Admittedly, we're only talking about 55 rep, but I find this worrying.

Yesterday my reputation was at 9206, and it seems like during the night I lost 55 points (currently my old reputation is still visible in the users list). There was no mention of this in the reputation lab or anywhere else in my profile.

I suddenly noticed that this question was deleted by community. Do we allow deletions of questions having answers with a positive score? Additionally, this leaves 15 points unaccounted for, perhaps another question was deleted and does not appear under "deleted recent answers?"?

Thinking about it, it is possible that the mentioned answer was accepted, which would explain all the difference in reputation. I don't remember and it doesn't seem that I can find out without undeleting.

  • $\begingroup$ That user was (manually) destroyed this morning as part of a sockpuppet ring, getting all their negative-score questions deleteted. We're currently discussing the case among the moderators. $\endgroup$ – Raphael Mod Feb 17 '18 at 11:59
  • 1
    $\begingroup$ Now that you mentioned, one of my answers got the deleted as well. Sigh, I was afraid that user would be giving trouble. Ignoring the rep, I really liked my answer (even though the question was silly). @Raphael Would it be okay if ask the question again (in a better format) and self-answer? $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 12:25
  • 1
    $\begingroup$ Sorry for keeping you waiting. We're spread across time zones; we're not forgetting. Please hold the line. =D (cc @Discretelizard) $\endgroup$ – Raphael Mod Feb 17 '18 at 17:52
  • 1
    $\begingroup$ @Raphael does this count as the answer (I think it does). And now it brings another question, is really the system of nuking proper? It may happen that someone put off-topic questions, but good answers should be in my opinion excluded somehow. It is not someones fault that puppet master asked questions. I get that 55ppints is not a lot, but the design seems flawed. Also that puppet master may ask good question, in that case maybe good questions could be chowned by Community? $\endgroup$ – Evil Feb 17 '18 at 17:53
  • $\begingroup$ from this it does look like that maybe removal of "designated" false votes by mods does not show up in (normally quite comprehensive) user rep tracking history (see following tab) aka "vaporized" which is indeed a bit confusing although other SE features work similarly eg deletions etc possibly intentionally. ie no loss seen here. but note that everyone touched by false votes (not just you) has same case. it could be significant impact cross-users-wide if the fake user voted alot. cs.stackexchange.com/users/27055/ariel?tab=reputation $\endgroup$ – vzn Feb 17 '18 at 18:01
  • 3
    $\begingroup$ @Evil This question has been asked across the network long ago, see meta.stackexchange.com/questions/199860/… . No response from the SE staff, though. $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 18:23
  • $\begingroup$ @vzn The problem here isn't fake voting, it's removal of questions with negative questions score, but which have an answer with multiple (at least one non-sockpuppet in Ariel's case) positive votes. $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 18:25

Thanks for raising this. I think this should be addressed now.

This was my fault. You are right that, normally, downvoted questions can't be deleted if they have an upvoted answer. Here's what happened. Yesterday I dealt with a sockpuppet ring. This triggers automatic deletion of all questions with a vote score below 0 from the socks. I manually reviewed all such deletions to make sure we weren't losing good questions, and rescued some of them. There were three remaining questions that I judged weren't worth keeping because the questions were so specific that I doubted anyone would ever have the same question again. So, I left them deleted.

However, after looking at this again, you raise a good point; for at least two of those questions, the answers do seem like they have lasting value. I made the wrong call. Sorry about that.

So, here's what I've done to save the good material. For two of those three questions, I did radical surgery on the question to get at the conceptual point that the answer is addressing, and I've undeleted those two questions. Hopefully now they are decent questions that someone might plausibly have again in the future, and that they'll be able to find if so. You can look at Detect whether one Turing machine invokes another and Solving diophantine equations -- does having a bound on the size of the solution help? if you want to review the surgery I did and make further improvements.

There is one remaining question that I just don't see value in. It's a "please give me feedback on my idea" question that had lots of symbols but doesn't seem to have any useful content that I could extract, and the answer basically says that. I don't see much chance that this will be useful to anyone else in the future. Perhaps you'll see a way to extract a useful interesting question that has a non-trivial answer, and if so great, but I couldn't. So, I'm going to leave this deleted. (Caveat: I suspect this link will only be accessible if you have at least 10K reputation, so you might not be able to see the deleted question.)

I hope this sounds like a reasonable resolution. Thanks for the nudge about this; I'm happier now that we could find a way to save the useful content that you and others wrote.

  • 1
    $\begingroup$ Thanks for make the effort to save the two 'patients'. As for the third, I've made a review of it's style here: cs.meta.stackexchange.com/a/1451/65339 , which basically concludes the style is awful. If you think the content has no value either, I agree with your decision to let this one deleted. Although I'm a bit disappointed in the deleted user for the behavior and the fuss, all is well, as far as I'm concerned. $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 18:37
  • 1
    $\begingroup$ I made a small change to this question. You removed the P vs NP implication, but this is inconsistent, as my answer directly addresses this implication. Therefore, I added the P vs NP implication at the end, such that this is merely in the background, but the answer remains consistent. $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 18:44
  • $\begingroup$ @Discretelizard, cool, thanks for the improvement! Perhaps I'm confused, but I think the question whose style you were reviewing in that meta post is yet a fourth question, distinct from the three I link in this answer and not the same as the one that remains deleted. $\endgroup$ – D.W. Mod Feb 17 '18 at 18:44
  • $\begingroup$ Oh whoops, my bad, one those questions was deleted and I just assumed it was that one (with my rep, I can only see the question numbers and the looked the same at first glance, although not at the second glance). Then I have nothing to say about the third question. $\endgroup$ – Discrete lizard Mod Feb 17 '18 at 18:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .