Root Fall

I wanted to learn, but I didn’t know what to search for. Root Fall is the answer I built: a discovery feed that does the looking, and a public shelf for what I keep.

How it works

Every morning a small tool on my Mac pulls papers from OpenAlex, a free, open index of scientific research, around a handful of themes: chaos, entropy, memory, meditation, micro ecosystems, and whatever those lead to. A small AI model running on the same machine writes one plain sentence about each.

I mark what’s interesting and skip what isn’t. Themes get stronger when I engage with them and fade when I don’t, and the tool suggests new ones along the way. What I keep goes into a library, where I rank it and write down my own take.

The shelf

rootfall.com shows the part of that library I choose to publish, ordered by how highly I rank each paper and how often I go back to it. The summaries are machine-written and labelled that way. Every entry links to the original paper.

No paid services: the model runs on my own computer, and the research index is open.