highlight · part ii · 2

Not all documents read the same.

All four can become data with a codebook.


Text

Prose that's already text. Emails, articles, transcripts, clean PDFs.

Read the words.

Email Interview transcript Court opinion
Scientific

Manuscripts, patents, reports. Figures, tables, equations — the evidence isn't only in the prose.

Read the text — and see the figures.

Demographics table from a clinical paper Forest plot with hazard ratios Correlation matrix
Visual

Forms, invoices, typed letters, well-scanned pages. The layout carries meaning.

Look before you read.

Sample ballot Nutrition label Herbarium specimen sheet
Squint

Ledgers, registries, field notes. The same structure on every page — clean or damaged.

Read each page the same way.

Boston Register of Women Voters, 1921 1940 US Census population schedule Ellis Island passenger manifest, 1913

On Data Mint, we add more reader modes as we find document types that need them.

Four disciplines of reading. One extraction, on Data Mint.
No matter where you interface with AI: pick the reader that fits the document.

@karlrohe