Suppose I am improving a bound in a problem in "some cases" by using the extra information available in that problem which was unused in the earlier bounds for that problem. However, in general, I am not able to show that my bound is better than the earlier bound. Can this result be considered to be a reasonable result?

  • this question is off topic here as it relates to a specific research problem. Try the math forum
    – Buffy
    Jan 5, 2019 at 12:12
  • 1
    There is no specific problem written out here, as far as I notice.
    – Tommi
    Jan 5, 2019 at 13:07
  • @TommiBrander Based on your profile, your research interests include PDE. This question may sound non-specific to you. But, can you imagine an English Literature professor would be able to answer the question?
    – Nobody
    Jan 6, 2019 at 6:47
  • 1
    @scaaahu No, presumably an professor in a field distant from maths could not answer the question. But there are many other such questions on the website. This is not a question about the content of research (how do I prove this or that bound). This is a question about academic writing and publishing. Maybe there is a meta-thread against this question, but at least the help center contains nothing about this question being off-topic.
    – Tommi
    Jan 6, 2019 at 8:12
  • 1
    I agree with @TommiBrander - asking whether improving a bound is generally publishable is not about a specific research problem. Actually, such a question would be applicable to some maths-related fields too. Moreover, we seem to allow analogous questions about algorithms, which a literature professor probably wouldn't be able to answer either.
    – Anyon
    Jan 6, 2019 at 17:18

1 Answer 1


The result might or might not be reasonable. The following would speak for it being interesting:

  1. The special case is interesting. A good argument for this is that others have looked into it or that it has applications or other relevance in other fields of science or mathematics.

  2. The method of proof is new.

But, in general, the only way to know is to write it down, talk about it and submit it to a journal and see what the response is.

When writing the result down, you will want to emphasize the relevance of the result and refer to similar results, applications and the new features in the proof.

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