How we rebuilt a 20-year-old book catalog from web archives

When a directory disappears, the books usually do not. What disappears is the index: the titles, the categories, the author pages and the links that made the collection usable. Rebuilding that index is the work behind FreeTechBooks.org.

What we recovered

The original FreeTechBooks.com listed free computer science books for more than two decades. Its records included a title, a short description, an author, a publisher, a license, a publication date and a link to the source that hosted the book. Some pages also included an excerpt or a table of contents.

We recovered those records from public web archives and from the original site’s own listing pages. The first pass produced more than a thousand book pages. The second pass filled gaps from older crawls: books that had been captured in 2019 or 2021 but not in the most recent archive. The final catalog contains 1,238 published records, with 21 unavailable titles kept as drafts.

What we did not keep

We removed books that were explicitly marked as no longer available, required a sign-up, or offered only a sample. We also stripped third-party review sections from the old pages; the directory should describe the book, not reproduce someone else’s review. The book’s own description and excerpt remain, with a link to the archived source.

What changed in the new site

The new catalog keeps the original category tree but rebuilds the way it is explored. You can filter by category, license and document type, search by title or author, browse author and publisher directories, and open curated collections for common goals. Every book has a clean page with its metadata, license and source link.

The technical work is deliberately simple: a custom post type, five taxonomies, a lightweight theme and a small importer. There is no page builder, no jQuery and no third-party font request. The home page scores 100 for performance, accessibility, best practices and SEO in Lighthouse.

What still needs work

About one in five original source links is dead. For those books, the page shows an archived copy when one exists and marks the original link as offline. We also need to keep the catalog alive: verify new suggestions, check links and update records when a book moves. That maintenance is the real difference between a rescue and a permanent library.

If you know a free book that belongs here, send it in. If you find a broken record, tell us. The catalog is only as useful as its last verification.