Licensing — tiny_schiller
This repository contains two separable artifacts under two different licenses. Read this document before redistributing.
1. Source content — public domain
Friedrich Schiller died on 9 May 1805. Under the EU
Schutzdauerrichtlinie (Directive 2006/116/EC, life + 70 years) and the
equivalent term rules in the United States (works published before 1928
are in the US public domain) and the United Kingdom, copyright in
Schiller's original works expired no later than 1 January 1876 and
has not been recoverable since.
The text content of tiny_schiller.txt and tiny_schiller.parsed.txt is
therefore in the public domain and may be freely copied, modified,
redistributed, and used for any purpose, including machine-learning
training and commercial use, with respect to authorial copyright.
Edge cases
- Critical / scholarly editions. Modern critical editions (e.g. Nationalausgabe, recent Hanser editions) contain annotations and apparatus that remain in copyright. None are used here. The Projekt Gutenberg-DE digitisations use old, non-critical print editions whose body text is plain reprinting of public-domain works.
- Typography and spelling. Editorial styling, not original authorship; creates no new copyright.
- EU database
sui generisrights (Directive 96/9/EC). We do not redistribute Projekt Gutenberg-DE's database structure, metadata, or HTML — only the underlying public-domain texts. See section 3.
2. Pipeline code — MIT
All scripts, configuration files, build tooling, and documentation in
this repository are licensed under the MIT License (see LICENSE).
This includes:
- everything under
scripts/ - the
schiller_char/andschiller_bpe/andschiller_cl100k/build recipes - this file,
DATA_CARD.md, andREADME.md
MIT does not apply to the textual content; the textual content is public domain (section 1).
3. Source distribution terms — DraCor / GerDraCor (CC0)
This repository intentionally sources its Schiller play texts from the DraCor API (German Drama Corpus, GerDraCor). DraCor distributes GerDraCor under CC0, which is designed to be redistribution-friendly.
Accordingly:
- Upstream plain-text exports are retrieved from
https://dracor.org/api/v1/. - The raw concatenation lives in
tiny_schiller.txt. - The canonical training artifact
tiny_schiller.parsed.txtis a deterministic derivative produced byscripts/parse.py.
Always verify upstream terms yourself if you need strict compliance guarantees (policies can change). This repo's intent is: CC0 upstream for the digitization layer, plus public-domain underlying works.
References
- DraCor (API + project entrypoint): https://dracor.org/
- GerDraCor corpus landing / docs (CC0 claim is documented there): https://dracor.org/doc/
- EU Directive 2006/116/EC (term of protection): https://eur-lex.europa.eu/eli/dir/2006/116/oj
- EU Directive 96/9/EC (database rights): https://eur-lex.europa.eu/eli/dir/1996/9/oj