The Human Side of Computer Science

Charles Dick-Kens, the inspiration for Dick Lipton and Ken Regan's joint alter ego, Pip. Image: public domain, via Wikimedia commons.

Charles Dickens, the inspiration for Dick Lipton and Ken Regan’s joint alter ego, Pip. Image: public domain, via Wikimedia commons.

Dick Lipton is a computer science professor at Georgia Tech who thinks P=NP, and Ken Regan is a computer science professor at the University of Buffalo who thinks P≠NP. Together, they are “Pip,” a Dick-Kens character.

Today I want to tell you about their blog, Gödel’s Lost Letter and P=NP.

Lipton and Regan write mainly about computer science, mathematics, and the history of those subjects, but they always put people first. Most posts start with a brief bio sketch of the person or people whose research they discuss, followed by a little background on the question they’re writing about and the technical content. In the “About” page, Lipton writes, “my real reason for making “who” central is to explain what researchers do when they work on hard open problems. They make mistakes, they go down dead-ends, they make progress. Also students may not know, but even experts sometimes forget, that ideas that we now see as easy were once unknown.”

The blog’s archives are extensive, and I’ve fallen down quite a few rabbit holes browsing through them. Rather than try to give any kind of a comprehensive overview, here’s an assorted (I dare not say “random”) collection of posts you might enjoy if you like creative, personal writing about some pretty technical topics.

I’m writing about this blog now because of how much I’ve enjoyed several of the most recent posts:
Avoiding Monsters and Non-Monsters, on Weierstrass’s “monsters” (continuous, non-differentiable functions) and some of their monstrous and non-monstrous friends
How to Avoid Leaving Clues, on how to have your memory and use it too
Is This a Proof? On the Weak Goldbach Conjecture and computer-aided proofs.
Proof Envy, on the theorems they wish they could use. “Call the phenomenon theorem envy: There are theorems that I have long known, but have never been able to use in any actual paper, in any proof of my own. I find this curiously unsatisfying, to know a cool result and yet be unable to use it in one of my papers.”
The Problem of Catching Chess Cheaters, on Regan’s recent work on the topic. Another good post on chess cheating is the Crown Game Affair.
In Praise of P=NP Proofs, on why reading P=NP proofs is a good idea, even if (so far) they’re all wrong.
I also stumbled on an entertaining post from 2010 about how well mathematicians have done at guessing whether conjectures are true.

Lipton and Regan suggested a few other posts to me, including some that are more computer-focused:
Perpetual Motion of the 21st Century, a double guest post from Gil Kalai, who does not think large-scale quantum computers are possible, and Aram Harrow, who does.
Quantum Chocolate Boxes, on using linear extensions to figure out whether your chocolate box is full of pralines or not.
Hedy Lamarr the Inventor, on amateurs in science and technology. In a similar vein, it takes guts to do research.
Finally, check out their Gödelinterviewseries, which gets more and more imaginative as it goes on.

This entry was posted in Mathematics and Computing and tagged , , , , . Bookmark the permalink.

2 Responses to The Human Side of Computer Science

  1. Paul-Olivier Dehaye says:

    I am not 100% sure that those blog posts have educational or recreational value, or a hybrid.
    I would suggest that the next blog post would not be a blog post, but rather a Google Hangout style video, where they use drama techniques (by lack of better words) playing different characters instead. I am about to explain this in my current running MOOC on massive teaching.

  2. Paul-Olivier Dehaye says:

    Forgot to post the address:

Comments are closed.