ezyang’s blog

the arc of software bends towards understanding


A compile-time debugger that helps you write tensor shape checks

A run-time debugger allows you to see concrete values in a program, make changes to them, and continue running your program. A compile-time debugger allows you to see symbolic values in a program, reason about them, and write the rest of your program, e.g. filling in missing tensor size checks, for example. Here's an example […]

  • April 6, 2018

The Edit-Recompile Manager

A common claim I keep seeing repeated is that there are too many language-specific package managers, and that we should use a distribution's package manager instead. As an example, I opened the most recent HN discussion related to package managers, and sure enough the third comment was on this (very) dead horse. (But wait! There's […]

  • September 2, 2016

Interleaving stream processors

Ghost in the state machine A long time ago (circa 2007-2008), I wrote perhaps the single most complicated piece of code in HTML Purifier—one of those real monsters that you don’t think anyone else could ever understand and that you are really grateful you have a comprehensive test suite for. The idea was this: I […]

  • March 23, 2011

On expressivity

Wherein I make fun of functional programming advocates. In this essay, I’d like to discuss the ideologies of “imperative programming” and “functional programming” in terms of the language features they lean on: in particular, the mechanisms by which they allow developers to express themselves in less code. I propose that the set of features that […]

  • March 16, 2011

Dead Edward Day

Should software engineers be required to implement the abstractions use before using them? (Much like how before you’re allowed to use a theorem in a math textbook, you have to prove it first.) A bit like reinventing the wheel for pedagogical purposes. (I've been sick since Saturday, so it’s a Dead Edward Day. Hopefully I’ll […]

  • November 3, 2010

OCaml for Haskellers

I’ve started formally learning OCaml (I’ve been reading ML since Okasaki, but I’ve never written any of it), and here are some notes about differences from Haskell from Jason Hickey's Introduction to Objective Caml. The two most notable differences are that OCaml is impure and strict. Features. Here are some features OCaml has that Haskell […]

  • October 25, 2010

Purpose of proof: semi-formal methods

In which the author muses that “semi-formal methods” (that is, non computer-assisted proof writing) should take a more active role in allowing software engineers to communicate with one another. C++0x has a lot of new, whiz-bang features in it, one of which is the atomic operations library. This library has advanced features that enable compiler […]

  • October 20, 2010

The HTML purification manifesto

I recently sent Greg Weber an email about his xss-sanitize package, cautioning about his reuse of the pandoc sanitization algorithm for his own package. He responded (with good justification) that a mere caution was not very constructive! So here is my response, the “HTML purification manifesto,” which HTML Purifier follows and which I think is […]

  • October 8, 2010

A radical Hackage social experiment

Prologue. This post is an attempt to solidify some of the thoughts about the upcoming Hackage 2.0 that have been discussed around the Galois lunch table. Note that I have never overseen the emergence of a language into mainstream, so take what I say with a grain of salt. The thesis is that Hackage can […]

  • August 11, 2010