Period — words per decade
All years
Words per decade as a table
| Decade | Words |
|---|
| Languages | Period | Encoding | Link |
|---|
An open catalogue of digital editions of Renaissance-era written documents, c. 1450–1700
All years
| Decade | Words |
|---|
| Languages | Period | Encoding | Link |
|---|
Renascor is an open catalogue of digital editions of written documents from about 1450 to 1700: manuscripts, printed works, correspondence and inscriptions. Each record states who made the resource, which languages and years its texts cover, how large it is, how it is encoded and where to find it.
“Renaissance” is used here as a historical and cultural framework centred on Europe, particularly Western Europe, between 1450 and 1700. The catalogue also includes texts arising from Europe’s contacts with the rest of the world (trade, diplomacy, travel, migration, missions and colonial encounters), written on either side of these exchanges, in any language, and considered within their own traditions.
The catalogue points to resources and hosts no texts. It lists only resources that give a transcribed text, epigraphic editions included. Page images or photographs without transcription, uncorrected OCR and bibliographies are left out.
A facsimile shows a page; a transcription makes its text usable. It can be searched, counted, compared and mined at scale, and it lets students and researchers experiment with the texts themselves: analyse them, annotate them, play with them. It also makes the documents accessible. A screen reader or a braille display can read a transcription but not a page image, and readers can enlarge or reflow it, or read it without first mastering old hands and typefaces.
A digital text is most useful when it can be shared, checked, analysed and mined. Paywalls restrict that use and put scholars without institutional means at a disadvantage.
The catalogue was first compiled by hand, then set aside before the size of the task. It became feasible with agentic AI coding assistants, which over the past year have made it possible to search for resources, check them and refine the records at scale.
The catalogue lives in a public GitHub repository. Anyone can improve it by opening an issue to propose a resource or a correction. Anyone who disagrees with its criteria can fork it and build a catalogue that follows their own principles.
GitHub’s continuous integration and deployment also keep the catalogue maintained: every change is checked against the schema and tested, the site is rebuilt and published automatically, and a weekly search proposes new resources. Upkeep is where earlier catalogues failed. The Renaissance Knowledge Network has been frozen since 2017, and Folgerpedia’s list of digital editions of English Renaissance drama was last edited in 2016. Renascor is built instead as a closed loop: AI agents and contributors propose changes, automated checks test them, a person approves them, the site is republished, and the published catalogue is the starting point for the next round of searches and checks. Because the loop runs on its own, the catalogue can last and keep improving itself.
That is the idea behind this project: it had to cost next to nothing to run and need no grant, because cheapness and sustainability go together. The site is hosted free on GitHub Pages, the checks run on GitHub’s free automation, and each weekly search costs a few cents.
Automation never acts alone. Every change arrives as a pull request, and a person reviews and approves it before it is published.
Collections that provide freely accessible Renaissance texts are eligible even when their contents overlap. Each collection’s size is counted before content deduplication; deduplication applies separately to the Renascor Corpus. Subscription collections and directories without texts are excluded; exclusions are recorded with their reasons.
Anyone can propose a resource or a correction through a form on GitHub, or by pull request. Each record is one JSON file in the public repository, checked against a published schema.
The maintainer reviews every record before publication: that the resource exists, that it provides transcribed text from the period, and that it is described accurately.
A weekly search uses Mistral Large 4, chosen for its multilingual capabilities and planned open-weight release, to propose candidate resources as pull requests. Every addition is reviewed by a human.
Further AI agents will be added to maintain the catalogue: checking that links still work, updating each resource’s status, licence and figures when its host changes, and spotting resources that have moved or closed. Like the search, they will only propose changes as pull requests, for a person to review.
Each resource is classed by the kind of document its texts are: manuscripts (including archival documents and letters), printed works or inscriptions, or several of these when its texts mix them. The Type filter uses this classification.
Countries are the present-day countries where a resource’s documents were written or printed (ISO 3166 codes; England, Scotland and Wales are kept apart), as defined in schema/countries.json. They appear in each record and can be searched.
This catalogue grew out of the Renascor Corpus Project, which is assembling as complete a corpus of Renaissance texts as possible for AI training; the corpus has not yet been released. It shows which Renaissance texts can be read online as digital text today.
A collection’s word count covers its texts dated 1450–1700, including those also found in other collections. The Renascor Corpus total counts each text once (October 2026). Each record explains how its figures were obtained.
Periods give the first and last year of the texts: a single year when they coincide, “c.” for an approximate date, and a century counted as its hundred years (16th century = 1500–1599). The period filter keeps any resource whose period overlaps the chosen years; periods given only in words are not matched. The period graph is approximate and includes dates estimated from author lifespans. For 63 resources, the record also gives the host’s last update, such as a GitHub push or a new Zenodo version.
Download the whole catalogue with every field of every resource.
Data version: commit Data commit 452599a on GitHub of 15 October 2026, catalogue version 1.1.1. Please cite the version you used.
Please cite the catalogue when you use its data.
The data are released under CC0 1.0: they can be reused for any purpose without permission or attribution, though a citation is appreciated. The code is released under the MIT licence. Citation metadata: CITATION.cff.
Missing a resource, or found an error? The forms open an issue on GitHub (a free GitHub account is needed); no knowledge of Git or JSON is required. The maintainer reviews every proposal before it enters the catalogue.
Renascor aims to meet the WCAG 2.2 guidelines at level AA (last checked 13 October 2026). To report a problem, write to cag30@st-andrews.ac.uk or open an issue on GitHub.