A single paper everyone should read? [closed]

Different people like different things in math, but sometimes you stand in awe before a beautiful and simple, but not universally known, result that you want to share with any of your colleagues.

Do you have such an example?

Let's try to go in the direction of papers that can actually be read online or accessible with little effort, e.g. in major libraries, so that people could actually follow your advice and read about it immediately.

And as usual let's do one per post and vote freely, vote a lot.

-
Why are so many answers big-picture papers and philosophical tracts? I'm sure many of them are good papers, but is this really what the question was about? Am I right in suspecting that posters only read the title of the question and not the question itself? –  Thierry Zell Sep 4 '10 at 0:23
Perhaps it's time to close this question. –  S. Carnahan Oct 22 '10 at 17:40
Agreed, as Thierry and Tobias say, there are too many recommendations for punditry. –  Robin Chapman Nov 17 '10 at 11:48
Vote not to close –  Gil Kalai Jul 9 '11 at 6:28
Vote to close; cancelling Gil Kalai's non-closure vote. [As this was requested once if it is not clear by name I add: I am a 'voting user'.] –  quid Jul 9 '11 at 10:23

closed as off topic by Kevin Lin, Andres Caicedo, Felipe Voloch, Hailong Dao, Bill JohnsonJul 9 '11 at 15:01

Questions on MathOverflow are expected to relate to research level mathematics within the scope defined by the community. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about reopening questions here.If this question can be reworded to fit the rules in the help center, please edit the question.

One paper that I've read a few times and always loved was Who Can Name the Bigger Number? (also available in Spanish and French, for those who prefer to read in those). It discusses how our concept of "big numbers" has evolved over time, and talks about Turing machines and the "busy beaver" numbers, which represent a non-computable function.

-

2N Noncollinear Points Determine at Least 2N Directions, by Peter Ungar. This is a beautiful short paper that proves the result in the title.

A general remark: If you have to choose a single paper (or a single paper of a mathematician selected in other answers), I would recommend more strongy to choose original papers of important basic results rather than large survey papers or "meta" paper about mathematics. (This is also closer to the original intention of the question.)

-
show 1 more comment

The Unreasonable Effectiveness of Mathematics in the Natural Sciences

by Eugene Wigner

Although Wigner is physicist, I consider this article about mathematical physics very important both for physicists and mathematicians. It's a wonderful feeling to realize to what extent our world can be mathematical.

The miracle of the appropriateness of the language of mathematics for the formulation of the laws of physics is a wonderful gift which we neither understand nor deserve. We should be grateful for it and hope that it will remain valid in future research and that it will extend, for better or for worse, to our pleasure, even though perhaps also to our bafflement, to wide branches of learning. E.Wigner

-
This is not only a historically important paper,it focuses on an aspect of mathematics that Western culture has hesitated to return to after the wholesale rejection of it during the Bourbaki era. –  Andrew L Oct 22 '10 at 19:04
Even if bashing Bourbaki seems to have become hip again (as it was actually during the Bourbaki era as well), and thus the myth of the "wholesale rejection" of applications of mathematics "during the Bourbaki era" has become generally accepted in certain circles, it still remains a myth, which like all myths contains a germ of truth surrounded by a lot of prejudices, misunderstandings and plainly wrong statements. –  Tobias Hartnick Nov 17 '10 at 13:44

Carl's Pomerance "A tale of two sieves", available at

http://www.ams.org/notices/199612/pomerance.pdf

It makes a quick introduction to subexponential factoring algorithms via their development from Fermat's Algorithm and then compares the Quadratic Sieve with Her Majesty the (General) Number Field Sieve, in a thorough, appealing and very understanable manner.

-

Toen's course on stacks. I don't know if this counts as a paper, but courses 2,3, and 4 introduce a really interesting approach to geometry using the functor of points approach that I've not seen before.

-
show 2 more comments

"An Introduction to the Conjugate Gradient Method Without the Agonizing Pain" by Jonathan Shewchuk at UC Berkeley

-

That's easy just off the top of my head,Illya: Nets And Filters In Topology by the late Robert G. Bartle;appearing in the 1955 Volume 62 of American Mathematical Monthly. I remember having a friend in the Stanford mathematics honor society who'd published papers by age 20,but had never heard of either nets or filters. I recommended it to him right on the spot.

-
show 1 more comment

I highly recommend this lucid, little book (with the length of a paper):
Mathematics: A very short introduction, by Fields Medalist Timothy Gowers

-

PDE as a Unified Subject by Sergiu Klainerman.
An essay on partial differential equations written by a leading expert in the field, I strongly recommend to anyone who aspires to know more on the subject as well as to those who are not interested strictly in PDE's, but would like to get a grasp of interactions between Mathematics and Physics. There are also many interesting references.

-

"Rigor and Proof in Mathematics: A Historical Perspective" by Israel Kleiner. Mathematics Magazine December 1991, 64:291-314.

This paper gives a very nice overview of how the understanding of rigor in mathematics has evolved from the early ages to the 20th century.

http://www.jstor.org/sici?sici=0025-570X%28199112%2964%3A5%3C291%3ARAPIMA%3E2.0.CO%3B2-Z

-
On the theorem of Pythagoras by by E.W. Dijkstra. (Did you know that in every plane triangle sgn$(\alpha + \beta - \gamma)$ = sgn$(a^2 + b^2 - c^2)$, a "theorem, say, 4 times as rich [as the original]"?)
I don't think this is any news to most mathematicians. This is even in some good German schoolbooks from the 1960's-70's. Often when there is a statement like "if $a=b$ then $c=d$" one could check whether $a\leq b$ implies $c\leq d$, and lots of geometric inequalities have been created this way from identities. –  darij grinberg Nov 17 '10 at 14:08