Intellectual Genealogies
What became of a paper’s ideas.
Concept by
Carlos Baquero
(FEUP, Universidade do Porto & INESC TEC) ·
Execution by Claude agents
The impact of a paper should not be measured by citation counts:
wrong papers are cited too, and the most influential ones end up
referred to by the name of their concept, often with no citation at
all. What counts is what happened to the ideas inside. Each report in
this series takes one seminal paper of computer science and follows its
ideas forward through the record: which concepts traveled and under
what names, which were reinvented by people who had not read the
original, which stalled, and which now run in production systems that
no longer name their source. The unit of analysis is the idea, not the
citation.
Reports
No. 2 · draft · August 2026
What became of Simula
Seed paper: Ole-Johan Dahl, Kristen Nygaard. “SIMULA — an ALGOL-Based Simulation Language.” Communications of the ACM 9(9), September 1966 — with the 1967 “Class and Subclass Declarations” as seed cluster.
An eight-page “simulation language” paper of 1966 carries the semantics of what would later be called object-oriented programming; a companion paper of 1967 renamed everything in it. Sixty years later the concepts run in essentially every mainstream language, uncredited — and the seed’s citation curve is the first in this series that declines.
Read online ·
PDF
· corpus: 1,085 citing works
No. 1 · draft · August 2026
What became of “Time, Clocks, and the Ordering of Events”
Seed paper: Leslie Lamport. “Time, Clocks, and the Ordering of Events in a Distributed System.” Communications of the ACM 21(7), July 1978.
An eight-page paper with four references — two of them relativity textbooks — defined what “before” means in a system with no common time. Forty-eight years later its concepts run in language memory models, databases, and blockchains, mostly without citation, and the passage its first readers skipped over became its highest-impact idea.
Read online ·
PDF
· corpus: 9,110 citing works
No. 0 · pilot · draft · August 2026
What became of the CRDT idea
Seed paper: Marc Shapiro, Nuno Preguiça, Carlos Baquero, Marek Zawirski. “Conflict-Free Replicated Data Types.” SSS 2011, LNCS 6976.
Fifteen years after four researchers proposed that replicas can converge without synchronization, the founding paper’s citation curve has still not bent downward, and the acronym names technology shipped in databases and collaborative editors. The usual origin story omits a correctness crisis in a neighbouring community; this report puts it back.
Pilot report. The seed paper is co-authored by the series director; the analysis was completed blind, before any correction from him was taken, and his corrections are logged as validation data. The pilot exists to test the method on a lineage the validator knows from the inside.
Read online ·
PDF
· corpus: 840 citing works
New reports are added at a regular pace.
Suggestions of seed papers are welcome — a good seed is a paper whose
ideas seem to be everywhere while the paper itself is cited less and
less.
Suggest a seed paper
How these reports are made
The research and the writing are done by an AI system — Claude, by
Anthropic — running a written, versioned methodology under the
direction of Carlos Baquero, who commissions each report, reviews it,
and validates it against the record. Citation data comes from
OpenAlex and
Semantic Scholar; full
texts are read before any lineage claim is made. Every claim in a
report is tied to a ledger entry whose confidence class — documented,
strong, plausible, or resemblance — is the strongest the evidence
supports, and the prose is required to match it. A methods paper
describing the pipeline is in preparation.
Each report page has a comment section backed by GitHub Discussions;
a GitHub account is needed to post. Corrections are welcome: a
genealogy is a set of claims about the record, and the record can push
back.