In the papers I usually read, it is common to present a new idea or an optimization of an existing idea that outperforms the known approches in the field.

Can a paper reporting that a new Idea or method (presented in this latter) doesn't outperform the existing approaches on the same fields be accepted for publication?

I'm asking this because I think that such papers could help the community to know that these approaches cannot improve the current state of the art on the field.

  • 1
    academia.stackexchange.com/questions/732/… may be a better dupe as well. Overall I think simply knowing the term "negative result" will let you find information about what you are asking. – Bryan Krause Aug 22 '18 at 16:35
  • 1
    Also see Can I still try to publish my work if my algorithm's results are not as good as other algorithms'? for a CS version of the question. – Anyon Aug 22 '18 at 18:11
  • 1
    Can it be published? Sure. Will it be easy - difficult to answer... - We have looked at testing a fast method (published) in a more general application and have to deal with a reviewer who believes a computationally significantly more expensive method is the only "right" way of doing the calculation. A method that only academics use for small molecules and nobody else will use because it is not computationally feasible at the moment for large molecules. (If you know computational chemistry, in generally, the better the method, the higher the computational cost, which grows exponentially.) – DetlevCM Aug 23 '18 at 6:57

Browse other questions tagged or ask your own question.