20
$\begingroup$

When I start reading a paper, is there some easy way to find a list of corrections for that paper?

For example, it happens occasionally that some result of a paper turns out to be wrong, or at least that the proof is incorrect and not salvageable, and I might not myself easily be able to spot the error (it will after all have been missed by both the original author(s) as well as the reviewers of the paper). And in such a situation, I would hate to end up using such a result in a crucial way only to later discover that it was not correct.

Another thing is that I might have been working on something similar which turns out to be a special case of such a result. In that case, if I am not aware that the proof (or possibly even the stronger result) is incorrect, I might stop working on that particular problem, even if I felt close to a solution, believing that it had already been settled.

In these cases, it would be nice to have some centralized list, where I could find the paper and see if it has had any corrections.

(I had a hard time figuring out what to tag this question, so please feel free to change the tags).

$\endgroup$
9
  • 16
    $\begingroup$ In an ideal world, in which people were not ashamed to confess their mistakes, such corrections would be published in whatever journal published the original paper. In our world there are rumors, gossip, and word of mouth. $\endgroup$ Jun 3, 2013 at 13:10
  • 3
    $\begingroup$ You can also look at arxiv, where many authors publish their papers, and also often publish a corrected version. The link is arxiv.org/archive/math. $\endgroup$ Jun 3, 2013 at 13:44
  • 3
    $\begingroup$ If the paper is published in a journal and has a published correction, sometimes MathSciNet will bundle the review of the original paper and its correction into a single entry; the extreme rarity of this practice is a clue to the extreme rarity of published corrections. Also, if you have downloaded the paper from the arXiv, keep a lookout for later versions on the arXiv or in a journal, either of which is quite likely to contain corrections/revisions/updates of earlier versions. $\endgroup$
    – Lee Mosher
    Jun 3, 2013 at 13:50
  • 6
    $\begingroup$ Some data to support some of the above comments may be found in this recent article in the Notices of the AMS: ams.org/notices/201304/rnoti-p418.pdf $\endgroup$ Jun 3, 2013 at 15:27
  • 5
    $\begingroup$ I'm surprised that nobody suggested asking the author. Assuming the author is still alive, it can be the surest way. $\endgroup$ Jun 4, 2013 at 3:04

3 Answers 3

3
$\begingroup$

I've found more errors in published papers when I went through their proofs, than as published corrections.

$\endgroup$
2
$\begingroup$

If the author finds a mistake (or someone finds and tells to the author) s/he usually publishes a correction in the same journal. If the journal is reviewed by Mathscinet or Zbl, they usually review the correction as well, so you can find it. Papers posted on arxiv are usually corrected, before or after publication. Be sure that you read the latest posted version. Another place to look for corrections is the author's personal web page.

Some authors have Collected Works published. This is another place to look for corrections.

$\endgroup$
3
  • 7
    $\begingroup$ Well, "usually" seems to be a slight exaggeration... $\endgroup$ Jun 4, 2013 at 21:08
  • $\begingroup$ If there is a simple misprint, the journal probably won't publish a correction just for that. (Published corrections are used only for more serious errors.) On my web page, I do list errata for my papers, including even simple misprints. $\endgroup$ Jun 5, 2013 at 12:27
  • $\begingroup$ I also do. I also replace the arxiv version if I find any serious misprint, even after the paper is published. And every time I asked a journal to publish a correction, it did. $\endgroup$ Jun 5, 2013 at 19:56
2
$\begingroup$

You can also try the author's webpages. Below my list of publications, I've included an errata (to which everybody is invited to contribute). Another example is Michael Harris, who has an "annotated publication list".

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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