This follows on a bit from etiquette (particularly on voting to close). I got a down-vote to my answer to this question.

I should probably make clear at the outset that I really don't care about getting a down-vote, even one without a comment. I'm an adult, I can bear the shame and ignominy.

But I don't think that this is the type of answer that should garner down-votes, at least beyond 0. It was an answer that solved the question in good faith, but because I didn't know about the \middle command, it was not the right answer. The point is that there are so many obscure commands and packages that no one person is going to know all about them. I'd rather people were encouraged to be helpful - as I was trying to be - than to hang back and wait to see if someone knows of a neat command already out there that does the job.

This isn't something to legislate on, but rather something that by our behaviour we can encourage.

(NB Regarding the time-line on that answer; when I started writing it, there were no other answers)

add comment

2 Answers

The way I do things on SO and SU is to downvote things that are actually wrong, don't work, etc. Things that will work, but aren't optimal or have a better way to be accomplished just aren't voted up. Since Google will return the entire page with all of the answers, the best should naturally rise to the top.

I think this is a fair policy –  Juan A. Navarro Jul 28 '10 at 16:15
Yeah, I think it's silly to downvote answers simply for being worse than some other answer. You're supposed to upvote answers that are useful, not also downvote all the others. Don't downvote answers unless they're actively harmful or terrible. (Though of course judgment on the last part is subjective, and you follow your own judgment…) –  ShreevatsaR Jul 28 '10 at 16:34
add comment

I can see your point. But imagine this question gets brought up in google search result 6 months from now. The up/down votes will be a clear indication the user what is "the TeX way to do it" and what is "a hack to bend TeX into doing something".

I'm ok for "Darwin" selection of answers even at the early stage of the site and particular question.

I thought accepting answers is the way to mark the right one. –  Caramdir Jul 28 '10 at 14:47
Accepting answer is a one-jury vote. Where as upvotes mean "public" vote. There is even a "populist" badge - Provided an answer that outscored an accepted answer with 10 votes by 2x –  Dima Jul 28 '10 at 15:08
Such an answer is unlikely to garner any up-votes once a proper answer is in place so I don't think that this would be an issue. –  Andrew Stacey Jul 28 '10 at 15:11
One of the problems is that strange hacks may have already got several up-votes before anyone gives the "right" answer. And it may take a lot of time before the right answer gets much more up-votes than the wrong answers have already got. Especially if there are already lots of strange hacks, people lose interest and might not even bother to check the latest answers carefully. I think that in some cases it might be ok to speed up the ranking process by using down-votes (at least if the wrong answer is by an established user, so that one isn't biting newbies). –  Jukka Suomela Jul 28 '10 at 18:43
Comments can more effectively explain what is wrong with an answer than downvotes. –  Charles Stewart Jul 28 '10 at 18:49
add comment

You must log in to answer this question.

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