Timeline for How much is simplicity of ideas/approaches in a paper appreciated by journals and reviewers?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 17 at 12:41 | comment | added | DCTLib | @zx-81 I am not sure if really many authors present their results in a too-complicated way. But in some case, the temptation is definitely there. I once got three reviews for a paper that all said "Wow, these are strong results, funny that nobody did it before, but your proofs are too simple -- reject". The paper ended as a tech report because I didn't find a solution that I didn't hate. Still, I believe that the problem is best dealt with by side-stepping, for instance by one of the approaches in the answer. I've seen many papers in CS that look like approach A from my answer was used. | |
| Jun 17 at 10:22 | comment | added | zx-81 | "likely to recommend rejection when it looks like there is no deep insight in your contribution, and they may use the complexity of it as proxy for the amount of insight." Yes, exactly my point. So I think many authors tend to publish more-than-neccessarily complex exposition of their ideas just to be on the safe side and avoid the risk of "sounding too trivial". This is why I really like the "previous work" sections where authors describe the substance of ideas of OTHER other authors often in very simple and understandable terms. | |
| Jun 17 at 10:12 | vote | accept | zx-81 | ||
| Jun 17 at 10:11 | comment | added | zx-81 | Thanks. This is a very heplful answer, exactly to the point of my question. Especially this advice: "Formalize the key insight behind your new approach, and add this as theoretical analysis. If you keep the paper in a way that the paper is still readable without the theoretical analysis, it's still a helpful paper even for those who want to skip the theory" | |
| Jun 17 at 8:53 | history | answered | DCTLib | CC BY-SA 4.0 |