Please get rid of “vote locking”

Not for the first time, I recently revisited an answer I had upvoted, and discovered a technical error that gave me second thoughts about the upvote. I don't want to mislead new readers into believing I endorse the incorrect answer. But the software will not let me unvote, saying my upvote has been "locked in."

Of course, in the future I should be more diligent in carefully checking all answer before voting on them. But it would also be nice if I were allowed to correct my mistakes.

What is the point of this feature? Another meta thread from years ago (http://meta.math.stackexchange.com/questions/861/locking-in-unvotes) says this behavior is to stop "tactical downvoting" but I don't understand the disease, or why is needs such an annoying cure. I propose removing vote locking from math.stackexchange; what are your opinions?

• There's a dirty workaround for users with reputation $\ge 2000$. Edit, vote. – user2468 Sep 11 '12 at 21:55
• This has been raised previously here and on meta.SO. It was closed as (status-bydesign) before and I don't think the result will be any different this time. – Rahul Sep 11 '12 at 22:20
• Am I missing something? The top three answers, including the accepted one, on the meta.SO are supportive of removing the locking... we can't vote on and set our own policy for math? – user7530 Sep 11 '12 at 22:31
• @user7530: Yes, unfortunately we aren't the ones who decide what gets implemented in the code that runs StackExchange. The users voted on and accepted the answers, but the developers tagged it (status-bydesign), i.e. "not a bug; won't fix". – Rahul Sep 11 '12 at 22:52
• I wanted to add: Please don't take Jennifer Dylan's advice! She emphasized dirty for a reason, we don't want to encourage pointless edits. – Eric Naslund Sep 12 '12 at 1:43
• @Eric I agree that the workaround makes me uncomfortable, and I am reluctant to do it. But surely it's preferable to leaving an upvote on an incorrect answer? – user7530 Sep 12 '12 at 8:26
• Tactical downvoting is explained here. (Remember that if you undo a downvote you get back the 1 rep point cost for doing the downvote.) While it may not be as much of a problem here on Math.SE, it is a genuine problem on bigger sites like StackOverflow. It is not unreasonable for Stack Exchange Inc to only want to maintain one version of the software. – Willie Wong Sep 12 '12 at 8:45
• @WillieWong: The tactical voting in that question is only valid for answers, not for questions. However, the vote-locking feature also exists in questions. – user1729 Sep 12 '12 at 12:43
• @user1729: (a) I neither defend nor oppose vote locking; I see it as already a part of the established system, and I learned to work with it. (b) My comment was to point to an explanation of "tactical voting" which the OP didn't understand, with the caveat that something that doesn't happen on Math.SE may happen on other Stack sites. (c) Ostensibly, the motivation for the OP asking this question is about answers (see the first sentence). (d) You are welcome to debate the pros and cons of the vote locking system; I, however, am not interested in being your opponent (or partner). :-) – Willie Wong Sep 12 '12 at 13:34
• @user7530: There seems to be an easy fix to your problem. One can leave a tactful comment about the possible error. That has far more effect than an anonymous vote, up or down. – André Nicolas Sep 13 '12 at 6:11
• @AndréNicolas That is probably one very good argument for the locking mechanism - if the reason for changing one's mind (and vote) only occurs after some time and second thoughts, then that reason is definitely worth mentioning – Tobias Kienzler Sep 13 '12 at 10:05
• @TobiasKienzler: In that case, one could make the rule that one can unlock the vote for a limited time by making a comment, provided the vote is older than a certain time (to prevent tactical downvoting). If you don't undo your vote in a certain time frame after making the comment (say, five minutes, just as for editing), the comment is assumed to be unrelated to the vote, and the vote gets locked again. If one wants to get extra security against misuse, one could even store the connection to the comment, so if the user ever deletes the comment, the vote gets automatically redone. – celtschk Sep 16 '12 at 14:24
• One could even make the connection between the comment and the un-upvote visible, thus discouraging its use (and especially its misuse), because unlike normal voting, reversal of votes would not be anonymous. And of course one could restrict the feature to a certain minimum reputation. – celtschk Sep 16 '12 at 14:27
• @celtschk That's a nice idea, you should feature-request this at meta.stackoverflow.com – Tobias Kienzler Sep 16 '12 at 17:23
• @Willie, how about allowing to change vote, but with no effect on the voter's reputation after some fixed period? i.e. if you down-vote a question and later decide to take it back, you will still have -1 resulting from the down-vote. – Kaveh Sep 16 '12 at 22:41

I agree, it should be removed from all stackexchanges, not just this one. We should always be able to change our votes if new information comes to light and we realize our previous votes were wrong. Otherwise future readers are mislead by the permanently incorrect votes.

1. Tactical downvoting is an imaginary problem.
2. Even if it were a major problem, vote locking doesn't fix it.
3. Even if it did fix it, the cure is worse than the disease.
• Tactical downvoting of competing answers is not the only form of bad downvoting. There also is a form o revenge voting, given by some of the less mature persons here. And tactical downvoting is real. People have been suspended before because of tactical downvoting. – Michael Greinecker Sep 16 '12 at 18:56
• @MichaelGreinecker: Vote locking doesn't fix tactical downvoting or revenge voting, so I'm not sure why you're bringing that up. – endolith Sep 17 '12 at 0:09
• @Michael: Beware of making innocuous comments like yours here. They have a history of disappearing... – cardinal Sep 17 '12 at 2:13
• @endolith Well, I didn't bring it up. You did. – Michael Greinecker Sep 17 '12 at 8:38
• @MichaelGreinecker: The question mentions that vote locking is supposed to fix the tactical downvoting problem. I explained that it doesn't. It doesn't stop revenge downvoting, either. That's why you lose rep when you downvote answers, to disincentivize revenge downvotes. – endolith Sep 17 '12 at 14:29
• You wrote that "Tactical downvoting is an imaginary problem." I provided evidence to the contrary. – Michael Greinecker Sep 17 '12 at 14:30
• @MichaelGreinecker: What evidence? Seems you provided an example of revenge downvoting, not tactical downvoting. Anyway, I'm sure tactical downvoting exists, but it's not a significant enough problem that it warrants locking votes. – endolith Sep 17 '12 at 14:32
• @cardinal What's wrong with his comment? And why would a perfectly civil, non-spam comment "disappear"? If this kind of suppression is standard practice by the moderators here, I strongly oppose it. – user7530 Sep 17 '12 at 14:47
• @user7530: There is nothing wrong with it (hence, innocuous), as far as I can tell. I am also quite sure Michael knows what I am referring to. :-) – cardinal Sep 17 '12 at 14:48
• @Cardinal: the rest of us don't, unfortunately; I do not have the impression that "this kind of suppression is standard practice by the moderators". If you or other users have the opposite opinion, please do speak up (clearly). Otherwise, I would appreciate it if you can clarify for user7530 that that is not what you had in mind. – Willie Wong Sep 24 '12 at 9:33
• Dear @Willie: I'm very sorry my comment to Michael has caused some unintended consternation; I attempted to make it very oblique so that it would not arouse any outside interest and yet communicate my point to him, but the former intent (at least) seems to have failed. Simply put, I believe the mods on this site do a wonderful job, by and large, with its management. I certainly did not mean to imply that "this kind of suppression is standard practice by the moderators". I am going to flag this answer now and provide a little private elaboration. Cheers. – cardinal Sep 24 '12 at 14:05
• @cardinal: no apologies was necessary. I know from first hand experience that comments on meta can easily spiral out of control when other users attempt to read between the lines into what may or may not be there. I've found it most convenient to just nip those discussions in the bud, in a manner of speaking, but appropriate clarifications. Of course, I also wanted to make sure that if there's anything wrong we the moderators can do something to make it better. Cheers. – Willie Wong Sep 25 '12 at 7:35

Would it be possible to modify the system as follows: there is no vote locking. But if you undo a downvote after the current "lock-in" time period has elapsed, you do not get your -1 reputation back.

Unless I'm overlooking something, I believe this proposed system would deter tactical downvoting as much as the current one (just like it is now, nothing stops you from downvoting competing answers, but you aren't able to later recoup the reputation) while also allowing you to correct voting mistakes.

• +1 Yep, this would have exactly the same effect on tactical downvoting (arguably none), while not causing any collateral damage. – endolith Sep 17 '12 at 0:10
• I have not thought this completely through, but am tempted to suggest taking it one step farther: Apply an additional "surcharge" of one reputation point to remove a previous vote, regardless of whether it was an upvote or downvote. The main disadvantage I can see is that there are instances where accidental voting occurs for technical reasons (e.g., delays in TeX rendering which cause changes in the page flow). – cardinal Sep 17 '12 at 1:48
• @cardinal: At the moment we can remove votes right after we cast them. If removing a vote within this same time window (5 minutes?) remains free, there should be no issue with accidental voting. And removing a vote should also remain possible after edits. – Joonas Ilmavirta Nov 24 '15 at 9:39