It happens, that users before writing a full answer drop a sketch of their solution in comments. As it is very easy to do (you don't need to care about the details) such comments often appear before any other posts and the following answers are kind of tainted with "you have used my solution" despite the author working on his own.

This happens with answers too, but incomplete posts can be downvoted, and one does not have to click the "$n$ new answers" bar. On the other hand, the comments usually contain some clarification and accidentally reading an answer's sketch in the comment ruins all the fun.

Is this alright? Is there any policy about it? Are there any ways to deal with it?

Edit: Some clarification.

While I approve of giving hints, and understand that sometimes the content of a hint "is not worthy" of being an answer (I do sometimes write such comments), my question is about posting an in-comment sketch with an intention of writing a full answer later.

-

If someone mentions something useful in a comment, that does not preclude you from using it too. Even if you got the idea directly from the comment, you can still use it, as long as you attribute the source. (There is no ethical problem with using other people's ideas; that is called "scholarship". There is only a problems if you use other people's ideas without giving them credit.) If you had the idea independently, you have the option to point out the comment, or not, as you think best. Either way, there is nothing "tainted" about the answer. The purpose of the site is to provide answers to questions; comments are peripheral.

For example, this answer begins with "As Giuseppe Negro said in a comment…" In this case I had the same idea as Giuseppe Negro, but chose to point out that he had said it first.

I also like JSchlather's suggestion of marking one's answer community wiki when it adds nothing substantive to an answer already given. Here are a few examples: 1 2 3

-
Funny coincidence. –  Did Mar 5 '13 at 8:51

I drop hints in the comments when I want the OP to figure out the answer on their own. I believe this is likely to be of more benefit to the OP than reading a ready-made answer. Sometimes this works: the OP solves the problem and then I encourage them to post the solution as an answer. Other times, they remain stuck and an answer is eventually given (by me or someone else).

If someone unfairly accuses you of plagiarism, flag the comment.

-
I think the first sentence of my question was unclear, I added some clarification. –  dtldarek Mar 4 '13 at 19:21
@dtldarek After the edit, I think my answer still stands. I leave a comment with the possibility of writing an answer later (if the comment does not suffice to get the OP unstuck). If someone leaves a comment only to "claim priority" while writing an answer at the same time, that's just sad and pathetic (but not a violation of any rules). –  user53153 Mar 4 '13 at 19:40
I like to (semi-naively) believe that since no one is getting tenure, a grade, or a promotion based on whether they have original, non-copied answers posted to Math.SE, we can all work together to make sure that questions have complete and helpful answers, regardless of who writes what first. –  Todd Wilcox Mar 4 '13 at 20:01
@ToddWilcox But we do get promotions here. For the younger users, these may be some of the more meaningful promotions they ever got. –  user53153 Mar 4 '13 at 20:23
I know I'm greatly looking forward to hitting 10k and getting access to moderator tools ;). –  JSchlather Mar 5 '13 at 2:56
@JSchlather Sure, there are over 30000 deleted posts for you to enjoy. What a hidden treasure, and it will be all yours. –  user53153 Mar 5 '13 at 3:01
I primarily ask questions and the hints/sketches in the comments are helpful because I often continue to work on the problem after posting it and update my work while reading incoming comments and answers. –  GovEcon Mar 5 '13 at 23:30

I don't see how this could this be made policy of any sort, nor for the matter why it should be. With sufficient vigilance, cases of actual plagiarism can be dealt with swiftly. In many cases answers will resemble strongly comments, and other answers; more so for the frequent elementary questions.

When I leave a comment as opposed to an answer, I consider that giving permission for this comment (observation/hint/request for clarification) to be used by anyone. While this is maybe only my understanding, it strikes me as a reasonable interpretation.

As Todd Wilcox remarks in a comment to 5pm's post, I too would like to give the benefit of the doubt to users and believe that, friendly competitiveness aside, most users prefer to cooperate. We should not make such cooperation more difficult by imposing unnecessary rules.

-
"When I leave a comment as opposed to an answer, I consider that giving permission for this comment [...] to be used by anyone." – I cannot agree more with you on that, however, it seems there are those who disagree. –  dtldarek Mar 4 '13 at 20:37

I will add in addition that if one is unsure, one can always post a community wiki answer in the event that you feel much of the credit is due to a commentor. For instance this question is probably in need of such an answer given that it's fully answered in the comments.

-