Here in my journal is an entry:

2016-07-27

1557

Oh man - I realize there’s the possibility of a github h-index for stars. n projects with n stars. Oh man. I’d have an h-index of 4 or so.

This is as breathless as a scientific work journal can get - if you can’t tell from the repeated use of “oh man.”

As I was working on another project, I came to reflect upon the architecture of academia. Mapping the structure of human intellectual pursuit has been an interest of mine for years. However, in this moment of architectural reflection, I began thinking about applications for such a line of research. Then, I began wondering about good research questions that could motivate the applications I came up with. Somewhere in the middle of all that, I thought about all of the researchers using GitHub without getting credit for their work, and that’s when it all came together in my mind.

After all, Academia really isn’t so different from GitHub. Stars are kindof like citations, and uploading is kindof like publishing. Perhaps the architecture of academia is kindof like GitHub, too? On the basis of this question, I turned to a bibliometrics algorithm called the H-Index that has proved useful for estimating academic influence.

If I could estimate influence on GitHub, then I could give people one way to get credit for the work they publish on GitHub!