17

I have been researching a specific topic in computer science for a couple of years now and a well established professor and his students have recently published a couple of publications in that topic too. In their work they do reference the well known and well cited previous work that basically everyone in that topic references, but I have noticed that they ignore (don't reference) a couple of publications that are doing essentially what they are publishing about, i.e. extremely related work. And I am wondering even though there is a gap of a couple of years between the work they have published and the available previous literature how did they miss referencing the relevant previous work? Was it done intentionally or did they just do a hasty job at finding more recent related work?

Either way my main concern here is if there is anything that can be done to remedy the situation now, since the paper has already been published?

  • 9
    1) "Was it done intentionally or did they just do a hasty job at finding more recent related work?" - this is a question that only the authors can answer. 2) "Is there anything that can be done to remedy the situation?" - why? Are those ignored papers yours? If so, write to the authors and point them to the relevant works. If not, I'd say there's no point in taking any actions. But if you really want to: write to the authors and point them to the relevant works. – corey979 Jun 17 '18 at 8:37
  • 5
    They can't update the references of a published article. The best they can do is citing those references in the next related paper. – The Doctor Jun 17 '18 at 9:09
  • 15
    So their literature survey was not perfect ... tough, many many papers could show similar omissions. They can’t do much except go forward and neither can you... – Solar Mike Jun 17 '18 at 9:42
  • 18
    Email: "Dear X, I saw your interesting paper on Y. You might be interested in the following paper of mine on Y that addresses a similar question." – Michael Greinecker Jun 17 '18 at 9:42
  • 7
    @Michael Greinecker that would actually be a good answer. – henning Jun 17 '18 at 12:44
35

Realistically, there is nothing you can do to change an already published paper. There are some options for extreme cases, but they are unlikely to apply.

What you can do is try to have the record correct in future publications. Unless you have hard evidence to the contrary (and usually even then), it is best to communicate on the basis that this is not an act of malice. You can write the authors of these papers (and authors of preprints on the topic) a polite email stating that you liked their interesting work and have done related work that you think might be of interest to them. What they do with that information is up to them then.

  • 9
    For the interested, here's a couple of those extreme cases described on Retraction Watch: Example 1, and Example 2. – Anyon Jun 17 '18 at 23:33
12

What do you expect to see in references? Every article on the topic?

References should contain the articles that the article in question actually reference. They publish some results. If these results are in the area that those "well known" articles defined and set up - they reference those. If they use certain results found in other articles - they reference those. But they don't owe you a reference just because you do similar stuff in parallel. If your results weren't used in their work, there's nothing to reference.

If they are writing a survey, they should include every notable result and reference it. If this is the case, you could actually contact them and ask (in a non-aggressive manner) why your work hasn't been included. But everyday articles should just cite the articles that they base the work on.

If you suspect they simply don't know about your work, feel free to write something friendly like "hey, looks like we are working on the same stuff, I noticed that you found Fact1 about Topic1 in Article1, I managed to find Fact2 in Article2. do you think Fact3 could be true? have you managed to do any work on that yet?"

  • 2
    I strongly disagree with the second paragraph. If you are writing a paper about doing X, and there was a publicly available and reviewed paper doing X (or a very closed task X'), then you have to reference and discuss it. You cannot just keep reinventing already known stuff and claim it as new ignoring the past work of everyone else. – Clement C. Jun 18 '18 at 14:03
  • 1
    @ClementC. I disagree. You should only reference what you need/used for your paper. The existence of a paper doing the same thing if anything should merely be reason for an editor to reject the publication of your paper if they agree it is too similar to the previously published work. – guifa Jun 18 '18 at 18:17
  • I couldn't disagree more... I don't know what your field of research is, but in mine omitting citations in the "Relevant work" section is a major sin. – Clement C. Jun 18 '18 at 18:20
  • @ClementC your field has a section of the bibliography called "Relevant Work"? We literally call ours "Work Cited", if we don't cite it, we don't include it. Not having read every single article is common. Even the best Cervantista will find it impossible to have read every single work on Cervantes, so even if you felt citation were mandatory, I wouldn't call the lack of it a sin, because that implies that it is possible to (a) know if every single article in existence on the topic, (b) have access to them all and (c) read them all. – guifa Jun 19 '18 at 7:25
  • 1
    @ClementC. I think we have a different understanding of the problem here.I agree that if you find an algorithm that does X in n^2.15 time you should discuss and compare it to the previous result that same in n^2.18. I read the question more like "we investigated the P=NP problem from the [..] viewpoint. The problem was set up by the famous papers [x] and [y]" and that might not need to reference everyone (or maybe no one at all) working on P=NP in the last few years. – Džuris Jun 19 '18 at 17:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.