What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

I flagged an answer on SF yesterday that, while long, basically said "please clarify." It belonged as a comment, so I flagged it as "Not An Answer." No action was taken by whatever moderator looked at it and I lost 10 flag weight for a bad flag.

Today, I hopped into chat.SF and asked one of the mods that hangs out in there what I missed, since I thought it absolutely deserved to be flagged. Another, higher rep, user chimed in and said he agreed with me and flagged it as well. Within minutes the answer was deleted. I still lost the 10 flag weight, though, even though an answer that I had originally flagged was responded to in the end. This ends up being a 20 point swing in weight.

Is this a bug, or just an oversight?

share|improve this question
11  
It's one of the side effects of publishing a number - you start worrying about it needlessly. Yes, sometimes you will have a flag marked bad when it's good, and sometimes the opposite is true. It's not worth worrying about, though. Keep flagging well, and you'll generally go up. –  Adam Davis May 31 '11 at 15:16
5  
@Adam, it directly effects the number of flags that users are allowed to use in a given day, though. It's not an entirely useless number. –  MDMarra May 31 '11 at 15:17
6  
@MarkM True. Do you run out of flags every day? If so, then you'll rocket to the top anyway, even with a few bad calls. If not, then you don't need the additional flags. It's a self-healing system. –  Adam Davis May 31 '11 at 15:19
3  
@Adam - A bug is a bug. It wasn't a "bad call" as you put it. If there were a similar bug where votes/rep weren't being tallied there would be a huge outcry. I realize that a lot of people think flag weight is useless, but if it's published and it effects feature(s) on the site, then I believe that bugs in calculation should be handled appropriately. –  MDMarra May 31 '11 at 15:22
5  
@MarkM This is not a calculation bug. In this case one moderator disagreed with you, and a completely different moderator agreed with you. It's a human issue. It's the same as if your good answer got a downvote that was undeserved. You can't post a complaint here and ask for your two rep back because it's obvious your answer is correct - you got hit by someone's bad vote. It's not an error in the system at all. –  Adam Davis May 31 '11 at 15:24
    
@Adam, Users voting is fundamentally different than mods responding to flags. I think we're just going to have to agree to disagree on this one. –  MDMarra May 31 '11 at 15:26
    
Also, note that it's sometimes a good thing old flags aren't re-scored. Eg, a question might be flagged badly (and marked invalid) and later flagged by someone else for something different (that is valid) - you don't want flag weight then given to the original bad-flagger when the new (good) flag is acted upon. –  DMA57361 May 31 '11 at 15:26
3  
I disagree with your agreeing to disagree. Just to disagree, you see. –  Adam Davis May 31 '11 at 15:28
3  
"Not An Answer" is the most controversial of all flags. Moderator opinion has a high variance on this. If you are awfully concerned about your flag weight, then avoid flagging those. If you feel strongly about it, flag a post with "[x] other" instead and provide an undisputable reasoning as explanation. –  mario May 31 '11 at 15:35
    
@mario, are you suggesting that we do away with the "Not An Answer" flag? –  MDMarra May 31 '11 at 15:38
3  
From my point of view, Adam Davis has the right attitude towards flag weight. If your flags are generally valid, you'll wind up at 500+ weight sooner or later. –  mmyers May 31 '11 at 15:46
1  
@MarkM: if the answer is blatantly not an answer ("Thanks!" "Help!" etc.) then "not an answer" flags are all that are needed - mods can see they're inappropriate with barely a glance. But answers that should be comments are often less obvious. YMMV... –  Shog9 May 31 '11 at 16:00
1  
@mmyers; under that thinking if I shoot you with a gun, and you survice, it's ok because either way you'd end up alive. Just because some outcome is eventually achieves doesn't mean the process isn't broke. –  Chris S May 31 '11 at 17:24
1  
@ChrisS: Oh, I'm sorry, I didn't realize reaching 0 flag weight causes instant death. In that case... I'll be right back. (I do apologize if this was unnecessarily snarky, but I can't actually tell if your comparison of flag weight to bullet wounds is serious.) –  mmyers May 31 '11 at 17:44
2  
@mmyers, perhaps my example was a bit over-dramatic, and your response a bit snarky; but where is the line drawn? Obviously the potential for instant death is over the line; but dinging a (admittedly mostly insignificant) number on the net is ok because it's not common. I think MarkM would be good with the response "It's a flaw, we aren't fixing it because it ain't worth our time; keep calm and carry on", but instead you're justifying it as 'right' because it's insignificant/uncommon. –  Chris S May 31 '11 at 18:04

1 Answer 1

up vote 8 down vote accepted

Is this a bug, or just an oversight?

Neither. The system is working as designed. Moderators are free to ignore (as valid or invalid) or act upon flags at their discretion, and may not always agree with you, each other, or previous or future versions of themselves as to how a flag should be dealt with.

Regardless, the validity of one flag at one point in time should not imply that previously-dismissed flags were valid or invalid. Situations change. Posts change. Perhaps not in this case, but often enough.

I still lost the 10 flag weight, though, even though an answer that I had originally flagged was responded to in the end.

So what? It's just flag weight - it doesn't actually matter, except to the moderators fishing the flags list. Keep flagging things that need to be flagged; you'll get it back.

Also, read this comment posted above:

"Not An Answer" is the most controversial of all flags. Moderator opinion has a high variance on this. If you are awfully concerned about your flag weight, then avoid flagging those. If you feel strongly about it, flag a post with "[x] other" instead and provide an undisputable reasoning as explanation. – mario

"Not an answer" is terse. It better be blatantly obvious that the answer isn't in order for the moderator to do anything about it - we don't need moderators running around deleting perfectly good answers just because someone didn't like them. Obviously, more users flagging will also bring more attention to a specific post - this is also by design.

share|improve this answer
    
If flag weight doesn't matter, why is it there? Isn't it kind of stupid to have something that doesn't matter, needlessly adding complication to the code base? –  Bart Silverstrim May 31 '11 at 18:02
2  
@Bart: it matters to the moderators, or more specifically to the list of flags they see - higher flag weight pushes flagged items higher up the list. Originally, it wasn't shown to users themselves... But folks wanted to know where they stood, so that was changed. Which is fine, so long as you don't obsess about it to the point where you're unhappy when it drops even though your original goal (addressing the problems with a flagged post) was accomplished. –  Shog9 May 31 '11 at 18:10
    
@Shog9, I hardly call this obsessing about it. My questions was "X happened, then Y happened, is this expected behavior in this very uncommon situation." Then it got turned into a debate about flag weight as a whole and my question still remains unanswered. I don't care much about the 20 flag weight, what I do care about is finding out whether or not this is expected behavior. –  MDMarra May 31 '11 at 18:18
    
@Shog9: It sounds like @MarkM was penalized for a flag that was, in the end, deemed valid. He still lost flag weight. I would think it's a bug that should be addressed to keep the math correct in the system, no? –  Bart Silverstrim May 31 '11 at 18:21
    
It may not be a huge or site-crippling bug, but I would think something like this isn't so difficult that it can't be fixed so calculations are kept correct. If it's a trivial feature that mods don't use or take seriously, it makes sense to get rid of it altogether. If it is in use, it makes sense to fix it. –  Bart Silverstrim May 31 '11 at 18:23
    
@MarkM: sorry if I've misjudged your intentions here - if the tiny drop in flag weight doesn't bother you, then good! To answer your question, I would have expected this behavior - assuming your flag was dismissed, the weight should drop, regardless of what happens in the future. Consider a scenario where a post was flagged that didn't require moderator attention, then an argument broke out in comments and it was flagged again - should the fact that it eventually did require moderator attention validate the flag that preceded this situation? I don't think so... –  Shog9 May 31 '11 at 18:27
    
@Shog9 - I think part of what is throwing me off (and I didn't post this, maybe I should edit) is that in the Flagging Summary page, it shows the resolution for my flag as "[diamond] post deleted" like it would for any successful flagging. –  MDMarra May 31 '11 at 18:33
    
@MarkM Any action that occurs after the very first flag on a post is recorded there, not specifically action taken in response to your own flag. As for Shog, while I do agree with the general point of "Once dismissed, that's it", as I've argued that side before, I think the -20 for what should've been -10 may be something to better investigate. –  Grace Note May 31 '11 at 18:34
    
@Bart: ever driven a car with a real-time MPG readout? Makes you a bit paranoid, if you let it... "Oh no, I sped up to pass that truck and my fuel economy dropped by 0.1!" It's still a useful tool when used correctly, but... keep your eyes on the road. –  Shog9 May 31 '11 at 18:36
    
@Grace: I think you're reading the question incorrectly. He said he lost 10 points, for a total 20 point discrepancy (assuming he would have gained 10 points if the flag had been deemed valid). –  Shog9 May 31 '11 at 18:37
    
That could be a possible interpretation of it, and would make sense in which case the system is indeed working correctly. –  Grace Note May 31 '11 at 18:38
    
@Grace, @Shog is right. By 20 point swing I mean that I lost 10 when I expected to gain 10, causing a twenty point discrepancy in what I expected. I did not lose 20 weight total. –  MDMarra May 31 '11 at 18:39
    
@Shog9: Yes, I have that gauge on my car, and it's useful only in that I know it's telling me the current MPG. Perhaps I don't understand the flag weighting system; the equivalent as I see the attitude towards @MarkM's question is, so what if you lost five bucks? It's just five bucks. So what if he lost those weight points? It's just 10 points. Hey...I want my five bucks back, and unfair is unfair. But like I said, I may not understand the flag weighting system. –  Bart Silverstrim May 31 '11 at 19:15
    
@Bart: money is a great analogy... If I look in my pocket and there's $5 missing, I'm concerned - where did that $5 go? Did I drop it somewhere? Did someone steal it? Did I spend it and forget? But when I'm checking my 401K, if it goes up or down by $5... Meh - I'm in it for the long haul, worrying about tiny daily fluctuations is counter-productive. Flag weight should be looked at the same way: if you suddenly start seeing a massive drop, then... Ok, look into that - maybe you're flagging when you should be doing something else. But little drops/increases? Meh. –  Shog9 May 31 '11 at 19:23
    
Isn't there a saying along the lines of "keep track of the pennies and the dollars will take care of themselves?" :) –  MDMarra May 31 '11 at 19:38

You must log in to answer this question.

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