Horculas en el migo. And so it is.

Looking back, it is clear that growing up in Buenos Aires shaped my view of the world, what to expect from the day to day of life. And this happens in really unexpected ways: the shape of buildings, the layout of streets and cities, the birds and the trees (it is frustrating to try to explain the peculiarities of the hornero, the little bird of the endless pampas building its mud house on top of alambrados, to someone and not being able to convey all the memories it brings or why it matters).

Coming to Italy is a bit like going back. So much of the culture feels like home, from the language to the hand gestures to the food. But perhaps the most significant connection for me is: la passeggiata.

People actually go out to walk for the sake of it, for the pleasure of strolling about with their friends, their family, not just to walk the dog, to shop or to exercise. People in normal clothes, pushing a pram, or grandma in a wheel chair, kids chasing each other around, couples holding hands. And they do so in the open air, perhaps taking in the rays of a beautiful sunny Sunday, in public spaces, sidewalks, no commerce involved.

Buona Domenica!

]]>

One of the main goals of the AIM workshop is to be able to explicitly compute the -function of hypergeometric motives. This means computing numerically enough of the ingredients of the -function in detail so as to, for example, check that it satisfies the expected functional equation. In fact, testing the functional equation is a good way to find a few of the ingredients, say the conductor, that might be missing. Once we have a large computable set of higher degree -functions (i.e., with Euler factors of degree bigger than two say) we can begin exploring numerically various conjectures about their zeros, special values, etc.

So, what is a hypergeometric motive? The best place to start is the Legendre family of elliptic curves

Deuring computed the number of points of this family over the finite field and showed that modulo is given in terms of the polynomial

Igusa later pointed out that this polynomial satisfies the same hypergeometric differential equation as a period of over the complex numbers.

On one hand we have for a two dimensional vector space with an action of Galois and on the other the two dimensional space of solutions to the hypergeometric differential equation. These are two incarnations of an abstract object: a “motive”.

Now the point is that we can start with *any* hypergeometric equation with appropriate parameters and it will yield a family of motives parameterized by . (Because we have a rigid system the hypergeometric equation is determined by very simple discrete data; in other words, there is no moduli.) For each rational value of we get an associated -function. These are what we want to be able to calculate.

The key fact is that though it is not entirely clear how to describe the motive (for example, how do we come up with the Legendre family of elliptic curves starting from the differential equation?) this description is not strictly necessary! Indeed, there is a formula like that of Deuring’s above that gives the trace of Frobenius on the motive. This finite field analogue of the hypergeometic series (due to N. Katz) is quite computable and it actually gives the trace as a -adic number and not just modulo . With this we can compute the Euler factors of our -series at least for all but finitely many primes up to some bound.

The challenge that remains is how to compute the rest of the data of the -function. Namely, i) the Euler factors for the remaining primes, ii) the Gamma factors, iii) the conductor and iv) the sign of the functional equation.

To complete this program some further description of the motive would in fact be helpful. But there isn’t necessarily a unique or canonical way to exhibit the motive. Different ways of presenting it bring their own advantages. To my mind a motive is somewhat like what genes are in biology. The same gene can appear in the DNA of very different life forms.

An example we worked out with Henri Cohen is a good illustration of this point. The motive has weight zero and corresponds to a Galois extension of with Galois group the Weyl group of of order . The -function of the motive is the Artin -function attached to the reflection representation (of dimension ). As it turned out the motive also sits naturally in of an affine cubic surface with an Eckardt point. This surface has lines that are permuted by the Galois group. This description gives us an explicit degree extension with Galois closure . From here it is not too hard to give the associated Artin -function directly.

]]>

I’ll leave aside the late night taxi ride where the driver, a young woman, told a few off-color jokes (“you are not a prude man, are you?”); including one that I am pretty sure was popular when I was in primary school. (Let’s just say that it involved traveling by camel in the desert.)

More curious was the guy with the fourth dimension. It started innocently enough: “what’s your business?” or something like that. When I said I taught math (often a risky move) he went off about the fourth dimension and extra-terrestrial beings living among us. He actually showed me a pile of issues of a magazine called “The Fourth Dimension”. I also recall something about a meeting of extra-terrestrials in Washington DC the following month. And this was just a few minutes into the ride.

I then made a really reckless move. I mentioned that in fact, I was currently working with some physicists (this was at the time of my collaboration with Ph. Candelas and X. de la Ossa) and that in string theory the thinking was that the universe was actually ten dimensional.

The response was a chilly silence. I distinctively recall thinking: the driver must be saying to himself “Now, there is a nut in this taxi … and it is not me”.

]]>

The question is this. Take x_1,…,x_k some n digit numbers, say in base 10. When you add them with the usual elementary school algorithm some of the columns of k digits will contribute a carry, a number in the range 0 to k-1, to the next column. What is the distribution of these carries for random x_i for large n?

The question has an interesting answer, which I won’t spoil but refer you to the original beautiful paper by J. Holte (mathscinet).

Surprisingly, Diaconis and Fulman found a connection between this question and card shuffling, Foulkes characters, symmetric functions,…

I love it when we discover something new and interesting in things right under our noses.

]]>

The question arose from our joint work with Hausel on the geometry of character varieties. We later found and instance of this phenomenon already discussed by Reineke archiv (see below).

I presented three examples.

1. The Grassmanian G_n,k. Here for large n and k the Betti numbers are distributed as a Gaussian. No surprise perhaps; by Poincare duality the distribution is symmetric and by the Hard Lefschetz theorem it increases up to the middle of the range and then decreases.

2. The Hilbert scheme of n points on the plane. Here the Betti numbers are distributed as partitions of n according to length (by Goettsche’s work). There is a limiting distribution as proved by Erdos and Lehner. It is the Gumbel distribution exp(-exp(-x)) that appears as an universal distribution for max(x_1,…,x_n) of independent, identically distributed random variables x_i (given the known height of a river for the past 10 years how high could we expect it to get this year?) You can see the case n=500 and the corresponding limit case in the notes of the talk. The distribution is no longer symmetric.

3. The toric hyperkaehler variety (Hausel-Sturmfels archiv) associated to the complete graph K_n. Here the Betti numbers are given by the coefficients of the reliability polynomial of K_n. These are known to have a limiting distribution: the Airy distribution (the same as that appearing in Reineke’s case mentioned above.) This distribution appears in a number of different other contexts (hashing algorithms, area of Brownian motion, large graphs with fixed genus, etc.)

It is remarkable that the distributions in 2 and 3 (Gumbel and Airy respectively) are in fact very close (scaling and shifting appropriately) to each other but are definitely not the same. (In the early stages we thought that perhaps we always got the Gumbel distribution; it makes you wonder how much to read in the comparison of continuous graphs to discrete data.)

Is there some kind of universality?

After the talk Stavros Garoufalidis mentioned that the Airy distribution also appears in various matrix models in Physics.

]]>

I’ve seen a few dollar coins too. Usually all of them at once unfortunately. At least 15 years ago, if you bought a NJ transit train ticket in a machine at the station and paid with a 20 dollar bill the change came clanking down in a stream of dollar coins. They are heavy.

And just the other day reading a post at the The Oil Drum encountered the word copacetic for the first time in my life, according to World Wide Words a uniquely American slang word meaning “fine, excellent, going just right”. (Not much use for it these days.)

On my way back from Vancouver once the US customs officer that checked my passport, after finding out I was a mathematician, told me: [thinking pause] “I hope things add up for you”.

I hope the post does.

]]>

A quick search uncovered this interview with Ruth Sabean, the assistant vice provost for educational technology at UCLA, where she discusses the decision. It is interesting and so is the posted discussion with readers at the end. In particular, I was led to this insightful post at e-Learning, a blog by Michael Feldstein from Oracle, who claims that “The monolithic closed source LMS [Learning Management Operating System] is dead meat.”

Incidentally, UT uses Blackboard, which according to the Washington Post article “was founded in 1997 by a few 20-somethings who quit comfortable jobs to start the company. The dot-com boom swept up Blackboard, and it weathered the subsequent bust before going public. Last year, it had sales of $180 million.”

Dead meat or not the question for me is: are we really improving the way we teach? or are we just using new tools to make the usual easier? We all know that writing a paper in TeX (and who doesn’t?) does nothing to its content, neither does giving a powerpoint talk.

On a related topic what I would really like to see is a web environment for collaboration in Mathematics. There are a number of systems out there that could be adapted for this purpose but I have yet to find one I am really satisfied with. Sakai looked promising but it doesn’t seem quite ready for what I have in mind yet.

I’d love to have something like this: a browser based interface (so that one could login from anywhere, whatever the operating system is) that allows one to up/down-load files, keeps an easy-to-use history of the files as they evolve, create webpages, link to documents, etc. It should have a broad array of communication tools: voice, video, whiteboard, chat, forums, etc. For mathematics the chat should have an automatic TeX formatting filter that would allow people to type formulas in TeX while talking via Skype, say. Finally, a powerful and smart search feature (within the system as well as the archive, numdam, JSTOR, GDZ, etc.) including, if they so allow it, other people discussions.

It seems to me that if technology is going to take us to a higher level of doing research it will only do so by increasing the opportunities for the random associations that fuel it. Powerful and smart searching is crucial for this. Am I the only one annoyed by, say, the searching capabilities of mathscinet? Unless you know an author’s name exactly chances are you’ll never find the paper you’re looking for.

As an example, the following funny thing happened to me recently. I was looking for material on a few integrals that have value a rational multiple of pi squared, which Coxeter talks about in the preface to his book “Twelve geometrical essays”. These arose from some volume calculations and are in one of his earliest papers. I quickly found the paper by Wagner, Peter, “Solution to a problem posed by H. S. M. Coxeter”, C. R. Math. Rep. Acad. Sci. Canada 18 (1996), no. 6, 273–277 (related to a different integral actually). The review in Mathscinet has the phrase: ” The method seems to be due to Tortellini in Crelle’s Journal 34 and is explained in Dirichlet’s Bestimmte Integrale.”

I was intrigued both by the alluded method and the name of its author, Tortellini. I had never heard of a mathematician of that name (of any era). A search in mathscinet with the author’s name gave nothing. At least the reviewer, H. W. Guggenheimer, had included the reference to Crelle’s Journal volume 34. A search in GDZ (not that straightforward either actually, I searched for Crelle in title, then clicked on the Journal’s actual name “Journal für die reine und angewandte Mathematik”) yielded the table of contents of volume 34 showing a paper by Barnaba Tortolini! Talk about a Freudian slip.

Actually, who was Barnaba Tortolini?

I find myself spending a lot of time searching for something (on the web, my hard disk or my office), which I know I have found before… (For example, I had to reconstruct the above Tortellini/Tortolini story all over again.) I could, of course, be more organized but wouldn’t it be great to have your computer help you out?

]]>

I spent two weeks in Benasque, Spain for the workshop P-adic analysis, Periods and Physics . We ended up being a very small group of participants, unfortunate in a way but the result was a very charming workshop. And what a place! Benasque is in a valley in the Spanish Pyrenees not far from the border with France, a small mountain town where you walk everywhere, surrounded by amazing mountains full of unbelievably beautiful hikes. String theorist have been coming here every two years for quite a while. We should have some more math workshops in Benasque!

I’m becoming somewhat obsessed with technology and how it’s changing our daily life in ways we hardly have time to think about, let alone understand. I grew up in Argentina at a time when phones were a luxury; apartments had them or not. Here I was in Benasque sitting under a tree facing a soccer field and some towering green mountains, talking to my brother in Argentina on Skype through a wireless connection to my laptop; it was right about 3pm, before the Benasque kids came out of their siesta to play and my brother had to take his kids to school.

Since writing the post on communicating mathematics today I became more and more convinced that we, the mathematics community, are not exploiting the available technology as much as we could.

Talking about this in Benasque my friend Maria mentioned Moodle a open source management system for courses. It is now installed at UT and I will give it a try for my graduate class this Fall. It looks really well done, with lots of customizing possible: forums, chats, blogs, quizzes, you name it.

I plan to use its “workshop” assignment module, which allows students to grade anonymously a number of other students homework. My goal is to give students the chance to read and assess somebody else’s mathematics and write a report about it. It is after all what we working mathematicians spend quite a bit of our time doing. Hopefully the anonymous feature will also give them a flavor of the referee system for publishing papers (without unleashing too much of the nasty sadism that this can involve). I am all in favor of assigned work for graduate classes in any case. To paraphrase a rock fashion designer: “If your pants don’t hurt it ain’t rock and roll”.

Of course technology can be a mirage too. In a memorable story John Tate was once asked how he dealt with papers when he was a graduate student, an era without photocopying machines. John, without missing a bit, answered: “We read the papers”.

On the other hand, as Philip Candelas remarked, it is likely that at the time one could be on top of pretty much everything being published in a given topic.

I’ve been looking around for some open source software to manage conferences, seminars, lectures, etc. Found some things but they’re not quite it yet. Any recommendations?

Finally, a shameless plug. My book “Experimental Number Theory” has finally appeared, published by Oxford University Press. It contains many computer scripts for the wonderful Number Theory package PARI-GP on hopefully interesting mathematics. It was a lot of fun to write and I learned a great deal in the process. In particular, I encountered my currently favorite elementary math problem (due to D. Knuth): A certain baseball player has a batting average of .334. At least how many times did he bat? (No, is not 3.)

The solution goes back to an algorithm of Gosper included in Hakmem , the remarkable Hacker’s memorandum of the early 70’s.

Of course, you can also check my book…

]]>

Over the years I’ve read articles about Twyla Tharp, the american dancer and coreographer. Somehow, I always found great affinity to her ideas. A few months ago, in a New York Times article, she described how she sometimes went for days without dealing with numbers (hid all the clocks in her house, for example) to give the intuitive side of her brain more prominence. Despite being a number theorist, I found the thought strangely appealing rather than heretic.

This morning at the Harvard Coop I found she just wrote a book “The creative habit” (published by Simon and Schuster) that I hadn’t seen before. I just had to buy it. Here’s a representative bit:

“I will keep stressing the point about creativity being augmented by routine and habit. Get used to it. In these pages a philosopical tug of war will periodically rear its head. It is the perennial debate, born in the Romantic era, between the beliefs that all creative acts are born of (a) some transcendent, inexplicable Dyonisian act of inspiration, a kiss of God on you brow that allows you to give the world the Magic Flute, or (b) hard work.

If it isn’t obvious already, I come down on the side of hard work. That’s why this book is called ‘The Creative Habit’. Creativity is a habit, and the best creativity is a result of good work habits. That’s it in a nutshell.”

Other ideas in the book remind me of L. Pasteur’s quote: “Chance favors the prepared mind”. To students I have often compared doing mathematics to fishing. You mostly sit around waiting for ideas to bite but when a big one does you better be ready.

Then a strange thing happened. I was sitting on the lawn of Harvard Yard reading the book when a guy comes over to ask me if he could borrow it. He wanted to get a picture sitting on the lawn of Harvard Yard reading a book. “Isn’t that what people do at Harvard?”, he said. When he finished posing for the photo and was ready to leave he asked me “And who the hell is Twyla Tharp?”

]]>

The standard reference (“the scriptures” as Schiffmann put it in his talk at AIM) is Ian G. Macdonald’s book: Symmetric functions and Hall polynomials. Second edition. Oxford Mathematical Monographs. Oxford Science Publications. The Clarendon Press, Oxford University

Press, New York, 1995. x+475 pp. ISBN 0-19-853489-2. (The second edition is more that just a cosmetic change of the first.)

This book is truly amazing. There’s absolutely no fat in it. This, however, can make it a bit hard to get into. In my case, the motivation to read it came from needing his description of the irreducible characters of , with a finite field (chapter IV), which is in turn based on the original description by Green. I was hooked.

Symmetric functions are at the very least a fantastic computational tool. They can be used to package quite a bit of information in a very concise form. Typically this information is combinatorial or related to representations of certain finite groups (the symmetric group, obviosuly, but also, as mentioned above, and so on). This is already evident in the description by Frobenius of the irreducible characters of the symmetric groups. In a situation that is quite typical the values of the characters (the character table) appear as the transition matrix (i.e. a change of basis matrix) between two natural bases of the ring of symmetric functions in infinitely many variables: the power sums and the Schur functions. (I can’t refrain from quoting Macdonald to the effect that elements of , itself a limit of polynomial rings, are neither polynomials nor functions but we have to call them something! See his paper “A new class of symmetric functions”, Publ. IRMA Strasbourg, 1988 372/S-20, Actes 20e Seminaire Lotharingien, p. 131-171, were he introduces the Macdonald polynomials. )

We also know from the work of Haiman that symmetric functions are relevant to the geometry of Hilbert schemes. My experience with our work with Hausel and Letellier is that they are also crucial to understanding the cohomology of character varieties. I am convinced that tough we use them mostly as a computational device the connection goes well beyond that.

Non-sequitur: Farshid Hajir pointed out during my talk in Amherst last Fall that Frobenius appeared in it in three different forms: 1) the Frobenius substitution, the absolute bread and butter of Number Theory, 2) the trace formula expressing the number of points of a character variety over a finite field in terms of irreducible characters of the target group and 3) Frobenius algebras, the building block of topological field theories. Frobenius was just amazing! I highly recommend his Collected Works; I would bet money that there’s still a lot to be dug up from them.

For the history on how Frobenius developed the theory of characters (a pretty convoluted one and hard to fathom from modern accounts) I recommend Representations of Finite Groups: A Hundred Years, Part I by T. Y. Lam.

]]>