Recently, we had a pretty heated series of discussions about the deletion of old questions that are informative, but would be tossed aside if they were asked today. From this, several software changes were made to SO (and some are still pending), and the moderators adopted some policy changes about how they deal with such questions.

I'm not interested in whether you like the eventual outcome. I don't care about which side of the argument you are on. My question is this:

How do you feel about the process itself? What do you think of the methodology used to come to this conclusion?

Specifically, I would like to know the following:

  1. Do you feel that the process proceeded fairly for all involved? If so, in what ways can you cite where there was clear fairness to one or more sides? If not, then in what ways was the process unfair to one or more sides? Please try to provide specific examples no matter how you answer.

  2. Do you feel that the conclusion and general resolution proceeded logically from the substance of the discussion? That is, do you feel the resolution came from things brought up in the discussion, that the various discussions were moving towards the resolution? Or did the resolution seem to come from out of left field, that the substance of the discussion did not contribute to the outcome?

  3. Do you feel that the conclusion and general resolution was a result to the substance of the discussion or a response to the vehemence of the discussion? That is, do you think that the louder or more prolonged argument win or the better, more effective argument win?

  4. After having been through this process, how has your opinion of Meta Stack Overflow changed? Do you respect the site as a place for making policy more than you did before or less? Or is it the same as before? Explain why in either case. What makes you feel this way? Is it the outcome itself that makes you feel this way, or the process that went into achieving it?

The goal with this is to gauge the community's feelings about the discussion itself.

A reminder: This question is about the process itself, not the conclusion!

  • 1
    I had experienced a lot of rant with no examples/proof whatsoever. And oh, rage quitters... – Tamara Wijsman Mar 8 '12 at 22:05
  • 12
    Why? If you're looking to change the process, Steven Jeuris is like 8 posts ahead of you – Michael Mrozek Mar 8 '12 at 22:13
  • @ConradFrix There isnt a path to there yet. – Naftali aka Neal Mar 8 '12 at 22:15
  • 2
    @MichaelMrozek: I'm looking to do exactly what I said: "gauge the community's feelings about the discussion itself". Making suggestions to change something without knowing how people feel about what they currently have is putting the cart before the horse. – Nicol Bolas Mar 8 '12 at 22:19
  • 6
    To be fair in disclosure, I downvoted and added a vote to close: If you don't already know that for real issues the processes on meta really work for everyone, then you haven't been on meta enough. For the less severe issues, the ones that only the OP feels passionate about, the process also works well, just not the way they want it to work. – jcolebrand Mar 8 '12 at 22:55
  • Somewhat tangential to the issue at hand, but I was actually happy this wasn't a feature request masquerading as a discussion. I also don't see anything particularly wrong in soliciting feedback in this case, personally. – Tim Stone Mar 8 '12 at 23:02
  • 3
    @jcolebrand: "If you don't already know that for real issues the processes on meta really work for everyone, then you haven't been on meta enough." If this is true, then why would you be against a discussion of it? Simply declaring something to be true does not make it true, nor does declaring people who don't believe it to be true to be wrong make it true. If something is true, then it is true, and discussing it will not make it less true. So why the need to downvote and vote to close? – Nicol Bolas Mar 8 '12 at 23:16
  • @MichaelMrozek You might want to point out you should be looking at the most down voted questions there. ;p That would be an interesting feature, arrange be 'unpopularity'. I could use that for my cause. – Steven Jeuris Mar 9 '12 at 0:29
  • 2
    @jcolebrand "then you haven't been on meta enough" ... oh how that is my favorite anti-argument. :) It's an argument from authority. – Steven Jeuris Mar 9 '12 at 0:39
  • 1
    I'm rather tired of reading post-mortems about this. I think it all worked out just fine. My answer to each of your questions would be "yes". Most people are going to say the same thing, or they're going to bitch about how they didn't get what they want. I don't really want to read that. As someone else mentioned, Steven is providing us with plenty of fodder on that front already. – Cody Gray Mar 9 '12 at 2:34
  • 4
    @CodyGray: You don't have to read it if you don't want to. That's no reason to shut down discussion of it. Also, the point isn't that "it all worked out just fine." Just because something worked out doesn't mean that it's good. I don't understand this refusal from some people on MSO to even talk about these sort of things, to do any kind of self-analysis. Well, no, I understand that. But what I don't like is the idea that self-analysis is so wrong that nobody else can do so either. So they go around closing questions for no good reason. – Nicol Bolas Mar 9 '12 at 3:31
  • 6
    @CodyGray: You mean a question tagged "discussion" on a discussion site might solicit... discussion? You could use that excuse for any question with the "discussion" tag. – Nicol Bolas Mar 9 '12 at 3:42
  • 1
    @NicolBolas "So why the need to downvote and vote to close?" because you haven't read either the downvote arrow's hovertext or the reason to close. It is not useful. The question is likely to draw a lot of debate and arguments. Those are my reasons for closing and downvoting. – jcolebrand Mar 9 '12 at 16:02
  • 2
    @jcolebrand: In order to have a reasonable conversation about something, both sides must be willing to: 1) Discuss the matter in a reasonable way, not shut down discussion that they don't like. 2) Present evidence to defend their position, not supposition or speculation. 3) Use reasonable arguments based on evidence, not logical fallacies. 4) To accept the other side's position if sufficient evidence and logical argument is presented for it. Saying, "the system works so long as you understand the system" does none of those. It's a statement of faith, not argument or reason. – Nicol Bolas Mar 9 '12 at 17:12
  • 1
    @jcolebrand: "The question is likely to draw a lot of debate and arguments." So was every question on the deletion issue, and probably 50% of questions on MSO. So again, why single out this question when drawing "debate and arguments" is standard procedure on MSO? – Nicol Bolas Mar 9 '12 at 17:13
  1. Yes. The community got a fairer system for dealing with historical off-topic questions, and the moderators were (happily) released from the burden of having to deal with "this is ancient, off-topic, and should go away, but it will never achieve the needed delete votes" moderator flags.

  2. The solution was carefully considered (my understanding is that it involved pizza and unicorns, at great expense), and took into account the concerns of all involved.

  3. Haters gonna hate. Although some voiced their opinions in a less than civil fashion, their opinion was not discounted, nor was it given undue weight.

  4. I am profoundly impressed with the way this played out. The gears didn't fall out of the universe, the community was heard, and constructive changes were made, in a timely fashion. Compare to the process that the United States Congress uses, if you have any doubts.

  • 1
    "The community got a fairer system..." My question was "Do you feel that the process proceeded fairly for all involved?" I specifically said that it's not about the resolution. I'm not taking a poll on whether you liked the outcome or not; I'm asking how people saw the process that created the outcome. – Nicol Bolas Mar 8 '12 at 23:33
  • 1
    @NicolBolas: If the outcome is fair, isn't the process fair by extension? What about the process would be unfair if the outcome was satisfactory? – user102937 Mar 8 '12 at 23:34
  • 5
    @Robert: "If the outcome is fair, isn't the process fair by extension?" Allow me to make my favorite Babylon 5 quote: "If you do the right thing for the wrong reasons the work becomes corrupted, impure and ultimately self-destructive." Even if good ends come from a bad process, that doesn't absolve the process from being bad. You can't trust a bad process to produce good results, even if it has done so thus far. That's why SO exists: because people saw that forums, which produced OK results, were a bad process, so we developed a better one that produces better results. – Nicol Bolas Mar 8 '12 at 23:38
  • 1
    I'll leave it to others to come up with specific examples of where they thought the process was fair or unfair. My answer seems to adequately state my own opinion; I'd rather not single out anyone else by pointing out specific instances. – user102937 Mar 8 '12 at 23:39
  • 3
    Changes were made because the site blew up and had the second-most daily traffic it has ever had. These issues were not addressed until the team was facing a risk of serious community backlash. I'm happy with the result, but I wouldn't say "impressed". (I think they would have been addressed these issues eventually, but I can understand others' skepticism on that point.) – Jeremy Mar 9 '12 at 0:47
  • 5
    @Jeremy: Do you really think that the traffic was a direct result of people upset about question deletion? Or do you think it was more likely the result of a prominent banner displayed on Stack Overflow that linked to a Meta question explaining why people suddenly lost a lot of rep? In other words, I disagree with the correlation you're asserting. I think that the spike in traffic on Meta would have been there despite deletion of old favorite questions, just because a global rep recalc was performed. – Cody Gray Mar 9 '12 at 2:36
  • I agree with @CodyGray on the traffic assertion. I'd also add that there was a ton of social media traffic and sharing (Google+ and Twitter, specifically) linking to meta, so that probably played a role as well. – Josh Darnell Mar 9 '12 at 2:47
  • 1
    I forgot about the banner; I concede that it may be the cause of most traffic (I had been attributing most of it to Twitter). The point stands that the deletion issue wasn't addressed until there was a mass of traffic, regardless of the cause. – Jeremy Mar 9 '12 at 4:15
  • 4
    FWIW: the vast, vast majority of that traffic was from Stack Overflow itself - I think we can safely blame the banner. – Shog9 Mar 9 '12 at 4:26
  • 2
    @CodyGray: That's a reasonable position. However, the traffic did coincide with the debate itself, which means that people who would never have even known about the debate found their way to it. Whether that actually had an effect on the debate itself or the outcome, I can't say. Someone would have to look through the archives and see if there was a dramatic change in the tone of the debate on or about the 2nd. – Nicol Bolas Mar 9 '12 at 17:24
  • Forgive me, I've been busy with work this week. Would you link to a post describing the actual resolution to the old question deletion process? I had a quick look and I can't find one. – Greg Hewgill Mar 9 '12 at 20:39
  • @GregHewgill: blog.stackoverflow.com/2012/03/… – user102937 Mar 9 '12 at 20:48
  • @RobertHarvey: Thanks much. – Greg Hewgill Mar 9 '12 at 20:49

It was like every other big fight on meta: badly organized.

But that is the inevitable consequence of (1) using a Q&A engine for a discussion and (2) the whole thing being highly asynchronous. There is no getting around that.

Now to address the points as asked...

  1. Obviously "yes" in the sense that everyone had their chance to speak up. Indeed to speak up over and over again if they so desired. Just as much "no" in the sense that crowd behavior can be pretty nasty even when the individual participants are trying to play nice. A lot of people got really hammered with downvotes for asking if there could be some consideration. 'Course, they got at least part of what they wanted in the end, and meta rep is worth less than the usual variety.

  2. The conclusions are within the realm of that supported by the many and varied opinions, which is not to say that there was a firm consensus.

  3. ... Who can say? Just plain shouting won't change minds, but even the best argument won't either unless people see it. You've gotta be at least partly right and loud.

  4. We've been through this before. Maybe not at this intensity, but the essence is the same. Why would my opinion change?


How many "meta's" should be prefixed onto this discussion? ;-)

At any rate, this seems like as good a place as any to toss in a few things I've been thinking about:

  • Most of us are developers and have firm ideas about how things ought to be designed. We know that it's usually a bad idea to listen to users and do exactly what they ask, but we aren't used to being part of the unwashed masses of userdom. MSO can be infuriating as it gives the illusion of being part of the design process, but after a while it becomes clear that important decisions are made elsewhere. An answer that is voted down is sometimes the one that gets implemented. By the way, that's probably for the best—we just need our position clearly spelled out.

  • I'm going to go out on a limb and say the fixes did a good job of compromising between the various positions. The comments on the blog post seem overall positive.

  • Almost all of the issues that caused the perfect storm were raised long ago. Like old sticks of dynamite they were ignored until such time as they blew up. As a developer, I have a feeling I know why: fun tasks, like new features, drive out the boring or difficult tasks, like defining how the deletion process should work. Early on, I remember being annoyed that so many requests were marked as . It's especially troubling when the answer was: "We don't have time to do this right now."

    The result was to establish a culture where important, but uninteresting-at-the-time problems needed to be re-broached over and over, and louder and louder in order to be noticed. (That's probably more of a perception thing than reality. If my ultra-critical problem was ignored last time, I need to make it appear ultra-ultra-critical. In reality, it's just annoying the developers.)

  • The sooner MSE is separated from MSO, the better.

(I didn't really answer your question point, by point. On the other hand, I didn't follow every twist and turn of the process either.)

  1. “Fairly”? What is that supposed to mean? SE isn't supposed to be fair, they're supposed to compromise between what makes them the most money and their conscience telling them to make the Internet better. If they decide that one side of an argument is right, so be it. They are not under any requirement to be fair to the two (or the forty-two) sides.
  2. A lot was thrown around in the discussion, so most of the resolution can be found by assembling bits and pieces.
  3. To some extent, yes, I think shouting louder helps your side. It's not surprising: vocal people do need to be placated.
  4. My opinion hasn't changed, this was nothing exceptional.
  • " so most of the resolution can be found by assembling bits and pieces", good point. – Steven Jeuris Mar 9 '12 at 0:57
  • 2
    I'm really not sure why people disagree with this answer. I've been disagreeing with Gilles a lot recently (quite unusual, I might add), and I see nothing disagreeable here. I'd upvote if I weren't out of votes today. This looks like a rational assessment of the situation, and it thankfully avoids taking the opportunity to rant once again about how the changes made aren't exactly what he would have done. – Cody Gray Mar 9 '12 at 2:40
  • 1
    @Cody: Just guessing, yes, I think shouting louder helps your side. That's not exactly something I'd like to encourage. – sarnold Mar 9 '12 at 2:56
  • 2
    @sarnold: it sounds bad, but... a major goal for Meta is to give folks a place to let their voices be heard. When they actually use it, that's by and large a good thing. – Shog9 Mar 9 '12 at 4:28
  • @Shog9: I've definitely played Squeaky Wheel before when I cared about specific bugs. :) But consider the difference in tone in jalf's answer. The first version was definitely shouting; he got noticed but hurt his cause. The second version is persistent, reasoned, and makes sense. Persistent or impassioned might fit better than shouting. – sarnold Mar 9 '12 at 4:37
  • 3
    @sarnold: I don't disagree, but... There were sooo many problems with the original revision of that answer, it's hard for me to pick on just the tone. dmckee has it right, I think: you've gotta be loud and at least partially right. – Shog9 Mar 9 '12 at 4:40

You must log in to answer this question.

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