german-dynaword / data /gutenberg /gutenberg.md
grundtvig-bot's picture
cellar: parse in a forked child; gutenberg: licence section as H2
4ca878d
|
Raw History Blame Contribute Delete
5.31 kB
metadata
pretty_name: Project Gutenberg (German)
language:
  - de
license: other
license_name: Gutenberg
size_categories:
  - 1K<n<10K
task_categories:
  - text-generation
  - fill-mask
task_ids:
  - language-modeling
source_datasets:
  - original
domains:
  - Books

Dataset Card for Project Gutenberg (German)

German-language books from Project Gutenberg whose authors died in 1955 or earlier

Project Gutenberg is the oldest digital library of public-domain books. This subset holds its German-language titles: prose, drama, poetry and non-fiction, mostly 19th century, including the canonical authors (Goethe, Schiller, Kleist, Rilke, Fontane) and many titles that are otherwise only available as scans.

Dataset Description

  • Number of samples: 1.78K
  • Number of tokens (Llama 3): 198.60M
  • Average document length in tokens (min, max): 111.26K (69, 1.19M)

Dataset Structure

An example from the dataset looks as follows.

{
  "id": "gutenberg_2054",
  "text": "Iphigenie auf Tauris.\n\nEin Schauspiel.\n\nJohann Wolfgang von Goethe\n\nPersonen:\n\nIphigenie.\nThoas, Kön[...]",
  "source": "gutenberg",
  "added": "2026-10-30",
  "created": "1749-01-01, 1832-12-31",
  "token_count": 30092,
  "license": "public-domain"
}

Data Fields

An entry in the dataset consists of the following fields:

  • id (str): A unique identifier for each document.
  • text (str): The content of the document.
  • source (str): The source of the document (see Source Data).
  • added (str): The date when the document was added to this collection.
  • created (str): The date range when the document was originally created.
  • token_count (int): The number of tokens in the sample computed using the Llama 3 tokenizer.

Data Fields

An entry in the dataset consists of the following fields:

  • id (str): A unique identifier for each document.
  • text (str): The content of the document.
  • source (str): The source of the document.
  • license (str): the licence of the document, public-domain for this subset.
  • added (str): The date when the document was added to this collection.
  • created (str): The date range when the document was originally created.
  • token_count (int): The number of tokens in the sample computed using the Llama 3 tokenizer.

Dataset Statistics

Annotation Overview

License Information

The books are in the public domain and Project Gutenberg publishes them under the Project Gutenberg License, which permits redistribution of the texts. This subset strips Gutenberg's header, transcriber credit and licence block from every file — which is what those terms ask of a redistributor — and keeps the work's own text alone.

Beyond the US public-domain determination Gutenberg applies, the build enforces the German term as well: only books whose authors died in 1955 or earlier are included, so copyright has expired under the 70-years-post-mortem-auctoris rule too.

Additional Information

Source Data

The titles and their bibliographic metadata come from Gutenberg's catalogue dump (pg_catalog.csv), filtered to Language == de and Type == Text. Each book is downloaded as its UTF-8 plain-text file (ebooks/<id>.txt.utf-8) and Gutenberg's own boilerplate — header, transcriber credit and licence — is removed between its START OF THE PROJECT GUTENBERG EBOOK and END OF … markers, which is also what its licence asks of a redistributor. A file without both markers is refused rather than shipped with the licence text inside it.

Annotations

Not available: the propella annotation release does not cover Project Gutenberg, so metadata.parquet carries nulls and every document keeps its text. The subset can be annotated from its text in a later batch run.

Limitations

  • The created range is a bound, not a publication date. Gutenberg does not publish composition or edition dates (the dates it does carry are transcription dates), so the range written for each document is the author's life span — the interval in which the text must have been written. It is deliberately wider than the true publication date.
  • Only books whose author died by 1955 are included, so that German copyright (70 years post mortem auctoris) has expired as well as the US term Gutenberg works from. Books whose authors carry no dates are excluded entirely: the term cannot be asserted for them. Of 2,415 German text entries in the catalogue, 1,8xx survive that filter.
  • The books are volunteer transcriptions, so the error rate varies by book; remove_duplicate_text has already dropped duplicates created by re-releases.
  • Translations are included when their translator's dates are known and expired, so a small share of the text is translated rather than original German.