From "Show HN: WhatTheDuck – open-source, in-browser SQL on CSV files" https://news.ycombinator.com/item?id=39836220 https://news.ycombinator.com/item?id=398362

From "Show HN: WhatTheDuck – open-source, in-browser SQL on CSV files" https://news.ycombinator.com/item?id=39836220 https://news.ycombinator.com/item?id=39836220 : > datasette-lite can load [remote] sqlite and Parquet but not yet DuckDB (?) with Pyodide in WASM, and there's also JupyterLite as a datasette plug-in: https://github.com/simonw/datasette-lite https://github.com/simonw/datasette-lite https://news.ycombinator.com/user?id=simonw https://news.ycombinator.com/user?id=simonw https://news.ycombinator.com/from?site=simonwillison.net https://news.ycombinator.com/from?site=simonwillison.net JSON-LD with https://schema.org/Person https://schema.org/Person records with wikipedia/dbpedia RDF URIs would make it easy to query on whichever datasets can be joined on common RDFS properties like schema: :identifier and rdfs:subPropertyOf sub-properties, https://schema.org/url https://schema.org/url, :sameAs, Plato in RDF from dbpedia: https://dbpedia.org/page/Plato https://dbpedia.org/page/Plato Today there are wikipedia URLs, DOI URN URIs, shorturls in QR codes, ORCID specifically to search published :ScholarlyArticle by optional :author, and now there are W3C DIDs Decentralized Identifiers for signing, identifying, and searching of unique :Thing and skos:Concept that can be generated offline and optionally registered centrally, or centrally generated and assigned like DOIs but they're signing keys. Given uncertainty about time intervals, plot concepts over time with charts for showing graph growth over time. Maybe philosophy skos:Concept intervals (and relations, links) from human annotations thereof and/or from LLM parsing and search snippets of Wikipedia, dbpedia RDF, wikidata RDF, and ranked Stanford Encyclopedia of Philosophy terminological occurrence frequency.