tea.mathoverflow.net - Discussion Feed (Constructive downvoting)

Harry Gindi comments on "Constructive downvoting" (5713)

2010-06-01T17:19:25-07:00

Absolutely.

Noah Snyder comments on "Constructive downvoting" (5712)

2010-06-01T16:48:41-07:00

One caveat to this is that once one person has commented giving a reason that it's a bad question/answer then I think it's totally fine for other people to downvote without adding further explanation. (Especially if they upvote that comment.)

Jeremy comments on "Constructive downvoting" (5678)

2010-05-31T17:29:07-07:00

I agree that it seems to encourage bad voting practices. In the link given above, and several other topics on meta stack overflow, people seem to not think it's a good idea because of "retaliation," or because they may not "feel like" giving reasons, or other superficial reasons... But, I think, in the context of math, requiring reasons is a pretty good idea.

I mean, when a friend or colleague tells me something that doesn't make any sense, I real-life down-vote them by explaining why they're wrong. If you submit a paper with errors in it, the referee will tell you what's wrong, they won't just send back an e-mail that says "-1". I don't see why things should be any different here.

I post questions because I want to see constructive feedback, and I read other people's because I want to see constructive discussion of their questions. Votes are useful to me, because when I read an answer for a topic I'm not familiar with, I would like to know what other people, who are familiar with it, think! And, e.g., things "very" wrong to someone who works in number theory, are probably not going to seem so very wrong to me, a physicist, whose background is in geometry. So a down-vote of "-1" on a number theory answer is going to be pretty useless to me without any explanation. And it will help me learn just as much to see *why* their answer was no good, as to see a correct answer.

Harry Gindi comments on "Constructive downvoting" (5673)

2010-05-30T16:20:42-07:00

Anyway, not to rock the boat, but I've recently been the target of such downvoting without comments. The three recent ones that I can remember were on questions that I answered that were followed by another person posting a similar answer. What's funny is, I received votes down (without comments) on a correct answer that was offered first. I suspect that the people responsible for those votes either did not understand my answer (or how it was the same as the other answer), or they decided to vote that way out of spite.

I offer this as anecdotal evidence that there should be reasons given for voting down a post. I think it's pretty irritating to have no explanation, and worse, it seems to encourage bad voting practices.

Harry Gindi comments on "Constructive downvoting" (5668)

2010-05-29T09:26:32-07:00

@Gerald: Upvoting and downvoting serve completely different purposes.

Pete L. Clark comments on "Constructive downvoting" (5666)

2010-05-29T05:31:46-07:00

@Gerald:

"If comments are required on downvotes, then they must also be required on upvotes."

I'm sorry, but this is not obvious to me. Could you sketch the proof?

geraldedgar comments on "Constructive downvoting" (5665)

2010-05-29T05:23:15-07:00

If comments are required on downvotes, then they must also be required on upvotes.

Maybe these "required" comments can be distinguished somehow from normal comments...

Jonas Meyer comments on "Constructive downvoting" (5655)

2010-05-27T19:16:41-07:00

I searched and found a lot of discussion of this on meta.stackoverflow.com. This led to the reminders when downvoting: "Please consider adding a comment if you think this post can be improved." Here's one link to get started:

http://meta.stackoverflow.com/questions/2263/require-comments-on-downvotes

Scott Morrison comments on "Constructive downvoting" (5654)

2010-05-27T16:23:46-07:00

It's not the sort of thing that's going to happen soon, but I'd certainly be happy to force comments with downvotes if we had software support. Is there an existing feature request over at meta.SE?

Pete L. Clark comments on "Constructive downvoting" (5653)

2010-05-27T15:46:36-07:00

@David: it's probably worth a try, at least. Over on Stack Overflow / meta.SE, I believe the prevailing wisdom is that you can force someone to leave a comment but you can't force them to leave a constructive (or even linguistically meaningful) comment and that this approach would lead to lots of spam. I find this slightly pessimistic and would be interested to know what would actually happen.

David Carchedi comments on "Constructive downvoting" (5652)

2010-05-27T13:11:21-07:00

Good point. But I still think the required comment-box is a good idea.

Harald Hanche-Olsen comments on "Constructive downvoting" (5651)

2010-05-27T05:36:40-07:00

@David: The usual mechanism for “undoing” a downvote is to upvote. This is, I think, the essense of what voting is about. I think your proposal for undoing downvotes just complicates things too much.

David Carchedi comments on "Constructive downvoting" (5650)

2010-05-27T05:13:34-07:00

@Jeremy: This is exactly what I was thinking. It would be nice to have a required comment-box (WITH the option to keep the comment anonymous). Perhaps even an option for the community to vote to undo the downvote in the event that the comment's justifications are either mathematically incorrect or, even worse, devoid of mathematical content (for example, merely a comment of style or preference), or simply blank (which is technically a subset of the former). Is this something reasonable for a feature request?

Jeremy comments on "Constructive downvoting" (5649)

2010-05-26T21:41:12-07:00

This kind of annoys me, too. I'll occasionally see things with down-votes that look like perfectly good answers to me, and I don't know if they've voted down because of an error that I don't see, because they aren't as applicable as I think, or someone didn't like them, or what.

It might be nice to *require* comments for down-votes, but leave a box to check to make the comment anonymous if you want.

Mariano comments on "Constructive downvoting" (5648)

2010-05-26T20:04:58-07:00

MO has a TOS?

Harry Gindi comments on "Constructive downvoting" (5647)

2010-05-26T16:00:19-07:00

I completely agree. The argument against it is that people want to preserve their anonymity, but I still wasn't convinced by the previously cited situations in which one would want to do so.

David Carchedi comments on "Constructive downvoting" (5646)

2010-05-26T14:50:31-07:00

Maybe this is something that can be put in the TOS, at least as a suggestion? I mean, I can live with being downvoted with no explanation, but, it just seems really lazy and not much good comes from it, whereas, if there was a comment, it would be quite helpful.

Harald Hanche-Olsen comments on "Constructive downvoting" (5644)

2010-05-26T14:38:53-07:00

I agree. I see the occasional unexplained downvote myself. However, there isn't much we can do about it other than asking everybody to please explain their downvotes. Or at least, upvote a comment explaining the problem with the answer (if there is one).

David Carchedi comments on "Constructive downvoting" (5643)

2010-05-26T14:05:52-07:00

Once in a while, one of my answers will be voted down, but there will be no comment explaining what is wrong with my answer. I would not propose an answer if I did not believe it would be helpful/correct, and I don't think anyone serious in the community would do so either. Wouldn't it be better etiquette to always provide an explanation when downvoting an answer? In particular, the person answering would very much like to know if they made a mistake, what it was. But moreover, everyone who has UPvoted the question might also want to know if it there were problems with it. I try to take the same point of view with students- I don't just tell them they are wrong, but explain to them why and how to fix it. I think such commenting would make the community better. Any thoughts?

-Dave