Convert LIT to PDF

Microsoft Reader launched in 2000 with LIT as its format and ClearType as its selling point, and for a few years it was genuinely the best-looking text on a screen. Microsoft stopped selling books in 2011 and retired the application in 2012. Nothing has written a LIT file since, and no other vendor implemented the format.

A LIT file today is therefore an object from a closed ecosystem whose doors were shut fourteen years ago. It will not open on current Windows, it will not open anywhere else, and the tools that still read it are maintained by volunteers. That is the entire argument for converting now rather than at some more convenient time.

Books bought from the Microsoft store carried digital rights management tied to an activation of Microsoft Reader on a specific machine, checked against Microsoft’s servers. Those servers no longer exist.

There is consequently no route — supported, unsupported, technical or commercial — to opening a protected LIT on a computer where it was not already activated. A reader who bought a hundred books in 2005 and reinstalled Windows in 2015 lost them, permanently and with no recourse. It is the clearest demonstration on this site of the difference between buying a licence and buying a file, and no conversion reaches it, because the conversion happens after the file has been read and reading it is the step that fails.

The format is a compiled HTML container derived from Microsoft’s help-file technology: HTML, CSS and images with a compression scheme layered over them. It is, structurally, a small website in a wrapper — the same idea EPUB arrived at independently a few years later.

The structure was documented well enough by independent work that free tools can read it, which is why an unprotected LIT converts cleanly when it converts at all. Text, chapter structure, images and most of the styling survive, and the table of contents generally becomes PDF bookmarks that a reader can navigate by.

Elsewhere on this site the argument runs the other way: converting a reflowable book to PDF freezes the type at one size and takes control away from the reader, which is usually a downgrade. That argument assumes the book is going to be read regularly on varied devices.

An archival rescue is a different job. What matters is that the file still opens in twenty years, on software nobody has written yet, and renders what it rendered today — and a fixed-layout document is better at that than any reflowable format, because there is less left to interpret. The fixity that costs a reader convenience is exactly what an archive wants.

A LIT has no pages. Like every reflowable format it holds text and structure, and Microsoft Reader decided where a screen ended based on the device and the size the reader had chosen. So a converter picks a page size, a typeface, a type size and margins, lays the whole book out once, and records where the breaks fell.

Choose those settings deliberately if the PDF is meant to be read as well as kept. A page roughly the size of a paperback with generous type is comfortable on a tablet; A4 with default margins is right for printing and unpleasant on a screen. Nothing about it can be changed afterwards, which is the whole property that makes it archival.

The selling point of Microsoft Reader was rendering: sub-pixel text on the low-resolution displays of 2000, at a time when everything else looked worse. That was a property of the reading application and the screen, not of the file, so none of it transfers.

It is worth knowing because it explains why these books can look unremarkable after conversion. What made them pleasant was never stored in the LIT; it was how one specific program drew them on one generation of hardware. The content is what survives, and the content is what was worth keeping.

The tools that read LIT are volunteer-maintained and the number of people maintaining them does not grow. Support for dead formats narrows over time — a plugin stops being updated, a dependency changes, a platform drops a runtime — and there is no organisation with an interest in keeping it working.

That makes this a task with a deadline nobody has announced. Converting a folder of old books is an afternoon; discovering in five years that the last tool which read them no longer installs is not recoverable. Do the readable ones in one pass and stop worrying about them.

If the books are going to be read rather than merely kept, convert to EPUB too. It is an open standard maintained by the W3C, implemented by dozens of independent applications, read by current Kindle hardware, and it keeps the reflow that makes a book comfortable on a phone.

The two copies serve different purposes and neither replaces the other: the PDF is the fixed rendering that will open in anything indefinitely, the EPUB is the version you actually read. Converting twice costs one extra pass over the same folder, and it removes the need to make this decision again.

One folder per book, or one filename convention across all of them, with the title and author in the name rather than only in the metadata — metadata is the part that gets mangled by a conversion, and a filename is the part that survives being copied between drives for twenty years.

Keep a short note of where the books came from and which ones could not be opened, so the next person does not repeat the work discovering the same locked files. And keep the original LIT files somewhere cheap: they are small, and they are the only evidence of what the protected ones were.