What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 129 Stack Exchange communities.

Sometimes you just happen to read an answer in a topic in which you are not an expert in, and it sounds convincing enough, it shows careful effort and attention to details in the writing, and the author seems knowledgeable enough about the subject (see notes below), but this is simply not your area of expertise and you can't honestly say if it's correct or not.

What do you do? Do you upvote?

Note: there's a distantly related phenomenon of upvoting because of author's high rep (e.g. Upvoting based on celebrity, too many upvotes (“excessive rep slurping”) and The Problem with Reputation: Does High Reputation Attract Too Many ‘Up-Votes’?). This scenario is different in that while a person's rep may play a factor, it's only a supporting one.

Ultimately the perceived high quality of the answer itself is why you think it may deserve an upvote, and the problem of whether or not it's responsible for you to do it is there even if the person has a low rep.

To complicate matters, and to anticipate the arguments that votes are based on usefulness not correctness, let's say that the answer is pretty much useless for you. Perhaps it's related to an obscure language that you don't plan on picking up any time soon, or perhaps it's an area not in your career direction at the moment, etc.

...and yet you somehow saw this question/answer. Hey, these things happen.

So the answer is useless, and you don't know if it's correct or not, and yet somehow (based on subjective factors, perhaps) you think maybe it deserves an upvote. Is it responsible to do so?

share|improve this question
no freehand circles? –  alexanderpas Aug 5 '10 at 21:23

9 Answers 9

If the answer is primarily technical (and not architectural) you should not be upvoting without knowledge. It is entirely possible to write a programming language where 2 + 2 == 4 would come out false. Why you would do that, I don't know, but you could. So if you have no technical knowledge of the area, even intuitively correct answers could turn out to be wrong, or worse, dangerous. (Encryption anyone?)

If the question is more subjective or architectural, and well-explained, this is to say, it is so well explained that even you can understand it then yes, I would say that you could (cautiously) upvote it. Well-explained answers are nearly always helpful, even if they don't work. This is because the thought process that goes into solving a problem is often enlightening, and users who have thought out and explained (put effort into) their answers are much more likely to come back and engage the OP on the occasions that they happen to be wrong.

Bottom line: If you don't know, Don't Vote.

share|improve this answer
+1 But I think the problem is beyond resolution. –  nb69307 Aug 5 '10 at 21:02

Ask yourself - If I only had one upvote to spend during the day, would I upvote this?

share|improve this answer
Ironically, I see three people may have already ignored my advice! –  nb69307 Aug 5 '10 at 22:08
No. It was sarcastically. –  Peter Ajtai Aug 5 '10 at 22:19
Double-ironically, nobody would notice your advice if everybody would follow it. In which case it was redundant. Which it isn't apparently. –  Nordic Mainframe Aug 5 '10 at 23:03
I would. ( +1 ) –  Pavel Shved Aug 6 '10 at 18:00

The old ethics test works here:

What would happen if everyone else behaved exactly like you?

If the answer, as it is in this case, is that the system would break, or be adversely affected, then you picked the wrong action.

Imagine there are two answers to a question:

  • One is right, but poorly (albeit clearly) worded.
  • The other is a well-written, thoughtful, well-intentioned response that happens to be factually incorrect (or useless to the OP, which is essentially the same thing).
  • Now assume that everyone who votes is like you, in that they don't know the answer, but vote for the clearly well-intentioned response, since it looks like it's probably right, and they don't know that it's not.
  • None of the voters in this world vote for the right answer, since none of them know that it's right, and it doesn't reflect the same thoughtfulness as the wrong one.

The outcome is that you have a wrong (or useless) response with a ton of votes on top, a correct answer with no votes buried below, and an OP and a bunch of visitors either doing something wrong, or thinking that the site is full of know-nothings.

If you can't tell that an answer is either correct or useful, up-voting it undermines the quality of answers on the site.

share|improve this answer
@Peter, I'm hoping you mean that this sometimes happens now, rather than that it usually does? But in any case, the question is really whether you want it to work that way. Assuming you don't, you shouldn't upvote an answer unless you have reasonable grounds to believe that it's correct/useful. –  Jaydles Aug 5 '10 at 21:02
(yes, edited) - This definitely happens sometimes. There are some clearly incorrect answers that get upvoted. It happens rarely, but it's very frustrating, since I feel like there's almost a mob mentality behind some of the surges of upvotes. –  Peter Ajtai Aug 5 '10 at 21:17

I vote only on topics in which I am somewhat versed and only if I think that the answer deserves more votes than it has already. If I'm in a callous mood, then I downvote answers with too many votes. I'm not very forgiving with errors or inaccuracies in answers by high ranked members. I assume that an novice member might have problems to communicate their thoughts due to a lack of experience or language barrieres (I bet many people play here to improve their communication skills), but a senior should be a role model in communication style and precision. People like me well may be the necessary counterweight to the celebrity voting the OP mentioned.

share|improve this answer
Is this intended to be ironic? It seems to reflect an low tolerance for posts that fail to exemplify good "communication style and precision" in a post containing typos and grammar issues. –  Jaydles Aug 5 '10 at 20:54
Because I am one of these people who are here to improve their communication style (and skills). English is not my first language, not even my second. –  Nordic Mainframe Aug 5 '10 at 21:00
Also I am dyslectik –  Nordic Mainframe Aug 5 '10 at 21:03
no disrespect or judgement of your English intended; I have no qualms with posts from individuals who are doing their best with a non-primary language. Bit it seems like you're judging others on their communication challenges. Put another way, based on your own standards, it sounds like you don't think you deserve to be a "senior" (which I assume to equate to "high-rep") member someday, and I'd respectfully disagree. –  Jaydles Aug 5 '10 at 21:10
@jaydles: [...]it sounds like you don't think you deserve to be a "senior" [...] member someday. With my current shortcomings, no :-( I tend to provoke misunderstandings and this is all bad. OTOH, this is my first month on SO and I'll see where I am next year. –  Nordic Mainframe Aug 5 '10 at 21:31

I think you should not vote if you don't understand both the answer and what the OP is asking.

Answers must be both accurate and precise. There is a set of accurate answers that get large numbers of up votes, despite the fact that they are not precise. In other words the answer is correct in its technical details, but it doesn't answer the question.

This, I believe, happens because many people vote without careful consideration. Sometimes they vote without technical knowledge, but more often they skim the question and read to the first answer that is not a syntax error, and upvote it.

Responsible upvoting means reading and understanding the question and the answer.

This answer prompted me to write the above. The answer is correct, but doesn't really answer the question. The question isn't phrased in the clearest manner, but it can be understood. The only reason that answer got so many upvotes was that people weren't upvoting responsibly. People were just acknowledging that range() can be a useful function in PHP.

To be honest, I upvoted that answer too. It took a comment from the OP for me to realize that they probably was looking for something else (Fair warning, that's my answer, but I think it's a legitimate illustration of this phenomenon)

So, it's really easy to upvote irresponsibly. You see a pretty answer, and you upvote it. It takes much more time to understand not just the technical details, but the context of the question. Especially since answers are often written in a clearer style than questions. There are definitely many beautiful questions on this site, but there are also many badly composed questions.

Finally, it can be said that accurate answers which are precisely NOT what the OP was asking for are still useful to someone, so they deserve an upvote, but I think the site would be much more useful and amenable to search if people, myself included, just slowed down and gave slightly more careful consideration to the upvote.

share|improve this answer

The responsible thing to do is to never betray your own opinion with your votes.

More than just being an indicator of what is useful, your votes are an indicator of what you think is useful. If you think an answer deserves your upvote, then it does. If you are indecisive or uncomfortable because of unfamiliarity, then withhold your vote until you decide.

Upvoting what others may skip or downvote is not irresponsible unless you're actually upvoting in bad faith.

share|improve this answer

Check comments to the answer, and if there's no negative responses from the users who are active in the relevant tags, upvote. Effort should be rewarded, and some domains just don't have enough people to reward good answers well enough.

share|improve this answer
Well, that breaks if none of the users that could spot errors haven't been around yet - it happens often enough to matter in e.g. [c++] and [c++]+[templates]. The absence of criticism isn't proof of an answer being correct. –  Georg Fritzsche Aug 5 '10 at 20:43

Sometimes you just happen to read an answer in a topic in which you are not an expert in, and it sounds convincing enough, it shows careful effort and attention to details in the writing, and the author seems knowledgeable enough about the subject (see notes below), but this is simply not your area of expertise and you can't honestly say if it's correct or not.

What do you do? Do you upvote?

Apparently, people do. For example, the accepted answer to this question has 33 upvotes, yet the main point about zero-sized arrays is completely wrong.

share|improve this answer

In spite of the drumbeat of encouragement to 'vote early, vote often,' I counsel erring on the side of zero.

First of all, nearly all questions deserve a zero. They present one person's particular problem in a readable form. Fine. 0. To deserve +1, they should be very well written, or manage to capture a very common problem in such a way that just reading the question will help other people navigate the space.

Of course, many questions deserve a -1, due to illiteracy, incompleteness, or presumption (plz snd me the codez).

On the answer side, a quick answer that contributes a useful fact deserves a 0. If it happens to hit the OP's problem on the head, and the OP votes it up once, well, fine. It doesn't need 2, or 27 me-too votes.

share|improve this answer

You must log in to answer this question.

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