Skip to content

Some reading recommendations

April 26, 2018

Quanta magazine has an excellent article by Erica Klarreich on the recent progress on the 2 to 2 conjecture, which I have blogged about before.  The article does not go into the technical details, but gives a good perspective on what’s been done and what are the challenges ahead.

Eric Posner and Glen Weyl have a new book on “Radical Markets”  which I view as to some extent taking a “computer sciency approach” to some of the basic problems of society, economics and democracy. It’s “computer sciency” in the sense that their approach doesn’t place too high of a premium on “backwards compatibility” and also in the sense that, while they talk about transforming basic notions of property rights and even human rights in physical countries, their ideas might end up being more applicable to “digital countries” such as cryptocurrencies and online communities. Indeed, as Eli Ben-Sasson wrote here, many of the unanswered questions in cryptocurrencies are less about crypto and more about governance. (A third more technical sense is that it seems that some of their ideas, such as quadratic voting, correspond to replacing the 1 norm with the 2 norm, which is of course a theme we have seen before. )

Avi Wigderson posted a self contained version of the epilogue of his book on mathematics and computation. This epilogue gives a panoramic view of TCS’s past, present, and future, as well as its connections with its near and not so near neighbors. Regardless of your background, you will learn something from this book.

Finally,  I’ve recently started to be interested in connections between physics and algorithms, and in particular the relations between statistical physics and algorithmic difficulty. I highly recommend these lecture  notes by Afonso Bandiera, Amelia Perry, and Alex Wein.  In a tough year for our community, Amelia was another brilliant student that we lost much much too soon, and these notes are dedicated in her memory.

Comments are closed.

%d bloggers like this: