GitHub

Five thousand years of writing.
Less than 2% translated.

Glintstone unifies the open cuneiform record — CDLI, ORACC, eBL, and a dozen more — into one platform where scholars can see exactly what's understood, what isn't, and where effort matters most.

Understand in New Ways

Glintstone aims to compound what CDLI, ORACC, and the broader scholarly ecosystem have built — making each resource more queryable, more connected, and more meaningful.

Exploration

New views across existing data.

With all open data connected, Glintstone allows you to explore and understand it in new ways: artifacts, composites, signs, readings, lemmas, and citations.

Search or use the API across all open data.

Curation

See the forest from the trees.

Connected data allows aggregation and analysis in new ways, linked to the ability to explore.

Automated and custom curation methods are baked in, across tablets, signs, lemmas, and meanings.

Scholarly Contribution

Catalyze your impact.

Future

Glintstone aims to unlock a new scale of scholarly achievement, and make sure you're recognized for your expertise and hard work.

We'll need lots of input from experts to make this happen, so we can build this system together.

Unified Artifact Pipeline

Every artifact tracked through five stages — from raw image to translated text. See what's complete, what's missing, and where scholarly effort has the highest impact.

01
Captured

The artifact exists as a digital photograph

02
Recognized

ML-assisted sign detection on the tablet surface

03
Transcribed

The text read into structured notation

04
Lemmatized

Every word linked to its dictionary lemma

05
Translated

Meaning rendered in modern language

Academic & historical context

Each artifact surfaces what scholars have written about its era, genre, and place of origin — drawn dynamically from trusted open sources. Administrative tablets sit alongside the political and economic world they recorded. Literary texts surface their reception history. Lexical lists connect to the scribal schools that produced them.

Cohesive Open Scholarship

Open data makes this possible, and there's over a century of expertise baked into that data.
Credit and source attribution are structural requirements.

CDLI Cuneiform Digital Library Initiative
CC0

Artifact catalog (353k records) with excavation contexts, seal descriptions, provenance, and collection data. Tablet photographs. Physical measurements.

Pleiades Ancient World Gazetteer
CC BY 3.0

Geographic coordinates for ancient places; provenience resolution for archaeological sites via ORACC GeoJSON data

OGSL Oracc Global Sign List
CC BY-SA 3.0

Cuneiform sign inventory (3,367 signs, ~15k reading values, Unicode mappings, MZL/ABZ concordance)

ePSD2 Electronic Pennsylvania Sumerian Dictionary
CC BY-SA 3.0

Sumerian lexicon (~21k entries), sign-level glossary data, lemma forms

DCCLT Digital Corpus of Cuneiform Lexical Texts
CC BY-SA 3.0

Cuneiform lexical tradition — sign lists, vocabularies, scribal training texts

CAD Chicago Assyrian Dictionary, ISAC / University of Chicago
TBD

21-volume Akkadian lexicon completed 2011 after 90 years; now being digitized by ISAC's Data Research Center

CompVis Heidelberg University
CC BY 4.0

Sign bounding-box annotations (81 tablets, 8.1k annotations); MZL-to-OGSL sign identity mapping

eBL annotations Electronic Babylonian Library, LMU Munich
CC BY 4.0

Sign detection training data; MZL sign concordance file used for cross-system identification

DeepScribe ISAC / University of Chicago
MIT

Computer vision pipeline for cuneiform sign detection and localization; primarily Persepolis Fortification Archive

BabyLemmatizer Sahala & Lindén, University of Helsinki
Apache 2.0

Neural POS tagger and lemmatizer for Akkadian; trained on ORACC annotated corpora

CDLI Publications & editions
CC0

Publications bibliography (16.7k pubs), artifact-publication links, edition histories

eBL bibliography Electronic Babylonian Library API
Research use

Fragment-level citation data; publication-to-artifact linkage via live API

CC0

Publication DOI enrichment; scholar ORCID identifiers for Assyriology-adjacent works

Free API

Citation and reference graph data keyed to DOIs from OpenAlex enrichment

KeiBi Keilschriftbibliographie, Tübingen
Research use

Comprehensive Assyriology bibliography; publication records with series designations and author data

CDLI Who's Who & ATF editor registry
CC BY-SA

Scholar name registry for cuneiform studies; ATF editor and author attribution

CC BY-SA 4.0 / CC0

List of Assyriologists for scholar name resolution; Wikidata QIDs as authority identifiers for disambiguation

Get Involved

Open source and built for collaboration.

Collaborate on GitHub

Browse the schema, review the data model, file issues, or contribute code. The entire project is open.

View Repository

Get Early Access

Sign up to get involved with early testing.

Collaboration Needed

Glintstone is in the exploratory stage and looking for ambitious collaborators.

Beyond cuneiform

The core infrastructure is built to be language-agnostic. Trust tracking, provenance chains, and annotation ecosystems apply to any extinct or undeciphered writing system. Cuneiform is the proving ground.

Linear B Linear A Egyptian Mayan Proto-Elamite Ugaritic Meroitic Indus Valley Etruscan Luwian Rongorongo Cypro-Minoan