About

The library is enormous. The shelves are nearly empty.

The Public Domain Funnel

There are an estimated 30 to 40 million public domain works in existence — books whose copyrights have expired, belonging to everyone and no one. Transcribing them into digital form has been the work of dedicated volunteers for over half a century.

Project Gutenberg, the largest and oldest such effort, has produced roughly 78,000 texts in 54 years — a monumental achievement, but still only about 0.2% of the public domain corpus.

Quality varies widely even within that small fraction. Some transcriptions are meticulous; others carry errors from scanning, OCR, or human transcription. Most readers, however, never encounter these texts directly. They reach them through distributors — sites like ManyBooks and Google Play Books — that take transcriptions and redistribute them with little to no effort to format them as an enjoyable modern reading experience.

Standard Ebooks has set out to change this, taking selected Gutenberg texts and transforming them into beautifully designed, professionally typeset ebooks. They represent the gold standard. But even their output — roughly 1,400 titles in 11 years — covers only a small fraction of what Project Gutenberg has transcribed, let alone the public domain as a whole.

The gap between what millions of people are reading and what they deserve to be reading — in a format that does justice to the work — is immense.

A matter of pace

Standard Ebooks produces each book through careful, labor-intensive volunteer work. At an overall rate of about 127 books per year (their track record), matching Gutenberg’s catalog would take over a century.

We built a production pipeline that brings the time per book from months down to days, sometimes a single day, without sacrificing quality. Every book is sourced from a reliable transcription, compared against original page scans to correct errors, and enhanced with proper typography and semantic structure. The output is an EPUB that stands alongside the best available editions — and the build process is fully reproducible from source.

I say “we”, but this is currently a one-person operation

I can produce one book every 1–2 days. At a production rate of 3–5 books per day, we could match Standard Ebooks’ entire catalog in under a year. In five years, over 9,000 carefully edited titles.

If we could do even more than that — we might actually start closing the gap.

Make the public domain accessible today

  • What if instead of one or two transcription efforts and ebook creation/publishing entities, we had a whole network of them?
  • What if we shared our best methods?
  • What if we collaboratively shared our efforts to minimize the overlap of our work?

I think that is how you close the gap and make the public domain accessible today.

Our standards

  • Open source. The complete source for every book lives on GitHub, so anyone can inspect, learn from, or build upon our work.
  • Free culture. Ebook editions are released under CC BY-NC 4.0. The underlying source texts are believed to be in the public domain.
  • High Tech.
    • Our automated pipeline begins with Standard Ebooks’ best-in-class software, then takes it further. We automate more, to meet our goal of getting more books to you sooner.
    • Our ability to spot errors (errata) in transcriptions is systematic and thorough. Each book is compared character-by-character between the existing transcription and the original page scans, catching errors that human proofreaders miss — and catching them in a consistent way. Where others use their eyes, we use code…and proud of it.

Get involved

We’re a small project with large ambitions. We’d like to hear from you.

Every book we publish is open source. You can inspect how it was made, suggest corrections, or learn from the process. The door is open.

The catch

We don’t actually take volunteers. Instead, we encourage franchising and tech transfer. You want to make ebooks? Clone Gerrata, our errata finding software. Fork our pipeline, make it your own, and give yourself a nice publishing name. Agree to collaborate to minimize overlap and make more books, better, faster.

Why “Impression Editions”?

An impression can be a printing of a book, or the mark a book leaves on its reader. We do the former so we can have the latter.

In traditional publishing, an impression is fixed — ink on paper, immutable once the press runs. Digital publishing changes that. Every book we release lives on GitHub, where readers can report errors, suggest corrections, and feed back into the text. A book that once took decades to correct can be refined in an afternoon.

So the name carries our whole philosophy: we publish to leave a mark, and we built our process so that mark can keep getting sharper.