Daniel Lemire's blog

, 2 min read

Why I hardly ever blog about my ongoing research

When I started my blog in 2004, my goal was to blog about my research. It never happened.

You may think that I am afraid a reader could steal my ideas, or that I might worry about looking silly. But I have no such fear. However, I am afraid it could hurt the quality of my work. I need a sandbox for my research. The purpose of the sandbox is to protect my ideas from the pressure:

What scientists really need is more time and more freedom to play with their ideas without pressure to fit in, to publish, to make up their mind. (Backreaction)

Working without pressure on your own ideas is essential for science. By opening up too early on your research directions, you can face at least two problems:

  • You commit to ideas before you had a chance of fighting self-delusion. Psychologists will tell you that once you write down and explain a position, you tend to stick with it. It is irrational, but true. I claim that this effect is less significant if you sketch ideas in your own private sandbox.
  • Early on, your ideas are still fragile: an expert can destroy them by apparently reasonable criticism before you had a chance to root them into a solid framework.

In the early stage of my research, I am the best person to hold a critical view on my own work. Only later, when I have had the chance to explore the idea to my satisfaction, do I need the criticism of others.

This is not just a theory. Several years ago, I shared an idea with a prominent scientist. He generously replied—in details—about why my idea was wrong. As part of his argument, he made a specific assertion which, if true, made my work uninteresting. A year later, I finally revisited the idea, and I determined that he was wrong, in a self-serving way.

Silence and long hours alone in front of a desk are necessary for strong original ideas.

Further reading: The notion of disputation arenas by David Brin.