What to do with a result too small to be publishable, but interesting nonetheless?

Question

Let's assume, for the sake of example, that I have found a new figure of merit which is of the same complexity as the one currently used, but is much more justified from a mathematical and physical standpoint. However, the whole development barely goes over one paragraph (maybe two), as the "proof" is extremely simple and straightforward.

Now, I was planning on saving this smaller result for a bigger paper on the same topic, where I'd introduce this new figure of merit along with other results, but the future of this other paper has become quite uncertain lately.

Is there any way I could still present this new (small) result?

There is a website somewhere where you can publish single figures, complete with authors and DOI and fully citable. If I remember the name I will post it as an answer. — gerrit, Sep 30 '14 at 15:29
I'm afraid I can't find it. I'm quite sure I learned about it either on [Academia.SE] or on [OpenData.SE], so perhaps someone else can. What I have in mind is basically an online-only "journal" that publishes single science-quality figures (with captions), complete with doi, list of authors, etc. — gerrit, Sep 30 '14 at 15:38
I have a couple of small things sitting on my webpage. One of them has been cited by two sets of authors (unlike my papers, sigh). One of them is also on the arXiv. — Jessica B, Sep 30 '14 at 18:34
@Michael Yes, I do! You could add that as an answer, as I think it might fit user8001's needs. — gerrit, Oct 01 '14 at 21:36
File it away in the interesting results tab of your file cabinet and move on or investigate more. Maybe in many years you or someone will have some results you cannot explain and the results from this earlier experiment will enter your mind and progress may be made. I believe Fleming merely mentioned the properties of what turned out to be Penicillin as something interesting but an "aside" it his early publication(s). I like the posting in on arXiv comment as well to take ownership should something develop from it. — If you do not know- just GIS, Jul 18 '16 at 01:55

Massimo Ortolano · Answer 1 · 2014-09-30T13:25:21.977

25

You can think of publishing it as a short communication, many journals accept this kind of contribution (they can be designated in different ways, depending on the journal). As I recall, the shortest article ever published was of about 3 lines [1] ;-)

[1] F. Lenz, "The Ratio of Proton and Electron Masses", Phys. Rev. 82, 554, 1951.

edited Sep 30 '14 at 13:25

answered Sep 30 '14 at 12:57

Massimo Ortolano

55,428
19
166
207

[...] the shortest article ever published was of about 3 lines Interesting. Can you provide any reference? – enthu Sep 30 '14 at 13:20
@EnthusiasticStudent: I've added a reference. I've found a picture of it, but adding this would probably result in copyright infringment. Actually, I'm not sure it's the shortest, but it would be surely difficult to write a shorter one. – Massimo Ortolano Sep 30 '14 at 13:27
2

Some other examples of short papers: http://mathoverflow.net/questions/7330/which-math-paper-maximizes-the-ratio-importance-length (the list includes papers of length 6 lines and 9 lines.) – Aru Ray Sep 30 '14 at 13:28
26

Nope. Shortest paper is The unsuccessful self-treatment of a case of “writer's block” – ff524 Sep 30 '14 at 13:28
1

@ff524 Wow! How is it a paper?! – enthu Sep 30 '14 at 13:33
@ff524: Lol. Definitely unbeatable, if we don't count note 1 :-) – Massimo Ortolano Sep 30 '14 at 13:34
2

Guys, they've only used lemon juice and x-rays for the initial review. We can do mass spec, smoke signals, and MRIs in our review and publish our findings 41 years after the initial report. – Compass Sep 30 '14 at 13:36
2

In many physics publication these go by the name "letters". Indeed there are Journals---such as Physical Review Letters---devoted entirely to them, and many other journals also accept them. PRL is notable for it's short time to press. – dmckee --- ex-moderator kitten Sep 30 '14 at 18:06
5

On a related note, this is the shortest abstract I know of: http://arxiv.org/abs/1110.2832. That's a grand total of two words! – terdon Sep 30 '14 at 22:05
@Compass it already has been replicated in a multi site study http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2078566/?report=classic – StrongBad Oct 01 '14 at 22:21

R.M. · Accepted Answer · 2014-10-01T17:19:49.467

The reason why you're publishing is because you think you have a new way of doing things that is better than the old way of doing things. Simply publishing a derivation of the new way isn't sufficient to convince people of that fact. You need to convince people that your new way is better.

So yes, your derivation is a simple one to two paragraphs, but in addition to that you should spend several paragraphs explaining why your new way is better. Why was the old way used? Why weren't the deficiencies not considered a problem until now? (Or if they were, how were they addressed with the old way, and why is your way better?) What are some of the objections that the people used to doing it the old way will raise? Why might they not want to switch to your new way? What do you say to those potential objections? Are there possible use cases might your new way not be appropriate for, and if so, how do you choose between the old way and the new way?

You'll probably also want to spend several paragraphs in an introduction, explaining how the field got to the old way, and outlining the deficiencies. You probably also want to include and example of using your new way, comparing it to the old way, illustrating why the new way is better.

Basically, you think your paper is too short because you're planning on just presenting the derivation and assuming its merits are obvious to others. They won't be. In all likelihood you'd be chewed up in review - not because your paper is too short, but because you won't have convinced the reviewers that your new method is worthwhile. Fleshing out the paper with necessary background and discussion to presents a coherent argument for your new method can easily take you into the 1-2 page range which is the typical size of "brief communications" (depending on field, journal, etc.)

Don't pad your article just to increase its size, but don't omit things which will bolster the arguments in favor of your new method.

Edit: In response to comments, here's an attempt at clarification: An academic paper is not just about presenting results. It's also about presenting a story and an argument. How does this result change the field? How should it change how the readers think about the topic?

I'm guessing that the original questioner thinks their paper is only one to two paragraphs because they were anticipating presenting just the short proof/derivation of the new figure of merit. My point was that the proof itself is insufficient for a decent paper. In addition to the proof you also need to convince the reader that the new figure of merit is better than the old one - and to do so for people who may be unaware that there was anything wrong with the existing way in the first place. Doing this properly can easily extend the paper from 1-2 paragraphs into a size which is more typical for a standard journal article. Write a proper paper, and it's no longer "too small to be publishable".

@Mindwin - If the edit is not sufficient, please elaborate on the area of your confusion. — R.M., Oct 01 '14 at 17:20
I will blame language barrier. But now I see you are correct in saying that communication is about what the listener understands, not what the speaker tries to say. What is obvious to Archimedes won't be so to Erasthothenes. +1'd — Mindwin Remember Monica, Oct 01 '14 at 17:56

score 2 · Answer 3 · answered Sep 30 '14 at 13:25

2

I would submit the manuscript as a short paper, or even a poster in a workshop. Workshop papers and posters in my domain (computer science) usually demonstrate more early results of ongoing research.

answered Sep 30 '14 at 13:25

george

1,607
13
18

2

A poster will only solve half the problem, though, won't it? Anyone not at the workshop won't be able to find out about it, and there won't be a permanent document that a future researcher could read and cite. Maybe poster + arXiv preprint, or poster + publish in some journal? – Nate Eldredge Sep 30 '14 at 16:40
@Nate Eldredge, regarding posters, I have in mind reputable conferences that publish posters online and in conference proceedings. See for instance the posters here, 2-page publications in fact, which can be read and cited: http://www2013.wwwconference.org/papers/companion.htm#P – george Sep 30 '14 at 17:40
3

Oh, I see. Thanks. In my field a poster is literally just a poster and there is no permanent record of its content. – Nate Eldredge Sep 30 '14 at 18:35

What to do with a result too small to be publishable, but interesting nonetheless?

3 Answers3

Linked