Spines

A spine traces the arc of a field — how it arrived at its present position, and where the next contribution sits.

Each spine is a working reference rather than a linear essay. It partitions decades of a field by what was actually being studied in each era, marks the bridges between them, and states plainly what carried across and what did not. They are maintained documents: entries get added, dates corrected, and framings revised as the field moves.


Available

Information and meaning
From Shannon information to embeddings
How information shed meaning in 1948 and spent seventy years getting it back — through semantics, pragmatics, distributional linguistics, and the geometry of embeddings.
1948
Shannon
1892-1970
Semantics
1955-1975
Pragmatics
1954-1990
Distributional
2003-2013
Embeddings
2017-2019
Contextual
Today
LLMs

In progress

Spines under development for the remaining research threads. Titles and framings are provisional.

Evaluation and measurement
Ground truth as an instrumented claim
How the field moved from treating labels as given facts to treating them as the output of a measurement process with its own assumptions.
In development — not yet published

On the word

“Spine” is used here in a specific sense. A spine is structural and load-bearing: it holds a body of work upright while it bends. The arc is the shape; the spine is the thing that has it. A spine is not a literature review, which aims at coverage, nor a survey, which aims at classification. It aims at sequence — what changed, when, and what the change made possible.