Academic Snarkives

June 4, 2010 7:45 AM

arXiv vs snarXiv. "A ran­dom high-energy the­ory paper gen­er­a­tor incor­po­rat­ing all the lat­est trends, entropic rea­son­ing, and excit­ing mod­uli spaces".

Suggested uses, depending on your location in the academic food chain.

posted by a womble is an active kind of sloth (50 comments total)

In my eagerness to post I did not include a link to arxiv.org, a collection of open access papers form the fields of Physics, Mathematics, Computer Science, Quantitative Biology, Quantitative Finance and Statistics from Cornell University Library.
posted by a womble is an active kind of sloth at 7:49 AM on June 4, 2010 [2 favorites]

Also of interest: SCIgen, and a generated paper the made it through a Bentham Journal's Peer Review.
posted by The White Hat at 7:55 AM on June 4, 2010

Reminds me of the Postmodernism Generator.
posted by The Winsome Parker Lewis at 7:58 AM on June 4, 2010 [1 favorite]

I'm sufficiently bad at this that I'm categorized as an "LHC doomsayer."
posted by a small part of the world at 8:02 AM on June 4, 2010

This is great.

I'm a math phd student and still suck at this. Way worse than a monkey, and I'm not even in my ninth year, yet.
posted by kaibutsu at 8:06 AM on June 4, 2010

As good as a monkey. That's where I'm at.
posted by jsavimbi at 8:09 AM on June 4, 2010

I only just now realized that the X in arXiv is supposed to be a Chi and that arXiv is supposed to be pronounced "archive". I haven't felt this kind of shame since I very-belatedly noticed that the Sun microsystems logo wasn't just abstract squiggles.
posted by roystgnr at 8:16 AM on June 4, 2010 [6 favorites]

14 of 25 (56%) — Better than a monkey
posted by bz at 8:17 AM on June 4, 2010

I'm getting good results by always choosing the longer of the two titles.
posted by motty at 8:23 AM on June 4, 2010 [1 favorite]

17 out of 25. I have no idea what any of this is about.
posted by The Mouthchew at 8:30 AM on June 4, 2010

At first I was like, "markovy goodness" but then I read the about post and it turns out it's actually based on a context-free grammar which is recursively expanded from the top down by picking one option randomly at each node.
posted by Rhomboid at 8:30 AM on June 4, 2010

roystgnr: "I only just now realized that the X in arXiv is supposed to be a Chi and that arXiv is supposed to be pronounced "archive". I haven't felt this kind of shame since I very-belatedly noticed that the Sun microsystems logo wasn't just abstract squiggles."

Hell, I thought this was going to be a comparison of text editors.
posted by Joe Beese at 8:32 AM on June 4, 2010

I only just now realized that the X in arXiv is supposed to be a Chi and that arXiv is supposed to be pronounced "archive". I haven't felt this kind of shame since I very-belatedly noticed that the Sun microsystems logo wasn't just abstract squiggles.

...at least you didn't have to have it pointed out to you by a guy who didn't figure it out until moments prior.
posted by griphus at 8:35 AM on June 4, 2010 [4 favorites]

I was doing really good for a while: Got 12 in a row right. And then I was all "Gamma Gamma collider"? That can't be real! And then it all went downhill from there.

How were people saying arXiv in their heads? AR-zive?
posted by PMdixon at 8:48 AM on June 4, 2010

I just picked the green one each time.
posted by Floydd at 8:48 AM on June 4, 2010 [1 favorite]

I was saying it as Ar-xxxx-I've, like when you say XXXXXtreme and everyone knows you're saying it without the first e.

It made me feel like it was or the really XXXXXXtreme science, but now I just feel silly.
posted by beepbeepboopboop at 8:51 AM on June 4, 2010

Did the Cox-Zucker Machine make it in?
posted by griphus at 8:52 AM on June 4, 2010

Hmm, I can't remember who did it, but it reminds me of an inversion of what this scientist did to mock the humanities/pomo thing...

He submitted some mumbo-jumbo paper and it got published, and he was like a ha!

And skeptic magazine and skeptical enquirer published about it...

I want to say it was Alain Aspect, but it appears, at least reading his wiki page, that it wasn't him. Does anybody know what I'm talking about?

Because this sounds exactly like that (except these "Snarxiv" papers didn't get published in scientific peer-reviewed journals)
posted by symbioid at 8:52 AM on June 4, 2010 [1 favorite]

http://en.wikipedia.org/wiki/Sokal_affair

There! Alan Sokal!
posted by symbioid at 8:53 AM on June 4, 2010 [1 favorite]

Now I'm wondering how y'all have been pronouncing Tex and LaTeX.
posted by kmz at 8:53 AM on June 4, 2010

Now I'm wondering how y'all have been pronouncing Tex and LaTeX.

"Edin-burrah" and "Glos-ter"
posted by griphus at 8:56 AM on June 4, 2010 [14 favorites]

Playing with arXiv vs snarXiv has got to be about 8000 times better than actually working on the paper that I'm, uh, supposedly going to eventually post to the arXiv.
posted by dorque at 8:57 AM on June 4, 2010 [1 favorite]

Now I'm wondering how y'all have been pronouncing Tex and LaTeX.

Based on my well-informed continental european colleagues I pronounce it 'Tech' or 'Lay-tech'
posted by a womble is an active kind of sloth at 9:15 AM on June 4, 2010

I laughed out loud when it gave me "D-brane models with ghosts" (not real by the way).
posted by atrazine at 9:19 AM on June 4, 2010

arxiv.org used to be at xxx.lanl.gov
As I heard it, www was adopted early by physicists trying to share data from particle colliders, and since this was sharing papers, you use the letter after w, hence xxx. Of course this name was whack, and combined with the lack of peer-review (by design), to this day I think of it as the go-to site for dirty unrated physics. I suspect the X in arXive is mainly an homage to that, with the "chi" being an excuse to preserve the snickering.
posted by Humanzee at 9:26 AM on June 4, 2010 [1 favorite]

I might have fallen for "D-brane models with ghosts"
posted by Humanzee at 9:28 AM on June 4, 2010

I was getting about 75% wrong and quit. However, 70% is statistically significant, better than a coin toss. I won't speculate on what this means.
posted by stbalbach at 9:30 AM on June 4, 2010

whatever stbalbach said
posted by infini at 9:38 AM on June 4, 2010

arxiv.org used to be at xxx.lanl.gov

ch-ch-ch?
posted by Splunge at 9:58 AM on June 4, 2010

Woohoo! Sleeping with a physicist has my score hovering around 65%! Take that English degree!
posted by Diagonalize at 9:58 AM on June 4, 2010 [1 favorite]

You know, a lot of the ones they said were fake papers have perfectly reasonable titles:

"Dark Matter in the CMB"?

One way to find fakes is words in the wrong context.

TypeIIA, for example, is a type of supernova. If you see it in a title with QCD, (quantum Chromo dynamics), its more than likely a fake.
posted by empath at 10:48 AM on June 4, 2010

No duh, empath. No duh.
posted by oddman at 11:06 AM on June 4, 2010

80%. I am awesome.

I earned my bullshit detection skills with four years in the English Major trenches.
posted by sonic meat machine at 11:19 AM on June 4, 2010

empath, TypeIIA could refer to superstring theory. Wikipedia on Type II strings.
posted by Schmucko at 1:00 PM on June 4, 2010

superstring theory

Still pretty sure that's made up...
posted by Potomac Avenue at 1:14 PM on June 4, 2010

Type II could also refer to a particular kind of superconducting state...
posted by Anderson_Localized at 1:22 PM on June 4, 2010

Well, learn something new every day.
posted by empath at 1:46 PM on June 4, 2010

Usually the simpler one is correct. The ones with grammatical errors are right out. And whoever wrote the fake ones has an unhealthy obsession with various branes.

Although I just got fooled by 'some conspicuous examples'.

Hmm.. "Supersymmetry in Classical Mechanics" actually sounds kind of interesting.
posted by delmoi at 2:01 PM on June 4, 2010

Actually the ones I had with confusing grammar in the title were the actual papers... The random generator at least has a structure it is trying to adhere to.
posted by Zalzidrax at 2:19 PM on June 4, 2010

I broke my best streak picking the likelier of "Electrons" and "Minimal Super Technicolor"
posted by rollick at 2:20 PM on June 4, 2010

Does anyone know all the rankings? From 40% to 70%, I've gotten the following:

9th year grad student
Guess harder
Same as a monkey
Slightly better than a monkey
Undergraduate
First year grad student
posted by rollick at 2:23 PM on June 4, 2010

There's also vixra.
posted by oonh at 3:18 PM on June 4, 2010

I am better than a monkey.
posted by Hairy Lobster at 3:54 PM on June 4, 2010

75% - 76% failed your orals
77% - 79% first year grad student
80% - 83% 2nd year grad student
86% - 88% postdoc
100% nobel prize winner

I'm guessing that somewhere between 90 and 100 there are various kinds of professor. I give up though. There is considerable overlap between the most reasonable snarXiv titles and the most unreasonable arXiv titles. I got one right by guessing that no auto-generated title would have laTex codes just lying about in it. Some of the others are just stupid. Bad grammar, terminology that's simultaneously vague and arcane, ... ugh. There's a reason a lot of arXiv papers never get published in a real journal.
posted by Humanzee at 4:01 PM on June 4, 2010

So my resident theoretical physicist just got home. He's been spending the last few months collaborating with the creator of the site on actual physics, so he was very confused when his work life and non-sciency girlfriend collided via Metafilter

He's also cranky that he got one wrong, but in any case, I'll have to get him to show this thread to David.
posted by Diagonalize at 4:30 PM on June 4, 2010

Actually, I just noticed it gives you stats on how many people get an article title wrong. So for example "Gluon Mass Generation Without Seagull Divergences" has had one person guess that it was a fake paper (me I guess -- seriously Seagull divergences?)

I also thought this one was fake: "The Non-perturbative Constraint on Sea Quarks --The Strange Sea Quarks in the Nucleon and the Soft Pion Contribution at High Energy--". Why does it end in a hyphen?
posted by delmoi at 6:41 PM on June 4, 2010

Seagull divergences

5 minutes of googling finds a reference to a type of Feynman diagram called a seagull diagram, I assume because of the shape.
posted by empath at 6:50 PM on June 4, 2010

Actually, let's make this thread about explaining physics jargon that pops up in these titles:

Here's one I understand:

Chiral Symmetry Breaking:

The laws of physics should (or I guess 'could') theoretically be right/left symmetrical (mirror symmetry, also called parity). If you reverse left and right in any system, the results would always be the same if physics had parity. However, some interactions involving the weak force have a bias toward either clockwise or counterclockwise spin, so flipping left and right would give you an opposite result, which means that it's chiral -- it is either left or right handed. Why this happens is a mystery.

Other symmetries in Parity are Time and Charge. If you reverse Time, Charge and Parity together, the laws of physics are symmetrical, as far as we know, but reversing them separately is not always.

You'll also see the word 'chiral' in relation to biology because almost all organic molecules have chirality (they lack mirror symmetry) -- amino acids are 'left-handed' and sugars are 'right-handed'.

Now, someone explain what an orbifold is.
posted by empath at 7:11 PM on June 4, 2010

uh, meant to say other symmetries BESIDES parity...
posted by empath at 7:15 PM on June 4, 2010

Take a manifold with some symmetry (say, a plane that is rotationally invariant). For specificity let's consider a plane rotated 90 degrees. You do this 4 times, and obviously you've gotten back to the starting plane (even if you had, say, painted a picture on the plane).

An orbifold is when you say ok, I'm considering the plane, but I'm going to insist that actually if you rotate it just once by 90 degrees, you get back the same space. In other words, I'm going to say that a point lying at x=1, y=0 is actually the *same* physical point as one lying at y=1, x=0.

A shorter way to say this, if you're ok with a little abstract algebra, is that you're considering the plane modded out by the symmetry group (in this case, by Z4).
posted by nat at 10:19 PM on June 4, 2010