Downvotes are a good thing. They separate the good content from the bad, and make a user aware that something is wrong with their question (or answer). Downvoting is an integral part of the Stack Exchange model.
It seems that here on ELL, we're a bit shy with the downvotes (and perhaps a bit too happy to give upvotes that might not be appropriate). To remind everyone what is worth of upvoting and downvoting:
Upvote when:
This question shows research effort; it is useful and clear.
Downvote when:
This question does not show any research effort; it is unclear or not useful.
We, as a community, do not seem to be following these guidelines. Out of 1,111 questions currently posted, only 17 have a negative net-score. I can absolutely guarantee that there are more than 17 low-quality questions on this site that do not display research effort.
We need to be willing to use downvotes when we come across poorly-researched questions that lack context. We also need to stop and think for a minute before we cast upvotes (because we really do seem to have a trend of upvoting just about everything). The main deciding factor in vote-casting is research effort. Is there context? Did the user make an attempt to explain why they don't understand something and exactly what they're looking for? Please pay attention to this and vote accordingly.
If we cast upvotes on everything, votes become essentially useless. Especially as we're still in the defining stage of beta, we very much need to utilize our downvotes to separate the good from the bad. Cast close votes when a question is off-topic or fits one of the other close reasons; cast down votes when a question is on-topic but is a bad question.
In short, let's please try to use our downvotes more often. In a perfect world all the questions posted here would be good, and we'd never need to use them; but we all know that's not the case. If the question shows no effort, downvote it, because downvotes are a good thing.
Please note, I am not saying we shouldn't try to comment and get the questions to improve first. This should absolutely be our first course of action! I'm just saying that downvotes are a useful tool, and that we certainly shouldn't upvote questions which don't show research effort.