This needs to be an open source script! I would love to use it to actually own my purchased books.

I find it puzzling that people are OK with allowing a mammoth, regularly hostile corporation know exactly what books they're reading as well as the exact details of their reading habits. Everything is accessible to Amazon - when and how often you access a book, how fast you read and when you linger on or return to a page. I wonder when they'll implement camera-based eye tracking so they know what word you're on?

The same public libraries that vigorously defend the privacy of our reading lists are simultaneously fine outsourcing all ebook access to Amazon where there's no expectation of privacy at all. Epubs at those libraries are now so well hidden they're not even mentioned anymore and access is buried multiple levels deep in the mandatory Libby app.

I love the ease of access and convenience of ebooks, but paper books are becoming more and more appealing by comparison.

Cool, but don't support Amazon at all.

I stopped buying books from Amazon long ago because their Android app is such dogshit. I can't imagine going to this much trouble to continue giving them money to reward their terrible engineering.

We need a better solution than a script reliant on a megacorpo service

Not including code makes me uninterested in using this knowledge.

Not all hero’s wear capes.

That was an interesting read.

If you wanna own your audiobooks you might wanna check out https://libro.fm/ - they allow you to download the bought audiobook without drm.
I gladly accept tips for similar sites when it comes to buying ebooks.

that was such a great read no joke. I love it when people use spite as a motivator.

I bought my first ebook from amazon

Just don't do that. If you give money to this company, they'll continue to go down the path they're on. Enshitification.

Spite: The greatest motivator!

That was a good read. I did think it was interesting that he decided to solve the randomized alphabet problem with the complex character recognition and image matching system he used. I mean, that was very clever and it clearly worked, so great! But an alternate option would have been the cryptographic method. There's a lot of software designed to crack replacement cyphers (especially in English) it probably would have been trivial to drop those characters into one of these and have it spit out the results.

Admittedly it would likely struggle with the other three alphabets, the italic, heading and italic heading alphabets, where there may not be enough words to be certain about success.

DRM's ever present weakness, if you can view it you will be able to crack it.

that’s one smart cat

I feel like a lot of what he attributes to malice or obfuscation is rather his unfamiliarity with the technical side of things.

Mapping numbers to non-sequential glyphs is exactly what happens in PDF, and PostScript before it. So is text positioning by Cartesian coordinates rather than in reading order.

Ligatures and multiple font variations have been around since fonts existed. Fonts that render incorrectly unless you use them in one specific way are legion (turns out font designers often have very little technical understanding, or just don’t care).

The only thing that sounds like obfuscation to me is changing the font encoding on every call. And if that is indeed its purpose, it’s a very poor attempt.

The level of pettiness by this person is immense, & I love it! We need more people like this in the world. Great short read. They obviously have some real skills & patience to pull this off, amazing!

Nice, hate cracking ebooks. The world needs more of this

What is evil is that this type of cracking is probably illegal through contract law (simply using the product).

I love that they did it.

For me, though; Yaar.

This post make me wish I could upvote more than once

Impressive work and an interesting article, but what I ultimately got from this was "don't buy shit from Amazon".

Very nice exercise and definitely easier to understand than tackling the DRM itself. For those interested, that's what DeDRM does: https://github.com/nodrm/DeDRM_tools