There’s been so much talk about “vibes” in programming lately that it gave me an idea for a Wikipedia tool.

If you accept the idea that Wikipedia page view data can be used as “fossilized attention” (indicators of public interest in a topic) then you can use information extracted from page view data — like derivation from average views (z-score) — as a benchmark of public interest.

Applying that analysis across an entire category to which a page belongs might find you other pages with similar scores, indicating that the pages could be “resonating” with a similar amount of public interest. Limiting the date analysis to a single day, and to a single category, raises the chances that the resonating pages might have overlapping roles in current events and will be relevant to each other.

This idea is yet another example of me playing with my favorite mental chew toy, “How do you ask for what you don’t know.” In this case I’m trying to group concepts together based on similar levels of public interest rather than any topical connection — temporal rather than contextual relevance, I suppose.

I’ve been trying it using Kendrick Lamar and May 4, 2024 for an example. The tool found both artists who worked with Lamar and artists who were mentioned in Drake diss tracks around that time; I was pleased to see that. The next step is to add an interface where users can turn the resonating pages into paired searches on Google.