I am currently reviewing a conference paper, which I found is just extended version of a previous conference paper. I have heard that a journal paper can be a new version of a conference paper with some new stuff added, while the paper I am reviewing is something like this. Would this paper be rejected simply because of the copyright issue?
It would seem that you should contact the organizing committee of the conference and get their advice. To me it seems kind of low to write papers that have very similar content but it does happen. If you are questioning it that much I would bring it up with the Organizing Committee and see what they say. You don't even have to mention names or specifics but just get their general opinion.
This depends entirely on the conferences involved. Some conferences will allow an overlap of up to 30% with other works (for example), and others require entirely original products. In either event, it is imperative that the earlier paper be referenced in the later one, ideally with an explanation of what is new.
I don't know if I read the question correctly. You are the reviewer? Not the author?
If you are the reviewer, the thing you should do is: Open your e-mail client, write to the jounral/conference editor that asked you to make the review, and tell them that you found this, and ask them if they will even consider it for the publication or not. It would be good to give them a link to the other paper, or send a copy attached (if you're legally allowed to do that).
The only case in which it is intellectually honest for conference paper to be an extended version of another is if:
- There is a major difference in the length of the two papers (e.g., 4-page extended abstract vs. 10-page full paper), and
- The paper declares the relationship up front with a proper citation (e.g., "This paper is an extended version building on the results already presented in [XXX]").
I see the relationship between conference and journal versions as different. In computer science, where the principle is typically "30% new for the journal version," the theory is typically that the journal version is a final and archival version which ties up all of the critical loose ends and inserts all of the portions omitted for space purposes in the conference paper. Thus, it essentially supersedes the conference paper. A second conference paper, on the other hand, should generally present a new piece of work not previously shown.
Thus, for a conference, if both of these criteria are not fulfilled, then you are dealing with a case of salami-slicing and possibly also self-plagiarism and the paper should be rejected.