Indeed. Sadly this story seems so familiar that I can't be very skeptical. Especially the idea of dueling citations based in being in the wrong "camp".
The problem is not the priority, the problem is ignoring the other competing team contribution. This indeed happens all the time.
Since most people get the electronic version, not the printed version, it is possible to ask the authors to update the online journal version to say: "A similar result was done by [0]." This is of course a measure that should be done only in exceptional cases, etc.
Joe Fitzsimons - 2014-09-14 20:18:09-0700 - Updated: 2014-09-15 11:43:51-0700
Sariel, while the final issue seems to be about the refusal not to publish a correction, a significant portion of the post seems to be aimed at establishing priority, and griping about loosing citations that were rightfully his. That's what I mean when I referred to priority disputes.
I also understand why the journal refused to publish a correction: corrections/errata are meant to correct a flaw in the argument in the paper, not to update the bibliography. If I was the editor, I would have responded almost identically, despite not knowing anyone involved, so I find the conspiracy theories deeply troubling. There is a very real worry that allowing a third party to force an errata just to include a reference, without a flaw in the argument, is an extremely dangerous precedent.
Like it or not, for every paper out there there are authors of other papers who feel aggrieved not to be cited.
I am not blind to the frustration here, I would estimate that it happens regularly to maybe 15-20% of my papers. However, sometimes you just have to live with the fact that people are sometimes disengenuous, and others simply fail to do the necessary background reading, relying instead on later papers to summarize earlier ones in their litt review.
Of course this is annoying to the original authors, but this does not necessarily imply that they should be able to do anything about it.
+Joe Fitzsimons "Of course this is annoying to the original authors, but this does not necessarily imply that they should be able to do anything about it."
Each person has to make a judgement based on their own case but a universal recommendation to be passive about infractions such as the one outlined by Pak seems unwarranted.
The journal has an obligation to maintain scientific accuracy and it seems that it has been violated in this case. If I have an algorithm/theorem that has already been published several years ago then not citing the prior work is some thing that the journal should help correct. At least that is my opinion.
Joe Fitzsimons - 2014-09-15 07:52:21-0700 - Updated: 2014-09-15 07:53:59-0700
Chandra: I think we disagree on whether it has been violated. A missing citation would not normally be cause for a formal correction, and as I mention, I would be wary about setting a precedent. I agree that the journal has an obligation to maintain the correctness of the paper, but minor errors rarely result in formal corrections, and this case does not concern the correctness of the results. I think it is worth keeping in mind that almost every paper will have multiple minor flaws, and no journal is going to accept that number of changes.
The crucial point I don't understand (since I don't know anything about the subject area) is this: does the 2013 paper prove anything new, or is its entire contribution subsumed by Pak's 2000 paper? If the former, then I think Pak's complaint is way over the top. If the latter, then the 2013 paper should be withdrawn. At most, they can publish it to say: "Here's another approach to the problem solved by the 2000 paper. Our techniques are different, and of independent interest."
The conspiracy theories shine a very unfavorable light on Pak. He is not-so-subtly implying that Niemeyer's career was significantly helped along by her friendship with Hiss. This is libel.
Finally, how significant is this result? He claims it's technically elementary. Then why all the drama? Is it worth slinging mud in public, even if he is right? 'Cause frankly, he comes out pretty muddy himself.
+Mikhail Belkin No, of course it's not ok to claim something false. However, I judge small inaccuracies much less harshly than gross errors. Let me elaborate what I meant in my first comment: If the 2000 paper showed how to deal with the unknown degree n only for some niche cases (say, n <= 10), then the 2013 paper is only guilty of small inaccuracies, and Pak's gripe amounts to, "But I had some result in this direction, too! Cite me! Cite me! How dare you not cite me?" Otoh, if the 2000 paper deals with any n, then the 2013 paper doesn't prove anything new, and should be withdrawn. In that case, Pak's complaint is entirely justified, albeit very inelegantly stated.
Magda: It is hard to disagree with you on these points but you are considering two extreme cases. Sure if there was nothing new in the 2013 paper it should be withdrawn. If Pak's result was trivial or extremely limited, there was no reason to cite it.
However, I don't think any of the parties is making either of these claims. In fact (from reading the correspondence) everyone seems to agree that there was novelty in the 2013 paper yet Pak's result was non-trivial and should not have been ignored.
In view of this, updating the arxiv/publishing an erratum would seem pretty reasonable course of action.
Personally, when reviewing a paper I would generally interpret a claim of novelty stated in the abstract as one of key contributions of the paper from the authors' point of view.
Since most people get the electronic version, not the printed version, it is possible to ask the authors to update the online journal version to say: "A similar result was done by [0]." This is of course a measure that should be done only in exceptional cases, etc.
I also understand why the journal refused to publish a correction: corrections/errata are meant to correct a flaw in the argument in the paper, not to update the bibliography. If I was the editor, I would have responded almost identically, despite not knowing anyone involved, so I find the conspiracy theories deeply troubling. There is a very real worry that allowing a third party to force an errata just to include a reference, without a flaw in the argument, is an extremely dangerous precedent.
Like it or not, for every paper out there there are authors of other papers who feel aggrieved not to be cited.
I am not blind to the frustration here, I would estimate that it happens regularly to maybe 15-20% of my papers. However, sometimes you just have to live with the fact that people are sometimes disengenuous, and others simply fail to do the necessary background reading, relying instead on later papers to summarize earlier ones in their litt review.
Of course this is annoying to the original authors, but this does not necessarily imply that they should be able to do anything about it.
Each person has to make a judgement based on their own case but a universal recommendation to be passive about infractions such as the one outlined by Pak seems unwarranted.
The conspiracy theories shine a very unfavorable light on Pak. He is not-so-subtly implying that Niemeyer's career was significantly helped along by her friendship with Hiss. This is libel.
Finally, how significant is this result? He claims it's technically elementary. Then why all the drama? Is it worth slinging mud in public, even if he is right? 'Cause frankly, he comes out pretty muddy himself.
"This eliminates the major gap in known algorithms, as they require the degree as an additional input."
Presumably this claim was important in getting the paper accepted in the first place.
Surely you would not suggest that the authors can claim anything, whether true or false, as long as there is some novelty in the paper?
However, I don't think any of the parties is making either of these claims. In fact (from reading the correspondence) everyone seems to agree that there was novelty in the 2013 paper yet Pak's result was non-trivial and should not have been ignored.
In view of this, updating the arxiv/publishing an erratum would seem pretty reasonable course of action.
Personally, when reviewing a paper I would generally interpret a claim of novelty stated in the abstract as one of key contributions of the paper from the authors' point of view.