Bibliometrics — Who Cites Who
Authorship and citation network analysis of the pathology literature, including a WHO-classification strand.
Purpose
Analyse the pathology literature itself — authorship, citation networks, and topic trends — and publish the results as an interactive companion site.
Data used
Open bibliographic metadata; OpenAlex is named as a source. Explicitly no private or licensed datasets. A related strand, WHO-CNS, concerns the CNS WHO classification and is registered as the ongoing project “Who writes WHO CNS”.
Methods
Metadata fetch, author disambiguation, and network/community analysis. Visualization with igraph and ggraph in R, and D3 on the published site. Repos: bibliometrics (main analysis notebook), whociteswho, whowriteswho, whowriteswho.github.io (published site). Site: whociteswho.patoloji.dev.
Repository topology — corrected 2026-07-26
That flat list of repository names hid which one holds what, and cost time to work out. Read directly:
| Repository | What is actually in it |
|---|---|
WHO-CNS |
The analysis. 1,918 tracked files, Python + R + Quarto, the data, the audits. See WHO CNS — Who Cites Who. |
whociteswho |
Rendered output only — _site/ with 171 files, 48 analysis HTML pages, and an empty README.md. No .qmd, no _quarto.yml, no source of any kind. |
whociteswho is a deploy target, not a codebase: WHO-CNS’s post-render hook copies output/ into it, and Cloudflare serves it. Anyone opening it looking for the analysis will find nothing, which is why this is written down rather than left to be rediscovered. It gets no project page of its own — there is no source to describe.
Scope of this correction: whociteswho and WHO-CNS were read on 2026-07-26. The bibliometrics repository named in this page’s repo: field was not re-read, so everything above about the OpenAlex strand still rests on the original Quarto-chapter sources and is unchanged. last_reviewed is deliberately not bumped for that reason.
Current state / open questions
Active. Onboarding first step: run the rendering pipeline against the bundled sample export. Documented pitfalls: author disambiguation is never perfect, so the ambiguity rate must be reported alongside any author-level metric; OpenAlex updates continuously, so the metadata pull must be cached per analysis run for reproducibility.
Derived from: bibliometrics/whociteswho.qmd, projects/projects.yml, onboarding.qmd.