96

I have recently stumbled upon this paper on arXiv: Stopping GAN Violence: Generative Unadversarial Networks (mirror), which is a parody/joke paper.

Is it ok to upload joke papers to arXiv?

8
  • 34
    viXra would be a better place for joke papers. Mar 12, 2017 at 0:08
  • 24
    I think it is okay when the paper is extremely funny, like the one you have linked.
    – Avery
    Mar 12, 2017 at 1:43
  • 11
    OK, according to whom?
    – Mad Jack
    Mar 12, 2017 at 2:28
  • 11
    @MadJack arXiv policy. Mar 12, 2017 at 2:54
  • 6
    To those in the know, "Under review as a conference paper at SIGBOVIK 2017" is also a strong indicator, SIGBOVIK being the Association for Computational Heresy's annual conference held on or around April 1 in in Pittsburgh.
    – Danica
    Mar 12, 2017 at 16:37

4 Answers 4

136
+100

There is a long tradition of posting joke papers to arXiv on or around April Fool's Day, especially in astro-ph - see the list below. The fact that all these papers were approved for arXiv offers some evidence (though not proof) that joke papers are okay.

It's probably best to limit joke papers to around April 1, though, when people know to be on the lookout for this stuff - or at least provide some indicator which is very clear to experts in the field that the paper is a joke, such as listing it as being submitted to Acta Prima Aprilia, SIGBOVIK, or the Annals of Improbable Research.

A history of arXiv joke papers

(mirrored on my website)


Note: since this has become what I think is the most comprehensive list of arXiv joke papers I know of, I've made it community wiki so anyone can add to the list. Any additions should be clearly joke papers and, in keeping with the topic of the question, must be posted on arXiv.

5
  • 2
    As another example there is also Donald Knuth's (quite hilarious) "The Complexity of Songs" which was published in ACM SIGACT - a quite serious and prestigious journal. So the tradition of posting joke papers does not just exist on arXiv but is a wider phenomenon at that.
    – Voo
    Mar 12, 2017 at 21:11
  • 13
    -1, because I was planning on using my time productively before seeing this. (+1!)
    – Nat
    Mar 13, 2017 at 1:58
  • 3
    So how do non-experts know when they're reading a joke article? That sounds unprofessional and childish to me.
    – Cape Code
    Mar 14, 2017 at 13:22
  • 5
    @CapeCode Articles on arXiv are targeted at experts. Other people generally won't understand what they're reading whether it's a joke or not.
    – David Z
    Mar 15, 2017 at 14:19
  • The best joke paper of all times is Gates, V., et al. "Stuperspace." Physica D: Nonlinear Phenomena 15.1-2 (1985): 289-293. The published version is available here: sciencedirect.com/science/article/pii/0167278985901733 and a free e-copy from here: languagelog.ldc.upenn.edu/myl/Stuperspace.pdf. It predates arXiv. Apr 8, 2018 at 0:29
38

Uploading joke papers to arXiv does not violate arXiv policy, according to this email sent by the arXiv moderation:

---------- Forwarded message ----------
From: arXiv Moderation
Date: Mon, Mar 13, 2017 at 4:38 PM
Subject: [moderation #189697] Removing a joke paper
To: spamme@gmail.com

Dear Spa,

We will not remove this paper as it does not violate any of our policies.

-- [retracted name]
arXiv moderation

On Sat Mar 11 18:27:47 2017, spamme@gmail.com wrote:

Dear arXiv-moderation,

This paper is a joke paper: https://arxiv.org/abs/1703.02528 (it was submitted to an April 1st joke conference: http://sigbovik.org/2017).

Could you please remove it from arXiv?

Best,
Spa Mme

This seems to be in contradiction with https://arxiv.org/help/moderation (mirror):

arXiv is distinct from the web as a whole, because arXiv contains exclusively scientific research content. Although arXiv is open to submissions from the scientific communities, our team has worked behind the scenes for a long time to ensure the quality of our content.

arXiv moderators will suggest the removal of a submission that violates arXiv policies in some way. Potential reasons for removal are: Inappropriate format. arXiv accepts only submissions in the form of an article that would be refereeable by a conventional publication venue. This excludes […] papers that contain inflammatory or fictitious content, papers that use highly dramatic and mis-representative titles/abstracts/introductions.

4
  • 8
    That might be slightly surprising. In ArXiv moderation system page we read: "arXiv is distinct from the web as a whole, because arXiv contains exclusively scientific research content." and "arXiv is an openly accessible, moderated repository for scholarly papers in specific scientific disciplines. Material submitted to arXiv is expected to be of interest, relevance, and value to those disciplines. arXiv reserves the right to reject or reclassify any submission." I wonder why ArXiv moderators classify clear joke papers as scholarly papers with scientific content.
    – PsySp
    Mar 14, 2017 at 10:29
  • 2
    @PsySp Yes I was surprised too (and disappointed). I wonder what it takes to be an arXiv-approved joke. Mar 14, 2017 at 16:19
  • 1
    Maybe these excerpts from the Moderation page of ArXiv should be added to your answer because it seem they directly contradict the answer you received from the moderator! arxiv.org/help/moderation (It is even in bold!)
    – PsySp
    Mar 14, 2017 at 16:31
  • 1
    @PsySp Sounds good, added. Apparently a joke isn't fictitious content, but is scientific research content. Mar 14, 2017 at 16:36
37

The arXiv (pronounced "archive")is a repository of electronic preprints, known as e-prints, of scientific papers in the fields of mathematics, physics, astronomy, computer science, quantitative biology, statistics, and quantitative finance, which can be accessed online.

So, please do not upload joke papers. These papers are not helping the community, and they are not helping the authors of such papers.

Having said that, if the paper has any scientific value, even if it is a funny paper, then it is more than welcome. TCS has a conference called FUN with algorithms the purpose of which is to publish amusing papers with certain algorithmic flavor.

10
  • 1
    @FranckDernoncourt I mean, in case the authors use their real names (which I doubt). In any case I do not see why anybody would want to create noise as you say.
    – PsySp
    Mar 11, 2017 at 23:57
  • 61
    I'm an arXiv moderator, and I disagree with this answer.
    – JeffE
    Mar 12, 2017 at 2:36
  • 83
    @JeffE please post your own answer?
    – smci
    Mar 12, 2017 at 3:14
  • 5
    @JeffE A good recipe to receove something like my last "Our volunteer moderators determined that your article does not contain sufficient original or substantive research to merit inclusion within arXiv. Resubmission of removed papers may result in the loss of your submission privileges." Mar 12, 2017 at 10:44
  • 4
    @StuartGolodetz My opinion is that a joke paper, published under Machine Learning tag, is borderline qualified for spam. There are many other places where such a paper would be more appreciated and could potentially invoke less inconveniences. Because, sometimes, it's hard to judge if it's a clear joke or not.
    – PsySp
    Mar 14, 2017 at 9:51
1

Also, realize that people have done this to real journals. Consider the Sokol hoax. Or the famous math paper about big game hunting mocking. Or the more subtle Physical Review Letter co-author F. D. C. Willard

I would avoid doing it too much. And if it gets done too much it will damage the resource. That said, a little bit is kind of amusing. Especially if skillfully done. But some pepper is good...too much is bad.

You must log in to answer this question.

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