Ernie's 3D Pancakes

Algebra IV

• “It looks like you're trying to multiply two numbers. I can help show you how to use the Method of Finite Differences to find a good approximation of your answer using only addition and subtraction. Would you like me to bring up a wizard so that we can get started on finding an appropriate power series?”

• “By 1869 the handwriting requirement and French had been dropped, but algebra had been added and students needed to pass a qualifying exam in the required subject areas.” (see also)

• “It's also designed to be a college-level math class for students who have zero skill at algebra. And if you're doing college level math and you can't calculate, well, then you're going to be conjecturing and proving and whatnot. If you're incapable of the background necessary to succeed in calculus, then you're in for a semester of learning how to make a logical argument stick together.”

• “This must be heartbreaking to his father, who loves learning so much that he’s chosen to subscribe to the premium digital cable package just so he can get the Algebra Channel.”
  

Horseshoes and hand grenades

Юрий Лифшиц, a postdoc at CMI in Caltech, sent me a link to his extensive Homepage of Nearest Neighbors and Similarity Search, which includes a large bibliography, tutorial slides (and soon video!), and a nice list of links to related conferences, data sets, and implementations. Thanks, Yury!

Ffrom Yury's page I found the call for papers for the 1st International Workshop on Similarity Search and Applications, which will be held April 11-12, 2008, in Cancún, Mexico, as a satellite workshop of ICDE, with proceedings published by IEEE Press. Submissions are due November 28.

And that led me to this this new book on spatial data structures by Hanan Samet, which I didn't know about, even though it's been out for over a year. Either I've been living under a rock, or somebody doesn't know how to advertise. Books, I mean.

King Algorithm?

Sunday's New York Times "Week in Review" included short article on algorithms that run on a mixture of computers and humans that somehow manages never to mention Luis Von Ahn.

It was the Internet that stripped the word of its innocence. Algorithms, as closely guarded as state secrets, buy and sell stocks and mortgage-backed securities, sometimes with a dispassionate zeal that crashes markets. Algorithms promise to find the news that fits you, and even your perfect mate. You can’t visit Amazon.com without being confronted with a list of books and other products that the Great Algoritmi recommends. [...]

What is spreading through the Web is not exactly artificial intelligence. For all the research that has gone into cognitive and computer science, the brain’s most formidable algorithms — those used to recognize images or sounds or understand language — have eluded simulation. The alternative has been to incorporate people, with their special skills, as components of the Net.

The article makes this symbiotic relationship between humans and machines sound novel (and rather sinister), but to me it doesn't seem fundamentally different than other aspects of augmented social intelligence. Hasn't human civilization always depended on algorithms performed by a combination of human and machine? Wasn't that the foundation of the industrial revolution, or parliamentary democracy, or the Macedonian phalanx, or written language? Hasn't the world economy always run on a combination of human decisions and careful arithmetic? What else is agriculture but an algorithmic (mechanical, step-by-step) response to the life cycles of certain plants and animals, the weather, and human hunger?

Unnecessary dairy overlap

Subway offends geometric decency.

Life, small holes

Okay, finally, I've new gone through the disk crash/replace disk/recover files from iPod/reinstall software/mouse button death/buy new laptop/move files/reinstall software/backlight death/move files back onto old laptop with broken mouse/send new laptop for repairs/learn to keep my thumb off the track pad/get new laptop back/move files onto new laptop dance, all during the first few weeks of teaching discrete math to 200 froshlings (or it is "froshling"?) and the first few weeks of Katherine walking more than she crawls, with lots of fun visits to the pediatrician thrown in just for laffs. (Kids, can you say "voiding cystourethrogram"? Sure, I just knew you could. Don't worry; Katherine's just fine.)

So. As I promised several people weeks ago, I've finally made the slides from my WADS talk available. If you wait long enough, you can also click through them right here, thanks to Keynote's Flash export. (Sorry about the occasional glitches; Flash doesn't understand polygons with holes.) Enjoy!

The talk refers to some unpublished (grumble) results about Vietoris-Rips complexes of points in the plane. I'll write about these in some future blog posts; meanwhile, see David's well-written Wikipedia article.

So say we all.

When all else is lost, and one’s scientific paper is beginning to make too much sense, read too clearly, and display too much insight and enthusiasm, I have one last recommendation that can help the author to maintain the essential boring tone. My advice is to make sure that all written statements, even trivial ones, must be supported by one or more references. It does not matter that these statements are self-evident or that they comply with well-established knowledge, add a reference, or preferable 3–5, anyhow.

Excessive quotation can be developed to perfection such that the meaning of whole paragraphs is veiled in the limited space between references. This technique maintains the boring quality of scientific publications by slowing down the reader, hiding any interesting information, and taking up valuable space. When authors are unsure of which paper to cite, they should always resort to citing their own work regardless of its relevance.

—Kaj Sand-Jensen, “How to write consistently boring scientific literature”, Oikos 116:723–727, 2007.

A scientific manuscript is intended to communicate new information and to teach new material to a willing audience. Many authors forget this simple fact; rather, they view the writing process as an opportunity to bolster their own egos and impress the reader, even discomfit the reader somewhat, either with too much material or too little.

—Stephen D. Senturia, “How to Avoid the Reviewer’s Axe: One Editor’s View”, J MEMS 12(3):229–232, 2003.

[via MF]

SODA, NIPS, Larry, Ed, Bob, Dick.

• SODA results are out; you can find the list of accepted papers on at least five other blogs. Two of my submissions were accepted: Finding one tight cycle (with Sergio Cabello, Matt DeVoss, and Bojan Mohar), and Empty-ellipse graphs (with Olivier Devillers and Xavier Goaoc). My other two submissions will not, alas, have the privilege of being in the same conference as UIUC PhD student Ke Chen's brilliant paper on approximate clustering with outliers, or UIUC PhD student Nitish Korula's brilliant paper on approximate orienteering, or the first observation-in-the-wild of the iterated inverse Ackerman function (thanks Mihai), or Mike Mitzenmacher's forays into practical hashing or hiring in Lake Wobegon, or the latest witness complex result from the Stanford topology crowd (as opposed to the Duke topology crowd, the INRIA topology crowd, or (dare I suggest) the Illinois topology crowd), or Uri Zwick's latest two-word title.

• A NIPS workshop on Topology Learning, to be held December 7 and 8 in Whistler, British Columbia, was recently announced on the computational geometry mailing list. The deadline for submitting 2-page abstracts is October 12.

• Psst! Wanna buy a formula? (cough cough)

• By Friday afternoon, the population had risen to a somewhat alarming number.

• Bob Dylan gives a good 10-second slide talk.

• And this year's Emmy for Outstanding Original Music and Lyrics goes to a song by the Erik Demaine of sketch comedy and some other guy in a fake beard. Meanwhile, this year's Google Research Award was won by the Andy Samberg of theoretical computer science and some other guy in a fake beard. Neither the song nor the research proposal is likely to be performed at the Emmy's, but who knows, maybe we can convince one pair or the other to make an appearance at SODA.

  What? Look, if Scott can talk about vaginae dentatae...

Aw, shucks.

Doron Zeilberger thanks me for my snide remarks!

See also Aek Thanatipanonda's further hopping.

Update 9/9: Greetings, carnivalleros!