Articles

Why every paper of mine has a public data folder

I lost a data set in 2019 and I do not intend to do it again.

11 October 20231 min read

A person's feet in canvas shoes resting in a dry field with trees behind

In 2019 I lost the raw data for a paper. Not to a hard drive failure, which would at least be dramatic, but because I moved it into a folder called “old” and then renamed the folder.

I found a copy in an email attachment eleven months later. I have not trusted my own filing since.

What I do now#

Every paper gets a folder in a public repository with the raw data, the scripts and a short README. The README says what each file is and in what order to run things. I write it before I write the paper, which is the only reason it exists.

Does anyone use them?#

Four people have emailed to say they ran the scripts. One found a bug, which I was very pleased about, and which took me a day to fix.

Arrow keys to move, Enter to open, Esc to close.