@ArtOfCode You invalidated the tpu feedbacks on this post. As stated in chat, that answer was plagiarized. I was under the impression from the Feedback Guidance that we were explicitly supposed to provide k/tpu feedback for plagiarism. Are we really supposed to base feedback on the contents, ignoring that it's plagiarized?
If that's the case, I'm fine with doing so, but it would be helpful to get the Feedback Guidance changed.
IMO, if the feedback is going to be used to train some ML implementation, then, unless that ML implementation is also taking plagiarism into account, then the feedback should be based on just the post/question contents without respect to it being plagiarized (perhaps unless it's plagiarized from content under the same question, which is data the ML could have).
Is there a way we can indicate that there is information which needs to be taken into account? Perhaps pinging you if we see a situation which might result in an issue (e.g. conflicting feedback on a post that's not clear, or that's clearly not the feedback provided without additional information)?
@MartinSleziak Please use !!/scan first. You can then use !!/report, if nothing is found by the scan. Doing so causes the post to be scanned by SmokeDetector and, potentially, have autoflaggs applied, which results in it being deleted faster.
[ SmokeDetector | MS ] Bad keyword in answer, bad keyword in username, blacklisted username, blacklisted website in answer, potentially bad keyword in answer: Travel Issues to Turkey by The Away Days on travel.SE
Feature request: If the post 99% (or some N%) matches any of the last 5 posts which have been deleted via manual flags (or manual + autoflags), then autoflags go to 5 (or 6).
[ SmokeDetector | MS ] Bad keyword in answer, bad keyword in username, blacklisted username, blacklisted website in answer, potentially bad keyword in answer: ATM fees in Turkey by The Away Days on travel.SE
@ByteCommander Relating to this: An RO needs to pin a message saying: "A new site called: has been created" with a link to the site when a new site is created.
@GrantGarrison Frankly, I consider it debatable. If an identical post (or perhaps multiple identical posts) has/have been recently confirmed by humans as spam, at a minimum it should have considerable weight as a reason. Why do you feel new, identical copies should not be automatically deleted?
[ SmokeDetector | MS ] Bad keyword in answer, bad keyword in username, blacklisted username, blacklisted website in answer, potentially bad keyword in answer: Where is Kechror, Turkey? by The Away Days on travel.SE
@Makyen I think just one human isn't too terrible of a price to pay. It also keeps the number 6 away from autoflagging in the code, which is also good.
@gparyani Hm. I want it to also be obvious to the moderators of the site, should they be curious about spam flags on a seemingly nonspam post. Hence a comment.
'I think a comment suffices for both purposes. I might be wrong, but I explained my reason.
@GrantGarrison oh okay; it seems like its core, not admin. I still can't access it though; core is based currently on feedback per month, and I hardly give feedback nowadays. I guess you'll just have to ask someone else. Never used it.
@K.Dᴀᴠɪs Added you['re]{0,3}\W.{0,9}bast[ua]rd to watchlist
CI on bf348ac succeeded. Message contains 'autopull', pulling...
[ SmokeDetector ] SmokeDetector started at [rev bf348ac (SmokeDetector: Auto watch of you['re]{0,3}\W.{0,9}bast[ua]rd by K.Dᴀᴠɪs --autopull)](//github.com/Charcoal-SE/SmokeDetector/commit/bf348ac) (running on Henders/EC2)
store => {
let results = [];
for(let index = 0; index < store.posts.length; index++) {
if ((JSON.stringify(store.posts[index].autoflagged)==="true")&(store.posts[index].user_reputation>1)&(JSON.stringify(store.posts[index].is_tp)==="true")){
results.push([JSON.stringify(store.posts[index].user_reputation)+" stackexchange.com/users/…);
}
}
return results;
};
It shows all users that gained reputation after a true autoflag. Reputation is on one side, link on the other
@gparyani So it seems the problem is with my cable -- measuring says 5v is coming out of the power supply (and the Pi is drawing 0.3a), while the other end of the cable says only 4.7V (!)