The training layer: a meta-aggregator / Project Gutenberg
Project Gutenberg
infrastructureThe corporaactive
The archive's reading. The archive reads Project Gutenberg as a source library held apart from the datasets made from it. Its own terms address two readers: the website 'is intended for human users only', and bulk and machine access go to mirrors and a machine-readable catalog. Its texts reach training and composition downstream, in the Pile's Gutenberg (PG-19) component (Wikipedia) and as allowed raw material for NaNoGenMo's generated novels; the name a text carries after that passage is outside its own pages, and Wikipedia records texts resold with references to the project removed.
| Relation to this census | infrastructure |
|---|---|
| Record type | library |
| The question it answers | training membership at the source: is a public domain work held as an eBook, at which landing page, and on what terms may a machine take it? |
| Kind | online library of free eBooks |
| Operator | the Project Gutenberg Literary Archive Foundation, which operates Project Gutenberg and accepts donations; per Wikipedia, a 501(c)(3) chartered in Mississippi in 2000, with Eric Hellman acting CEO after Gregory Newby died in 2025 |
| Evidence | digital library |
| Unit of record | an eBook: a digitized, proofread text with its own permanent landing page (e.g. www.gutenberg.org/ebooks/11), released in plain text and other formats |
| What it records | 'the world's literature' digitized and proofread by volunteers; per Wikipedia, mostly full texts of books or individual stories in the public domain in the United States, primarily works of literature from the Western cultural tradition, most in English |
| In the layer | public domain eBooks; per Wikipedia, the Pile carries a 'Gutenberg (PG-19)' component of 11.68 GB; Project Gutenberg's own saved pages say nothing of training |
| Under whose name | not recorded for authors in the saved own pages; Wikipedia records that 'Project Gutenberg' is a trademark that cannot be used in commercial or modified redistributions, that public domain texts may be resold once all references to Project Gutenberg are removed, and that books have been sold stripped of attribution to the project |
| Found and read back | each eBook has a permanent landing page; site search; a Gutenberg index file listing all books and a machine-readable catalog database (Offline Catalogs and Feeds); the website itself 'is intended for human users only', and bulk downloads go through mirrors |
| What the apparatus makes from it | not recorded on its own pages; downstream, per Wikipedia, the Pile carries a 'Gutenberg (PG-19)' component, and NaNoGenMo's rules allow a generated 'novel' that grabs 'a random novel from Project Gutenberg' |
| Method | volunteer digitization and proofreading; per Wikipedia, no selection policy (volunteers work on what interests them), copyright clearance recorded before a text is added, and close affiliation with Distributed Proofreaders |
| Coverage | geography: target audience United States persons; works not protected by copyright in the United States; most releases in English, many in other languages; period: eBooks since 1971 |
| Count | value: more than 75,000 free eBooks (own about page); nearly 78,000 items (Wikipedia, as of 22 March 2026); as_of: about page as read 2026-11-01; Wikipedia figure 2026-03-22 |
| Holdings | more than 75,000 free eBooks; releases in plain text and in HTML, EPUB and MOBI (Wikipedia); per Wikipedia, novels, poetry, short stories and drama, with cookbooks, reference works and issues of periodicals |
| Stated limits | 'Project Gutenberg’s target audience is United States persons over 13 years of age.'; 'Project Gutenberg offers no warranty or assurances about copyright status or freedom to access or use its materials outside of the United States.'; 'This website is intended for human users only.'; more than ~100 books per day are to be downloaded from mirrors; per Wikipedia, the Text Encoding Initiative criticized it in 1994 for not documenting which version of a text was digitized |
| Data access | free eBooks on gutenberg.org with permanent landing pages; mirrors for bulk download; the Gutenberg index file; Offline Catalogs and Feeds (a machine-readable database); an OPDS feed under stated rules for applications |
| Licence | the Project Gutenberg License (unread); 'All Project Gutenberg eBooks are completely free and always will be'; per Wikipedia, most books are public domain under US law and a few copyrighted texts are distributed with permission |
| Lifecycle | state: active; detail: operating since 1971; per Wikipedia, about 30 new eBooks a week (May 2025) |
| Version and time | count_wikipedia: 2026-03-22; count_own: more than 75,000 (about page); read: 2026-11-01 |
| Founded | 1971 (Michael Hart) |
What its record can settle
- whether a public domain work is held as a Project Gutenberg eBook, and at which landing page
- in which formats a release is available
- on what terms bulk and automated access is allowed
What it cannot settle
- which edition a text was digitized from (per Wikipedia, not always documented)
- whether a given eBook was carried into a given training set (the Pile's PG-19 component is known here from Wikipedia)
- how its texts are attributed once redistributed elsewhere
Kept apart
- NOT a scholarly edition: per Wikipedia, the version digitized is not always documented
- NOT open to automated access on its main site: its terms send bulk and automated downloads to mirrors
- NOT a training dataset: the PG-19 component is the Pile's, known here from Wikipedia
- NOT the downstream datasets or generated works made from its texts: the source library and what is drawn from it are different entities
- NOT any of its sister projects (Project Gutenberg Australia, Projekt Gutenberg-DE and others), which Wikipedia describes as independent organizations
Relations
- used_by: The Pile (Wikipedia's table of the Pile's sub-datasets lists 'Gutenberg (PG-19)', 11.68 GB)
- neighbour: Internet Archive (Wikipedia lists the Internet Archive as an affiliate: previous long-time backup distribution site and previous main host site)
- used_by: NaNoGenMo (NaNoGenMo's rules allow a generated 'novel' that grabs 'a random novel from Project Gutenberg')
Evidence
- self_description (the resource's own statement):
Project Gutenberg is an online library of more than 75,000 free eBooks.
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - founder (the resource's own statement):
Michael Hart, founder of Project Gutenberg, invented eBooks in 1971
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - method (the resource's own statement):
Since then, thousands of volunteers have digitized and diligently proofread the world’s literature.
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - free (the resource's own statement):
All Project Gutenberg eBooks are completely free and always will be.
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - mission (the resource's own statement):
To encourage the creation and distribution of eBooks.
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - operator (the resource's own statement):
The Project Gutenberg Literary Archive Foundation , which operates Project Gutenberg and accepts donations.
https://www.gutenberg.org/about/ · read 2026-11-01 · http 200 · sha256 4e1cfdb3edd2d17a… - audience (the resource's own statement):
Project Gutenberg’s target audience is United States persons over 13 years of age.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - limit_copyright (the resource's own statement):
Project Gutenberg offers no warranty or assurances about copyright status or freedom to access or use its materials outside of the United States.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - license (the resource's own statement):
See: the Project Gutenberg License and the Permission How-To .
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - human_only (the resource's own statement):
This website is intended for human users only. Any perceived use of automated tools to access this website will result in a temporary or permanent block of your IP address.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - mirrors (the resource's own statement):
If you want to download many books [1] manually or using an automated download software, download them from one of our mirrors, not from the main site.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - many_books (the resource's own statement):
Many books: i.e., more than ~100 per day
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - index_file (the resource's own statement):
If you want a list of all our books , download and save the Gutenberg index file
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - catalog_db (the resource's own statement):
If you want a machine-readable database of all our books, read the Offline Catalogs and Feeds page.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - landing_page (the resource's own statement):
When linking to a Project Gutenberg item, link to the main landing page such as www.gutenberg.org/ebooks/11.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - permanent (the resource's own statement):
We intend the landing page to be permanent.
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - opds (the resource's own statement):
Every application that uses our OPDS feed must:
https://www.gutenberg.org/policy/terms_of_use.html · read 2026-11-01 · http 200 · sha256 38e45e0cbc6a3f90… - count_wp (a third-party account):
Project Gutenberg had reached nearly 78,000 items in its collection of free eBooks.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - count_wp_date (a third-party account):
As of 22 March 2026
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - record_object_wp (a third-party account):
Most items in its collection are full texts of books or individual stories in the public domain .
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - western_wp (a third-party account):
These are primarily works of literature from the Western cultural tradition .
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - weekly_wp (a third-party account):
with an average of over 30 new e-books being added each week.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - formats_wp (a third-party account):
The releases are available in plain text as well as other formats, such as HTML , EPUB , and MOBI .
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - selection_wp (a third-party account):
Project Gutenberg is intentionally decentralized; there is no selection policy dictating what texts to add.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - clearance_wp (a third-party account):
Material is added to the Project Gutenberg archive only after it has received a copyright clearance, and records of these clearances are saved for future reference.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - trademark_wp (a third-party account):
"Project Gutenberg" is a trademark of the organization, and the mark cannot be used in commercial or modified redistributions of public domain texts from the project.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - resale_wp (a third-party account):
There is no legal impediment to the reselling of works in the public domain if all references to Project Gutenberg are removed
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - stripped_wp (a third-party account):
There have been instances of books being stripped of attribution to the project and sold for profit in the Kindle Store and other booksellers
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - tei_wp (a third-party account):
In December 1994, Project Gutenberg was criticized by the Text Encoding Initiative for failing to include documentation or discussion of the decisions unavoidable in preparing a text, or in some cases, not documenting which of several (conflicting) versions of a text has been the one digitized.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - foundation_wp (a third-party account):
In 2000, a 501(c)(3) non-profit corporation , the Project Gutenberg Literary Archive Foundation, Inc.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - ceo_wp (a third-party account):
After Newby died in 2025, Eric Hellman became acting CEO of the foundation.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - dp_wp (a third-party account):
Project Gutenberg is closely affiliated with Distributed Proofreaders , an Internet-based community for proofreading scanned texts.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - sisters_wp (a third-party account):
All sister projects are independent organizations that share the same ideals and have been given permission to use the Project Gutenberg trademark.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - ia_affiliate_wp (a third-party account):
The Internet Archive , previous long-time backup distribution site, and previous main host site.
https://en.wikipedia.org/wiki/Project_Gutenberg · read 2026-11-01 · http 200 · sha256 05de9e5673bcf1d5… - pile_component (a third-party account):
Gutenberg (PG-19)
https://en.wikipedia.org/wiki/The_Pile_(dataset) · read 2026-11-01 · http 200 · sha256 1ba82f3edd2a4647… - nanogenmo_use (a third-party account):
It could literally grab a random novel from Project Gutenberg.
https://nanogenmo.github.io/ · read 2026-11-01 · http 200 · sha256 d831aaf0d1a4efc0…
Unchecked
the Project Gutenberg License and Permission How-To; the History & Philosophy essays and Michael Hart's 2004 essay; the Offline Catalogs and Feeds page and the mirrors list; any eBook landing page (none saved), so the author and metadata fields of a release are not seen here
Cite: Lee Sharks, ed. "Project Gutenberg." The training layer: a meta-aggregator, traininglayerliterature.org, compiled 2026-11-01. https://www.traininglayerliterature.org/aggregators/project-gutenberg