 
user435118
Is there any plan to fix the inconsistent whitespace in all code owned by Charcoal?
 
@Daniil Enable dark mode? ;)
 
teward/Osiris: Executing automatic scheduled reboot.
 
@Daniil what do you mean? I'm pretty sure we're consistent within SD and MS, AFAIK that should be enforced by flake/rubocop
 
10:40 PM
[ SmokeDetector ] SmokeDetector started at rev 97b8f57 (Makyen: ChatExchange requirement --> Charcoal-SE's CE repo's master branch) (running on teward/Osiris, Python 3.6.9)
 
user435118
@NobodyNada Hopefully 3rd time lucky.
 
If you feel like taking that on... sure. ;)
 
Restart: API quota is 19922.
 
user435118
@Rubiksmoose ?
 
@Daniil no more whitespace in dark mode
 
10:41 PM
^
 
It's darkspace
 
user435118
...
 
user435118
@NobodyNada A random pattern of blank lines here and there.
 
@Daniil You keep tossing out things like this with context nearly completely missing. We can't read your mind to understand what you're thinking. Please provide more context when you're asking questions, so that the first thing that happens isn't people having to ask what you mean.
 
@Daniil Oh, you mean (e.g.) between functions rather than indentation? I think that's enforced by the linters too...
 
10:43 PM
@Daniil Listen all I can contribute right now is handling some reports and making bad jokes/puns. :P
 
user435118
@Mithical Sure, once I learn more git, finish my bot and do a whole load of other things. :)
 
sounds like a plan
 
user435118
@NobodyNada In functions (mainly).
 
user435118
@Makyen Probably because I've asked the same things before.
 
@Daniil Could you provide an example or two? AFAIK our formatting is usually pretty reasonable, and the linters prevent it from getting out of hand. That being said, line spacing within functions is generally left to the programmer's discretion, so I wouldn't be surprised if there was a function or two with some slightly funky spacing
 
user435118
@Mithical Yeah, probably in a few years :P
 
@Daniil There's around a thousand messages posted here each day; not everyone reads all of them and people forget over time -- so even if you've brought something up before, not everybody's going to remember the context
 
e.g. I do not remember being a part of any prior discussions about inconsistent code formatting
 
10:49 PM
It doesn't really sound like it's something that's causing an issue, or something that would be high priority, so I'd hazard a guess that no, there are no plans currently to go through and be consistent with whitespace.
 
user435118
Also do we use some weird PEP8 version because the line limit is 79, not 120...
 
@NobodyNada Agreed. That's usually something that's at the discretion of the person that's writing it, or editing that function/area. Each person generally has personal preferences in this area.
 
@Daniil Looks like we have the line length overidden: github.com/Charcoal-SE/SmokeDetector/blob/master/tox.ini#L15
 
@Daniil I'm not aware of you ever previously bringing up "fix[ing] the inconsistent whitespace in all code owned by Charcoal".
 
@Daniil consistency with what? The current spacing looks reasonable to me: the programmer wanted the whole function to read as one clause except for the Metasmoke check at the beginning. It's also consistent with the two other similar functions post_auto_comment and get_post_bodies_from_ms
 
user435118
10:53 PM
@NobodyNada F841 shouldn't be overridden IMO.
 
@Daniil Which is, IMO, a ridiculously short line length, based on archaic requirements, which are rarely encountered anymore (i.e. 30 years ago, it might have been reasonable, but really you have to go back 35 or 40 years). Line length limits that low, are, IMO, very detrimental to code maintainability, because they strongly encourage programmers to use variable names which are cryptic, instead of self-documenting.
3
 
user435118
@NobodyNada Consistency with a blank line before the payload.
 
fp feedback received on [MS] How to make a viable flying mount?
 
@Daniil that's not something that needs consistency
The blank line is to separate the MS key check from the rest of the function, not the payload from the rest of the function
 
10:54 PM
 
user435118
@NobodyNada Debatable.
 
It's an open source project that's years old, why would anyone expect consistency in any case
 
user435118
@Mithical Because it's open source.
 
To be frank, I can think of better uses of dev time than going through adjusting whitespace.
3
 
@Daniil yes, I'm inclined to agree on that one. However, it's possible the warning was added after SmokeDetector had already been around for a while, and whoever updated flake8 was in a hurry and decided to just supress the message
 
user435118
@Makyen I remember me asking you to change "[Note: On Macs, it may be necessary to remove the --user option from the last pip3 command line in the above instructions. However, the --user option is known to be necessary on other platforms. Further testing is necessary to resolve the discrepancy.]"
 
user435118
to "[Note: On Unix-based systems (like Mac's and Linux), it may be necessary to remove the --user option from the last pip3 command line in the above instructions. However, the --user option is known to be necessary on other platforms. Further testing is necessary to resolve the discrepancy.]" in the README...
 
user435118
10:57 PM
I think you might've forgotten...
 
fp feedback received on [MS] Get title from YouTube videos
fp feedback received on [MS] OPA attack when IV is FF?
 
@Daniil Make a PR?
 
@Daniil forgotten what? That seems like a reasonable clarification
 
user435118
10:59 PM
@NobodyNada I would ask for it to be enabled but it probably won't so there's no point.
 
@Daniil Our primary testing has been on unix/linux based system where the --user is definitely required.
 
user435118
@Makyen Same, and I always get errors.
 
"Local variable unused" warnings are, in my experience in other projects, great things to check in CI but utterly maddening if they're also enabled for local dev. I am not sure how Charcoal enforces this.
 
user435118
@Mithical Too trivial and no since I asked them before and if I can remember they agreed to do so.
 
@RyanM This one is a linter rule which runs only in CI or if you manually run the linter, so enabling it seems reasonable here
 
user435118
@Makyen I got errors on Mac and Ubuntu.
 
I've made PRs over single characters before. That's how open source works :P
 
Ah yeah that makes sense then. I don't know this lint tool at all, but does it support baselines? That's how my team at work handles enabling lint rules that are currently violated.
 
user435118
11:02 PM
@Mithical It would be much quicker to ask someone than to have the PR sitting around for a couple of months.
 
@Daniil If you want to enable it, test it locally, remove all the false-positives, and make a PR then feel free. However, if you're just planning to ask if someone else wants to do it, then you're probably right that it's not going to happen just because there are higher-priority things to do
 
user435118
@NobodyNada What higher priority things?
 
4 hours ago, by NobodyNada
However, that's just a short-term hack. For a long-term solution, we'll either have Metasmoke log into chat and post the messages itself (so runners won't need creds), or we'll have Smokey log in under two accounts (so runners will need creds).
@Daniil ^ first thing that came to my mind
 
@Daniil Whatever they feel is more important than fixing something that isn't demonstrated as an immediate problem.
 
user435118
Jun 12 at 11:08, by Daniil
(env) dmanokhin@MacBook-Air SmokeDetector % env/bin/pip3 install --user -r user_requirements.txt --upgrade
ERROR: Can not perform a '--user' install. User site-packages are not visible in this virtualenv.
 
user435118
11:05 PM
Jun 12 at 11:08, by tripleee
@Daniil you can drop the --user option when using a virtual environment
 
user435118
Jun 12 at 11:08, by tripleee
everything is user-installed now anyway
 
user435118
@NobodyNada I wouldn't say that's "high priority" but ok.
 
@Daniil At an absolute minimum, you should have linked me to the page which you want changed. By not doing so, you've increased the work I would have to do to make the change (i.e. I'd have to go find what you want changed; I probably know where it is, but still you should link it), which makes it less likely for me to make the change.
 
user435118
11:07 PM
@Makyen Really? The README is the front page...
 
@NobodyNada Heck, even clearing out the MS queue is probably more urgent, since that affects weights and autoflags.
 
@Daniil Ohhh, I understand what you mean now. Yeah, your change makes sense to me -- I didn't catch the venv/non-venv difference earlier. How about something like the following wording (cc @Makyen):
 
@Daniil Yes, you're asking me (or anyone) to do something for you, so you should make fulfilling your request as easy as possible for the person you're making the request of. Making your requests easy to fulfill makes it vastly more likely that your requests will be fulfilled.
 
> Note: On Unix-based systems (like Mac's and Linux), it may be necessary under some circumstances to remove the --user option from the last pip3 command line in the above instructions. However, the --user option is known to be necessary in other circumstances. Further testing is necessary to resolve the discrepancy.
 
user435118
@Mithical It wasn't urgent when I suggested it but now it suddenly is.
 
@Daniil Which front page?
 
naa feedback received on [MS] Merge branch 'master' of github.com... in commit message
 
@Daniil It's the first thing that came to mind that was more important than a NFC formatting/linter change. The current solution for posting messages under the MS account is a good stopgap but it has some rough edges
 
11:12 PM
@Daniil ?
 
user435118
@Makyen The README, is the front page of the SD repo. Do I really need to link to the README?
 
user435118
If you really need it ^
 
user435118
@Mithical I have suggested numerous solutions multiple times over the past 5 months to clear the queue, even started on it myself encouraging others. I have tried at least 5 times, even more but it's all been dismissed and ignored. But now it's suddenly a "high priority".
 
@Daniil You need to be specific enough to inform me of what you're talking about. I never look at SD's README.md file.
 
It's not the most urgent thing, no (although I've personally been a bit concerned about it for a while). However, it's objectively more important than stylistic changes in old code, since it has a direct impact on the day-to-day activity and behavior of the system.
Sure, consistency and style is nice, but it's not a priority, especially with limited dev time.
 
@Daniil One notable factor is that the review call-to-arms was a Teams post, which is a lot less transient than a chat message -- so lots more people saw it, and everyone was continually reminded of it
 
11:17 PM
That's all I'm getting at.
 
@Daniil No, it hasn't been dismissed and ignored. Your suggestion, which you've repeated multiple times, was to just dump the entire queue, without having people go through it. Yes, that was considered and rejected. The fact that you did not accept that that suggestion was rejected has been one of the things that has contributed to the friction in your dealing with this group.
 
[ SmokeDetector | MS ] Link at beginning of answer (32): MotionEyeOS without internet by user14267492 on stackoverflow.com
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword with email in body, potentially bad keyword in body, potentially bad keyword in title (252): WhatsApp:(+91 74082 09536) Buy Genuine IELTS Certificate Without Exam| Registered IELTS Certificate for sale* by pmp on meta.SE
 
tpu- feedback received
fp- feedback received on [MS] MotionEyeOS without internet
 
user435118
@Makyen Not because of that. I've suggested numerous times we go through the queue not clear it. Also, everyone doesn't listen to me from the day I suggested something people didn't like, or possibly even the day I came here.
 
user435118
I've also suggested alternate solutions.
 
@Daniil Getting an up-swelling of interested by multiple people in going through the queue and giving feedback is vastly different than one person repeatedly harping on wanting to just dump the queue.
 
user435118
@NobodyNada *14+ chat messages.
 
the thing about The Barbequeue is that it's a group effort -- we have multiple people reviewing all at the same time, and we're making very noticeable (and satisfying) progress which encourages people to keep going. Getting that to happen takes a considerable amount of both marketing and luck
 
11:22 PM
@Daniil OK. This is the point where I recommend that you step back and take some time away from here. Continuing from this point has generally resulted in negative situations.
 
Okay... @Daniil can I please ask you to disengage for the moment?
 
Makyen/EC2-linux: Executing automatic scheduled reboot.
 
fp feedback received on [MS] How does this end effector work?
 
user435118
@Makyen I'm not harping. I know you all have inbuilt impressions on me which are wrong. I clearly suggested alternative solutions like adding tpu feedback, adding a handled queue and asking people to kick things out of the queue which I reviewed.
 
2 mins ago, by Makyen
@Daniil OK. This is the point where I recommend that you step back and take some time away from here. Continuing from this point has generally resulted in negative situations.
 
user435118
@Makyen I'm having a discussion which is peaceful so why?
 
!!/blacklist-number- +91 74082 09536
 
PR#4764 ("user12986714: Blacklist +91 74082 09536") opened by SmokeDetector
 
user435118
Every time I try and make a point I have to go away.
 
[ SmokeDetector | MS ] Pattern-matching email in answer, potentially bad keyword in answer (100): Pinging in Linux terminal until getting N number of successful packet deliveries by Elizabeth seanpaul on askubuntu.com
 
!!/watch- IELTS certificate (?:\w*+\W*+){0, 5} without exam
 
11:28 PM
@user12986714 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.
 
@user12986714 Post 1: Already recently reported [ MS ]
 
tp feedback received
 
user435118
sdc reject 4764 "1 tp, enough weight."
 
@Daniil Closed pull request #4764.
 
11:31 PM
!!/blacklist-keyword- fabuloushacker\d*+@gmail\.com
 
PR#4765 ("user12986714: Blacklist fabuloushacker\d*+@gmail\.com") opened by SmokeDetector
 
@Daniil I know you're frustrated right now and it's affecting how you're asking things. That's part of the reason I'm recommending you step away. If you want answers to such questions, I suggest you ask them tomorrow.
 
user435118
sdc bisect IELTS certificate without exam
 
11:32 PM
Matched by the following regexes:
(?-i:IELTS certificate without exam)(?#Please leave the "without exam" on this blacklist as there are legitimate cases where an "IELTS certificate" may be asked about, e.g. on Travel.SE or Academia.SE) on line 2328 of bad_keywords.txt
ielts\Wcertificates? on line 2772 of watched_keywords.txt
 
Aug 27 at 17:13, by Daniil
Sorry everyone for the disruptions I have caused. I know my communication can be abrasive, I have a problem with this and could improve, but I need your help in this, please. In the future, please say something like "Stop Please @Daniil" so I know when to stop, take a step back, listen to you trying to explain the situation and, perhaps, guide as to other possible ways to communicate, or routes I can take to obtain what I appear to want.
10 mins ago, by Mithical
Okay... @Daniil can I please ask you to disengage for the moment?
 
user435118
sdc unwatch ielts\Wcertificates?
 
[ SmokeDetector | MS ] Pattern-matching email in answer, potentially bad keyword in answer (100): How to add new — and non-defined — metadata to an MP4 file? by kimperly lopez on superuser.com
 
user435118
sdc watch ielts\Wcertificates?(?! without exam)
 
11:33 PM
tpu- feedback received
 
user435118
@Mithical I should probably delete that.
 
user435118
@Makyen What questions?
 
(sigh)
 
user435118
@Mithical ?
 
11:36 PM
@Daniil I think it was a healthy response on your part and a good step towards mending relationships. It made me happy to see you had taken that step. You could certainly delete it, but it would seem to be a step back IMO.
8
 
user435118
I want to delete it because people are overusing/misusing it and just linking to it whenever they don't like my point and want me to shut up.
 
You were asked by multiple people now to step back for a little while, to prevent this from escalating. Nobody's trying shut you down; simply trying make sure things stay constructive.
I'm aware that you perceive it differently. I'm asking you to trust in the people around you, since the end goal is to help you and your interactions with the people here.
 
@Daniil I really can understand why it might feel like that. I honestly do. However, from what I've seen that doesn't seem to be the case from an external POV. I think people are reminding you of it though not to shut you up but [basically what Mith just said]
 
user435118
> Be nice to me and you'll get the same in return :)
 
user435118
(my chat bio, try it ^ )
 
user435118
11:42 PM
Also I should note it was my chat bio for the last few hours, not just now.
 
@Daniil If I may: I agree with Rubiksmoose that your message was a very healthy, mature response to the situation. The point at which it's often cited is when the conversation moves from discussion of ideas to solve problems to things like "everyone doesn't listen to me," which ceases to be about the ideas and becomes about the people discussing them. That's the sort of thing where even with the best of intentions, things can quickly devolve, especially in an emotionless medium like a text chat.
 
1) I hope that I have never been not nice to you and I'm always willing to apologize if I have wronged you.
2) That isn't fully the way that works. Really we (especially mods) should be nice regardless of how people treat us. We should respond to not nice behavior in proper ways, not in kind.
As frustrating as that can be at times...
 
user435118
@Rubiksmoose (in response to 2) I would like to do that but some people have pushed it too far and that's the only option I have with them.
 
user435118
@Rubiksmoose (in response to 1) I wasn't referring to you. :)
 
@Daniil I figured but just wanted to be clear. :)
 
11:47 PM
Everyone here is ultimately working towards the same goal. There just seems to be a bit of a personality clash, and things keep escalating. Until we can resolve the personality clash somehow, disengaging is a good way to prevent escalations and devolving conversions.
 
@Daniil Disengaging is almost always a better option than being unkind here. And listen, I've seen people cross that line with you here and I've even called some of them out (not always was I in a place where I could do so).
And really especially with people here, longtime very active users, the best option is to figure out where that friction is and figure out how to work around it (or, ideally, remove it).
 
@Rubiksmoose Yeah, I'll corroborate that.
 
user435118
@Rubiksmoose What do you mean by calling them out?
 
@Rubiksmoose You know what I'm going to own up to it: I really haven't done enough to call out line crossing where I should have for various non-sufficient excuses. And I apologize for that. Not that it's always productive (there's certainly a time/place component and until recently I hadn't really been or felt as part of the group) still I feel that some behavior towards you was not properly called out and I'll own my part in that.
On the other hand I don't think that ever justifies escalating the nastiness. From a practical level, it just doesn't help anybody. It breaks down relationships and causes toxicity. And from a site-running perspective it also looks bad coming from moderators.
 
We are only getting 1117 feedbacks today. What about we target 2000 feedbacks tomorrow?
 
11:55 PM
@Daniil That you haven't seen discussions with other people about how they behave doesn't mean the discussions didn't happen. Have all the discussions which you've had about behavior been here in public? Such things are often done privately. It's rare for there to be disclosure of such conversations, even of just the fact that a discussion was had.
 
user435118
@Rubiksmoose :) Means a lot, thanks :D
 
@user12986714 Speaking personally, I tend to try and enjoy the weekend doing other stuff and don't participate here as much. So that's one reviewer down. Definitely plan on resuming on Monday for sure, and doing some tomorrow.
 
@Daniil Basically saying: please stop, that behavior crosses a line. It's important to call those out, in a healthy way, to maintain a healthy group dynamic IMO. How to do it in a good way (one that results in changing behavior and not escalation) can often be complicated though.
 
user435118
So, now that I've calmed down a bit (I think). @Makyen Can you please make a change to SmokeDetector's README.md file. Specifically changing the bit that says: "[Note: On Macs, it may be necessary to remove the --user option from the last pip3 command line in the above instructions. However, the --user option is known to be necessary on other platforms. Further testing is necessary to resolve the discrepancy.]".
 
user435118
As after testing on Ubuntu I had the same error. I quite liked NobodyNada's suggestion.
 
11:59 PM
Yeah, I'll agree with Rubiksmoose here: I can think of one situation in which I think that you were treated a bit more harshly than you probably should have been (and there may be others I haven't seen). I understood how that person had gotten frustrated enough to get there, but we should hold ourselves to a higher standard than taking those frustrations out on others through harshness.
 

« first day (2491 days earlier)      last day (1345 days later) »