I need to know what it means​ when a paper starts with: “Revisiting [topic]”

I read enough on a topic to write a survey paper, and I want to publish it in a conference to get feedback and discuss it with the relevant community. I want to publish it as soon as possible, because if I don’t, somebody else will.

However, it might be the case that in the future, I can find more relevant information and my insights will develop (as I will continue researching the same topic in the future), which might extend the survey I propose. Is it okay to publish another ‘revisiting’ paper with an extended survey? Or does this leave a bad mark on the publication and the researcher? I do not want to delay publishing because it may be hard to reach a point where I am 100% satisfied with my knowledge of the topic. But I might be wrong and I should publish only when I reach this stage.

  • 2
    Research is rarely the final say. If you wait for a paper to be perfect you may never publish. IMHO, revisiting usually applies to something that got attention in the past and then was quiescent. For instance, I once published on compressing memory pages to fit more application memory in DRAM not disk. Wasn't the first and very much not the last: the same conference seems to publish a new paper on the topic about once every 5 years. Mar 26, 2017 at 14:00
  • 1
    I don't know if I'd say "bad" but much hinges on the sense of revisit (and your definition of bad). If revisit means to repeat the paper for no reason just to reminisce, then yes that's bad. If "revisit" means things have changed surrounding the experiment or thesis that dramatically reshape the claims it makes, then that could be "good" (depending on what you mean by good)
    – virmaior
    Apr 5, 2017 at 14:07

1 Answer 1


A survey paper shouldn't just be the result of 'I did lots of reading'. There are two key types of survey I've come across. One is when some major result has been proved, and so it is time to do a 'round up' drawing together all the different ideas that go into it and with it. The second is when you've done lots of reading and have something useful to add as a result (ie the paper contains new research, on some level, as well as survey material).

I would say the issue isn't whether you 'revisit' something, but rather whether you try to publish work that doesn't add enough to the literature. No-one is going to really mind if you freely release an extended paper as an eprint for those who want to read it. On the other hand, trying to get published two separate papers that are nearly the same is not going to look so good.

You must log in to answer this question.

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