Sod It

I came here from StackOverflow expecting the same great community.

I have made a sincere effort to participate and witnessed nothing but arrogance.

I'm not just talking about reactions to my own questions, but this one stands as a good example. People don't have to agree with the suggestion, but without having been rude or destructive, I'm blocked from posting. This is what's referred-to in Israel as a "proportionate response."

Enjoy yourselves, I'll make myself useful elsewhere. I no longer care about meta.

So I got into a discussion with someone about editing this crappy question and naturally this issue has been discussed before.

I started wondering: is it worth striving for some kind of consistent response from the community? To have a policy for sheparding these lost souls? I'm starting to think that having some distinctive way to flag this particular problem might be useful.

Motivating observations:

  • random anonymous downvoting doesn't really give feedback,
  • edits make offenders spoilt and lazy,
  • downvotes persist even after a question has been improved,
  • comments can just as easily be irritable, snooty and hostile as they can be helpful.

So, how about having a little 16x16 icon of a mud splat or something similar next to someone's user name if more than, say, 30% of that user's most recent questions have been edited by somebody (or are crappy in some other measurable way or can otherwise be measured to be ugly/childish).

Perhaps when editing someone's question there could be a "[x]This user isn't making an effort to express himself" checkbox to distinguish these edits from other, more benign changes.

It could also appear next to questions that have been edited since the user viewed it -- a signal that it might be worth re-reading and answering.

I expect that this would also cut down on artificially high downvoting on a badly-written question which is otherwise perfectly valid, and reduce the number of unfriendly responses (which bring down the tone of the whole site).

It's automatic and impersonal, removes itself as the offender improves, and has curiosity value: the user will be tempted to click on it and there would be an explanation, perhaps with a couple of their offending questions shown side-by-side in before-and-after fashion.

Good idea?

Another option, threaten the user with closing the question.

Moderators will close this question if you do not make an effort to express it clearly.

Or something like that.

  • 7
    With all the bitching we already have about downvotes and closed questions, just imagine what Meta would look like if we implemented this. – Cody Gray Jul 20 '11 at 14:47
  • 2
    Like a sort of scarlet letter, then? – ale Jul 20 '11 at 15:04
  • Yes, there are lots of crappy questions out there (measured by whatever standard makes them crappy) ...many of us can tell when we are looking at yet another post by someone who just throws up a question like it was their 12th ale in 5 minutes. Their rep will be low, they will have 200 questions, and few (if any) answers. They don't need a 'mud splat' ...those questions will be quickly recognized, flagged, and closed. This process rapidly moves the questions to the back porch, making room for the good questions. – IAbstract Jul 20 '11 at 15:15
  • 1
    Don't take the downvotes personally. Meta is like The Gauntlet; you toss an idea in, and if it survives the shouting and the rifle fire, it might actually be seriously considered. – user102937 Jul 20 '11 at 17:08
  • 1
    @RobertHarvey: And if it doesn't, you may be automatically barred from ever tossing another idea in. – mmyers Jul 20 '11 at 17:10
  • Am I replying to mmyers? Or maybe Robert Harvey? Heck, maybe even spraff. YOU DON'T KNOW – Michael Mrozek Jul 20 '11 at 18:20

So, how about having a little 16x16 icon of a mud splat or something similar next to someone's user name if more than, say, 30% of that user's most recent questions have been edited by somebody (or are crappy in some other measurable way).

How about this, instead:

closed as not a real question by 0A0D, David Heffernan, interjay, Bo Persson, Graviton

Seems to qualify as "something...next to someone's user name if more than, say, 30% of that user's most recent questions...are crappy in some other measurable way".

For when that's not enough, we also have this:

     -5 vote count

We don't really need anything else. If a question is salvageable through editing, and you didn't make those edits, then you're really just as much to blame for the declining quality of the site as the person who originally posted it. At least they might not have known or been able to do any better.

| improve this answer | |
  • 4
    Dunno... a steaming turd overlay on a crappy user's avatar would at least be cathartic... – user1228 Jul 20 '11 at 14:45
  • I don't think that's appropriate. A real question can be asked in an infantile mannar. Perhaps a threat-of-closure message "Moderators will close this question if you do not make an effort to express it clearly." would be do... – spraff Jul 20 '11 at 14:47
  • @spraff: The closure is the threat. Deletion is the next step. It doesn't count if you hide the "real question" where no one can find it or figure it out. And, as I mentioned, if you can figure it out, then you should edit it. No reason for anything else. You already get badges for that. Positive reinforcement works a lot better than negative reinforcement. – Cody Gray Jul 20 '11 at 14:48
  • Also, that -5 doesn't represent the current quality of the question. Editing it has become futile because it will now get ignored. – spraff Jul 20 '11 at 14:49
  • @spraff: I don't know what that means. 1) It's not futile to improve the quality of the site for everyone. Think about future generations! Like me when I'm looking at the page tomorrow. 2) After posts have been edited, downvotes can be removed, and even turned into upvotes. 3) Edits "bump" questions, so others can come along, look at it, wonder why it was downvoted, and cast upvotes of their own, effectively cancelling out all the negative scores. – Cody Gray Jul 20 '11 at 14:50
  • 1) Okay, "futile" is too strong. 2) Repairing the score of a question requires a direct effort. This may not happen. It also interferes with rep asymetrically. Better to avoid the downvote. 3) True, but who looks at a -5 question on the off-chance that it's turned into a good one? – spraff Jul 20 '11 at 14:54
  • 5
    @spraff: So who will look at questions with a "splat" or an steaming turd overlay? Why not just ban that user from posting questions altogether? Oh wait, we already do this. – Cody Gray Jul 20 '11 at 14:55

I reject at least one of your assertions:

  • edits make offenders spoilt and lazy

Can they? Sure. But I wouldn't say it's true 100% of the time or even a vast majority of the time.

There are perfectly valid reasons for a good, contributing user to consistently require edits to their posts. Perhaps English isn't their strong suit. Perhaps they're disabled and the utilities they use to post sometimes result in less-than-perfect posts which regularly require edits. So I don't think it's appropriate to brand someone as "frequently needs edits" as a mark of shame because I don't think it necessarily always is.


So, how about having a little 16x16 icon of a mud splat or something similar next to someone's user name if more than, say, 30% of that user's most recent questions have been edited by somebody (or are crappy in some other measurable way).

The bolded part of your feature request is already implemented. A user's reputation is displayed with their posts. That is the measurable way to tell if their contributions have been crappy.

| improve this answer | |
  • It collapses all dimensions of crappiness into one number. It's a rough indication of quality, intended to be so. Its existence doesn't preclude adding clarity to the situation. – spraff Jul 20 '11 at 14:41
  • @spraff what clarity are you seeking that reputation doesn't give? The temporal relationship (ie, your request states "30% of that user's most recent questions" whereas rep is their overall contributions)? If so, why is that necessarily more relevant/useful? – Daniel DiPaolo Jul 20 '11 at 14:43
  • 1
    "That is the measurable way to tell if their contributions have been crappy.". Well, you need a few downvotes to compensate for even one pity upvote. I've seen so many users with sufficiently high reputation and still a high number of crappy questions. – slhck Jul 20 '11 at 14:45
  • @slhck I'd be interested in seeing if your proposed measure would differ all that wildly from reputation in general, and if it did, which would be a "better" representation of which users are considered better contributors - if you're familiar with data.stackexchange.com you might be able to whip up some numbers to support your position – Daniel DiPaolo Jul 20 '11 at 14:50
  • I didn't actually propose any other measure -- I guess it'd be hard to come up with one. Also, what is crappy and what isn't is a very subjective decision, so probably it'd take a voting rather than just an algorithm to determine that. I could surely dig up some examples, but I guess that every site has users who'd fall into that category. In the end I don't think a measure like this would even be necessary at all. We should act on individual questions by using the up/downvote system and close-votes. – slhck Jul 20 '11 at 14:55

While I can understand the frustration, this isn't something I'd like to see. Some people (sometimes including myself) will make edits for niggling little typos that the OP missed, but don't really make the question impossible to understand (e.g. fixing an incorrect "your" which should be "you're"). Also, something like this could lead to undesirable behavior - e.g. "revenge edits" to SPLAT someone and make them look bad.

| improve this answer | |

It seems like the information that your "splat" would be conveying is already readily available. If I have a suspicion a user has a poor track record, I can just click on their username and get a good idea of their history. I can also track the edits on a post via the Edited link at the bottom.

I won't touch the downvoting issue because that has been done to death. If you mess with people's ability to downvote, it becomes impossible to gauge how the community feels. Just like an actual election, there are going to be trends, there are going to be people taking out their frustration, and there are going to be people voting for Sesame Street characters. Overall, though, the process is successful. I think (and this holds for the commenting and the editing) that you can micromanage the moment-to-moment operations of the site and create an environment that isn't as fun or conducive to helping people.

| improve this answer | |
  • True but you have to have suspicion and the inclination to investigate. An icon makes this immediately accessible, both to you and the offender. – spraff Jul 20 '11 at 14:39
  • 1
    @spraff Yes, but those who get concerned about these matters are more likely to go and look anyway, I would think. Personally, I'd love to have certain indicators be directly observable, but I think there's a tendency to try to keep the UI clean. – jonsca Jul 20 '11 at 14:44

IMHO, this wasn't worth the edits/rewrites. The original was next to meaningless, and the final product (which still sucks) bore almost no resemblance to the original.

That being said, the real difference between your proposal and the current practice is whether or not we consider the user's history when evaluating their posts. What we do now is take everything on a question-by-question basis: Questions are closed if they suck, and the only persistent penalty is the downvotes. You're talking about getting to know the user's past submissions, with escalating penalties for failure to improve.

I don't think this is such a great idea -- partly because it's highly subjective, and partly because SO works very well when questions are evaluated on their own merits. The reaction to a question should reflect only its quality, not who posted it; anything else becomes unfair to everyone involved. If you post a crappy question, I don't care that it's marginally better than your last twelve. I'm going to downvote or edit it, not waste my time picking through your history.

Likewise, if Jon Skeet switches to decaf one day and posts an answer that's just awesome instead of super awesome, we shouldn't downvote it just because he isn't improving. I realize that's not quite what you're proposing, but it's a logical consequence of evaluating the user instead of the post. Once we open the door to that, SO takes another step toward becoming a popularity contest, and downvotes take one more step toward meaning "I hate you" instead of "your question needs work."

What should really happen to questions like this is getting closed as "Not A Real Question." This is much worse than downvotes if you're struggling with a problem. Annoying users might not care about reputation, but failing to get an answer is enough of a penalty to make people try harder, especially if they really needed help with something. If they don't improve, they won't get help, and they'll leave us alone eventually. Problem solved.

| improve this answer | |

You must log in to answer this question.

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