[ SmokeDetector | MS ] Url in title, bad keyword in body, bad keyword in title, body starts with title and ends in url, potentially bad ns for domain in body, +3 more (392): chaterlimited.com/thermosculpt-pro/ by wodud on wordpress.SE
You have repeatedly posted dubious and highly promotional answers with links to your employer's web site. This is not acceptable promotion -- answers on Stack Overflow and Stack Exchange sites should provide information to solve the question at the top of the page, not circumstantial hopefully relevant commentary. — tripleee48 secs ago
@CalvT It's what SE shows on the post and delivers in the HTML version of the body via the SE API. There's already an issue on GitHub that SD should look at Markdown.
@CalvT @ArtOfCode Probably what happened was: The original version was what SD reported, without it being in code format. After seeing that basically nothing was displayed, the OP edited it during the grace period to put it into code format.
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, pattern-matching product name in body, pattern-matching website in body, +1 more (492): Finding The Right Weight Loss Product by Juanezxuten on stackoverflow.com
@Glorfindel @ArtOfCode would you mind if we added support to Metasmoke for sending iOS push notifications, or would you rather have that functionality on a separate server?
@ArtOfCode READi (or another iOS app) would sign up for notifications by POSTing an APNS token to MS, which would be stored in the DB. When a new post comes in, MS would send a request to Apple’s servers containing the notification info and a list of registered tokens.
We’d need, at the minimum, one table to store the list of device tokens. We’d probably also want a few more tables/columns to store notification settings such as site filters or mute options.
(I’m also imagining that when the app is open, it would ask MS to send silent push notifications for everything that would come down the websocket, because I’m hoping that would be better for battery life.)
@ArtOfCode well, I’d imagine that would require an Android app to go along with it — but most of the functionality could be shared; only the notification delivery would be different
@NobodyNada Sounds fine to me. Added bonus - if you do create a table to store notification settings, we should be able to use that for Smokey notifications as well and finally eliminate the my-notifications-got-lost problem.
@K.Dᴀᴠɪs That's a small version. Ones like this were still something you could carry. There were earlier versions that were intended to be mounted in cars. (source)
@thesecretmaster cool, thanks — I’ll do some research and write an issue soon. Winter break starts in a couple weeks, so I might be able to implement it then if you don’t get to it first.
Also, @ArtOfCode/@Undo to reduce noise in MS issues, do y'all think it would be worthwhile to have a separate place for long-term/roadmap type stuff? Like #257, #289-#291, #529, etc?
@ArtOfCode APNS is more than that, but it's not bad if you have someone who knows what they're doing (and willing to do the maintainance work, which probably isn’t me right now)
while the admins are around: I've noticed that since I've started hanging around here, I have been in quite a few disagreements about feedback in which I'm much more hesitant to flag as spam than others (I can give a few links if needed). I wanted to ask: are a few people being overzealous, has the room's culture changed, or am I being too conservative with my feedback?
@NobodyNada Nah, you're good. I'm typically more conservative with feedback as well. As long as things aren't getting nuked when they shouldn't have been, we're generally okay.
but the more I think about it, the more I think the numbers bear that out
I did some rough math a few months back. We know average % helpful stats - 95.4% non-SO, SO was somewhere in the 80s. Charcoal people average around 98%.
That said, helpful flags probably aren't the right stat for this...
Okay, invalidated feedback averages around 5%. Chance of an error is 5% per person. 0.05^6 is... 0.000000015625.
Recovered from KeyError: 'I have a problem compiling the following minimal example:\n\n\\documentclass{article}\n\\usepackage{listings}\n\\usepackage{tcolorbox}\n\n\\begin{document}\n\\begin{center}\n \\begin{tcolorbox}\n \\begin{...'
@ArtOfCode That would be true if feedbacks were independent, but I think that there is at least a bit of people taking into account what others have feedbacked.
@NobodyNada IMO, people have gotten more willing to flag as spam, rather than give the benefit of doubt. I see this on a regular basis, particularly on the first occurrence of either a site/domain, or post by a user. One of the more common is people providing TP feedback (and probably flagging) where there's a short answer that's recommending (promoting) a site, but where the question requests that type of information (i.e. it's promotion, but not unsolicited, which isn't spam by SE definitions).
I think this is partially caused by the implicit gamification to be the first person to feedback (as evidenced by the SD message saying X gave Q feedback). Maybe for those messages we just say that the feedback was received, rather than who gave the feedback (e.g. just say "Q feedback received" or the expanded "Q feedback received on [link to post]"). If someone wants to know who gave what feedback, that can be obtained from AIM or MS.
Admittedly, I like seeing who's working the review queue, as evidenced by strings of new feedback on older posts, but I'm willing to not have that, if removing this part of the unintended gamification results in people spending a little more time thinking about their feedback and flags.
@ArtOfCode Sounds good. I don't know if it will help, but there's definitely competition to be the first person to feedback, which probably does result in people spending less time thinking about their feedback/flag.
@EriktheOutgolfer It should still be available through MS and AIM (along with anything else that displays it based on the MS API data).
@EriktheOutgolfer It doesn't report it when the feedback is given via chat. SD only reports it when the feedback is given via the MS API or the MS HTTPS routes.
The link appears to have nothing to do with the question. It's to the download page (not Google Play or Apple Store) for an app supposedly about educating kids.