MarginaGet Margina
記事

公開

What Metadata Lives Inside Your Ebook

Your library app calls it 'Unknown Author' or shows a blank cover, and it feels like a bug. It usually isn't. Ebook metadata is smaller and more optional than most readers assume — here's what's actually inside the file, and what isn't.

What it is

Every ebook carries two things: the text, and a small bundle of facts about the text — its title, its author, the language it's written in, sometimes a cover image. That second bundle is metadata, and it's smaller than most readers assume.

The EPUB 3.3 specification only strictly requires three fields: title, a unique identifier, and language. That's it. Author — dc:creator in the file's underlying markup — is optional. A cover image isn't a metadata field at all; it's a separate declaration pointing at an image file bundled inside the ebook. A publisher can ship a perfectly valid EPUB with no author listed, and nothing in the format stops them.

This traces back further than EPUB itself. The vocabulary EPUB borrows — title, creator, identifier, language, and eleven other fields — is Dublin Core, a standard that started at a 1995 workshop in Dublin, Ohio and became an official web standard a few years later. Every one of its fifteen fields was designed to be optional and repeatable. That flexibility is why the standard spread everywhere from library catalogs to EPUB files — and it's also the root of why so much ebook metadata is thin.

Why it matters

Metadata isn't one thing. It splits into a few independent categories: descriptive (title, author, subject — what you'd search or browse by), structural (how the file's chapters are ordered internally), and administrative (file format, creation date, rights). These live in different parts of the file and can fail independently. A book can have a perfect cover and no listed author. It can have both and still lack a working table of contents — a separate, structural problem with its own causes.

That matters in practice because your library app leans on these fields to do basic things: sort by author, group a series, show something other than a placeholder tile. When a field was never filled in, the app isn't guessing wrong — it's showing you, accurately, that the data doesn't exist.

Where you'll see it

The most common symptom is "Unknown Author" sitting under a book you know has a name on the cover. That happens because dc:creator was left blank somewhere upstream — by the publisher, or by whatever tool last converted the file.

Missing or wrong covers are a second, separate failure. Current EPUB files declare a cover through a manifest property pointing at an image; older files used a looser mechanism, a <guide> reference that some modern reading apps don't reliably parse. A file converted from that older format years ago can lose its cover along the way even if nothing else about it changed.

PDFs have it worse. There's no cover field in the format at all, so reading apps fall back to rendering page one as a stand-in thumbnail — which is why a PDF's "cover" in your library is sometimes a title page, sometimes a blank sheet, sometimes a scanned watermark. It's a workaround, not a real cover.

Conversion is where most of this happens. Moving a file between EPUB, MOBI, and PDF, or re-scanning and re-sharing a document, tends to drop whatever metadata the earlier version had — and the format's own long tail of document types only adds to the inconsistency. It's also part of why a PDF renders differently from one device to the next: the file is carrying less structural information than you'd expect.

How Margina handles it

Margina reads whatever metadata is embedded in the file you open. It doesn't invent an author, generate a cover, or quietly overwrite what's there with a guess. If a field is missing, you'll see that it's missing — not a fabricated placeholder standing in for it.

That follows from how the app works more broadly: no account, no server-side catalog lookup running in the background to "clean up" your files. Your document's metadata is whatever is actually in the file, read locally, nothing more. If you want to try it, open a PDF or EPUB with incomplete metadata in Margina and see exactly what shows up — available on the App Store and Google Play.

Frequently Asked Questions

Why does my ebook show "Unknown Author"?

Because dc:creator, the field that carries an author's name, is optional in the EPUB spec — only title, identifier, and language are required. A publisher or a conversion tool can skip it entirely and the file is still a valid EPUB.

Why is my PDF's cover just a blank or ugly page?

PDF has no dedicated cover-image field the way EPUB does. Reading apps fall back to rendering page one as a stand-in thumbnail, so whatever is on that first page — a title page, a scan artifact, a blank sheet — becomes the cover you see in your library.

Can I fix an ebook's metadata myself?

Yes. Tools built for the job let you hand-edit title, author, series, and cover directly, or look a book up against a catalog to backfill what the file never had. Margina doesn't edit or replace metadata — it reads whatever is already embedded and shows you that.

Conclusion

Ebook metadata is small by design, and inconsistently filled in as a result — that's a property of the format, not a bug in whatever app you're using. Once you know that only title, identifier, and language are required, most of what you'll ever see go missing — an author, a cover, a table of contents — stops looking like an error and starts looking like what it is: a field nobody filled in.