[HN Gopher] Extracting content from an LCP "protected" ePub
       ___________________________________________________________________
        
       Extracting content from an LCP "protected" ePub
        
       Author : ColinWright
       Score  : 132 points
       Date   : 2025-03-16 12:50 UTC (1 days ago)
        
 (HTM) web link (shkspr.mobi)
 (TXT) w3m dump (shkspr.mobi)
        
       | ferbivore wrote:
       | > You can, for sure, publish information relative to your
       | discoveries to the extent UK laws allow. After study, we'll do
       | our best to make the technology more robust. If your discourse
       | represents a circumvention of this technical protection measure,
       | we'll command a take-down as a standard procedure.
       | 
       | Disgusting behaviour, as expected from the publishing industry I
       | suppose. This "EDRLab" outfit appears to be little more than a
       | non-profit front for Hachette.
        
         | JadeNB wrote:
         | > Disgusting behaviour, as expected from the publishing
         | industry I suppose. This "EDRLab" outfit appears to be little
         | more than a non-profit front for Hachette.
         | 
         | The quoted block is indeed disgusting, but it gets even weirder
         | in the context of the full discussion, where the correspondent
         | seems to be trying some sort of intellectual blackmail on the
         | author of this article, saying that, as long as nobody talks
         | about its deficiencies, DRM can be kept weak and inefficient--
         | and so trying to blame increasingly cumbersome DRM on the
         | people who want to access their material, rather than on the
         | publishers. For example, with a nice and patronizing start:
         | 
         | > You've found a way to hack LCP using Thorium. Bravo! We
         | certainly didn't sufficiently protect the system, we are
         | already working on that. ... If the DRM does not succeed,
         | harder DRMs (for users) will be tested. I let you think about
         | that aspect
        
           | Akronymus wrote:
           | > we are already working on that
           | 
           | so they worked on making harder to crack DRM before being
           | informed of the weak DRM...
        
         | nocoiner wrote:
         | The author's response was just perfect though in both tone and
         | substance.
         | 
         | "As you have raised the possibility of legal action, I think it
         | is best that we terminate this conversation."
         | 
         | Once someone shoots off about getting the lawyers involved,
         | there's really nothing more than can productively be said
         | (unless, of course, you are prepared to get your own lawyers
         | involved).
        
         | kristo wrote:
         | They make a software to help libraries lend ebooks for free.
         | Without their DRM you either wouldn't be able to borrow ebooks
         | because publishers would never agree to it, or would be limited
         | to kindle/libby to read them. They're not perfect but how is it
         | bad behavior to say you'll issue a takedown notice if your
         | copyright material is republished? I don't really understand
         | why they're being treated as the enemy here?
        
           | ziddoap wrote:
           | > _I don't really understand why they're being treated as the
           | enemy here?_
           | 
           | The gross manipulation attempt is what did it for me.
           | 
           | " _We were planning to now focus on new accessibility
           | features on our open-source Thorium Reader, better access to
           | annotations for blind users and an advanced reading mode for
           | dyslexic people. Too bad_ "
           | 
           | The legal threat at the end wasn't very cool, either.
        
           | MyOutfitIsVague wrote:
           | The effect, very often, is to force anybody with specific
           | reading habits to buy Amazon or be unable to read their
           | books. This is especially bad if you don't just buy books,
           | but read through the library or especially if you get ARCs
           | (advance reader copies, for pre-release reviews). Advance
           | readers who don't have Kindle are jerked around constantly by
           | DRM and especially changes in DRM schemes. It's really hard
           | to not see this as collusion, as it suspiciously always works
           | for the benefit of Amazon and the detriment of every single
           | other person and company involved.
           | 
           | > how is it bad behavior to say you'll issue a takedown
           | notice if your copyright material is republished
           | 
           | It's not. That's not what happened here, though.
           | 
           | It is bad behavior when you threaten legal action against
           | somebody working within their rights to legally allow people
           | to read things that they paid for on devices that they've
           | paid for. The DMCA has specific carve-outs for
           | interoperability. Threatening legal action there is bully
           | behavior. I'd argue that the ethics are pretty clear-cut here
           | too. A ton of copyright law is incredibly badly balanced
           | against the consumer and even against small artists in favor
           | of the biggest players. If this was illegal, it would be the
           | law that is unethical.
        
           | grayhatter wrote:
           | > They're not perfect but how is it bad behavior to say
           | you'll issue a takedown notice if your copyright material is
           | republished?
           | 
           | That's not what they said. This is how you should have read
           | their reply:
           | 
           | > If your discourse represents a circumvention of this
           | technical protection measure, we'll command a take-down as a
           | standard procedure.
           | 
           | If you say something we don't like, if we think we can make
           | the argument that the information about methodology and
           | implementation you share for free, is circumvention of our
           | DRM, we'll follow our existing strategy to abuse the legal
           | system silence you and prevent you from sharing information.
           | 
           | > I don't really understand why they're being treated as the
           | enemy here?
           | 
           | Because they are the bad guy, they're actively working to
           | make the world worse. They're pretending like if it wasn't
           | for their kindness, access to these ebooks would be
           | impossible. But in reality they only care about controlling
           | other people by force. The legal threats, insane arguments
           | about how it's better if how their DRM works is a secret, the
           | intent of the software they're defending, and the messages
           | they sent; are just ways or attempts to exert control what
           | other people are allowed to do, or are allowed to know
           | 
           | I'd also like to discourage this argument generally
           | 
           | > Without their DRM you either wouldn't be able to borrow
           | ebooks because publishers would never agree to it, or would
           | be limited to kindle/libby to read them
           | 
           | The (unfair) translation of this is: If it wasn't me abusing
           | you, it would be so much worse! You should be saying thank
           | you that it's me abusing you! Not complaining about how you
           | don't like how you're being treated!
           | 
           | Everything can always be worse, the point is to make it
           | better, not accept something harmful.
        
           | lxgr wrote:
           | > how is it bad behavior to say you'll issue a takedown
           | notice if your copyright material is republished?
           | 
           | Which copyrighted material is TFA republishing?
           | 
           | And where's the takedown notice? So far, there only seems to
           | be an attempt of emotional blackmail ("take this down or
           | we'll have to deprioritize our accessibility efforts").
           | 
           | > They make a software to help libraries lend ebooks for
           | free.
           | 
           | Free to the library (?), but not free to the reader. (Readers
           | indirectly pay for it via certification fees paid by the
           | ereader vendor.)
           | 
           | It might well be the lesser evil compared to Kindle (closed
           | ecosystem) and Adobe Digital Editions (words cannot describe
           | the pain), but it's still a DRM scheme and as such restricts
           | reading hardware/software choice, so I can see how its mere
           | existence upsets people.
        
           | Y_Y wrote:
           | > They're not perfect
           | 
           | Oh wow. Better make an exception in this one specific case
           | then.
        
       | spudlyo wrote:
       | Granted, Readium LCP[0] may be one of the less odious DRM
       | solutions out there for eBook contents, however it's still DRM.
       | Handcuffs are still handcuffs regardless of how comfortably they
       | fit.
       | 
       | DRM is in my view is too often used as a cudgel to mandate
       | hardware and software level restrictions that take away the
       | control of our own computing devices and environments. I
       | personally hold that intellectual property isn't property, and is
       | increasingly becoming a net negative to humanity as a whole. In
       | the case of this article, there is an ominous threat of legal
       | action against the disclosure of the author's work, potentially
       | stifling the speech of a fellow hacker.
       | 
       | While I'm not unsympathetic to the plight of creatives, and their
       | need to eat, I feel like the pendulum has swung so far to the
       | interests of the copyright holders and away from the needs of the
       | public that the bargain is no longer one I support.
       | 
       | Because of this stance, I find myself uncomfortably on the side
       | of AI bros like Sam Altman who argue for the expansion of the
       | fair use doctrine. I see AI as an accelerant in the erosion of
       | IP's relevance and enforceability. With AI being able to crank
       | out derivative works at scale, it blurs the lines between
       | infringement and transformation. My hope is that the flood of
       | such content makes enforcement impractical, and that it will
       | further demonstrate that the IP emperor is naked.
       | 
       | [0]: https://www.edrlab.org/projects/readium-lcp/
        
         | ferbivore wrote:
         | Altman isn't on your side, or any side except his own. OpenAI
         | insists both that they should be allowed to train models on any
         | text they can gain access to, regardless of copyright or
         | licensing (https://openai.com/index/openai-and-journalism/)
         | _and_ that you should not be allowed to train models on any
         | text produced by their models (https://archive.is/2025013013215
         | 3/https://www.nytimes.com/20...).
        
           | spudlyo wrote:
           | His ability to speak out of both sides of his mouth is why no
           | one trusts him, and why I find it so uncomfortable to agree
           | with anything he says.
        
           | npodbielski wrote:
           | Exactly. OpenAI and altman would be very happy to say that
           | intelectual property does not apply to them but then enforce
           | that law when they talk about their own intelectual property
           | being used without their consent.
        
         | lxgr wrote:
         | > I find myself uncomfortably on the side of AI bros like Sam
         | Altman who argue for the expansion of the fair use doctrine
         | 
         | Why? Are you training LLMs?
         | 
         | I highly doubt they'll fight a pro-consumer fight completely
         | incidental to their objectives (if not detracting: don't need
         | to buy the source textbook if you can ask ChatGPT about its
         | contents as soon as it's released).
        
           | damnitbuilds wrote:
           | In a courthouse in the near future:
           | 
           | "So, yeronner, I think you will agree that I was well within
           | my rights to share a torrent of the new Batman movie, not for
           | people to watch but so they could train their LLMS on it."
        
             | lxgr wrote:
             | Technically, you've got a point: https://xkcd.com/2173/
             | ("We trained a neural network to enjoy the latest Hollywood
             | movies...")
        
           | spudlyo wrote:
           | The enemy of my enemy is my friend. Anything done to
           | strengthen fair use is in my opinion a positive outcome.
           | Happy to see copyright holders go toe-to-toe against a bully
           | their own size. If both OpenAI and entrenched IP interests
           | get bloodied in this fight, it's a win-win.
        
             | lxgr wrote:
             | Contrary to this popular saying, real world
             | friendship/enmity is not an anti-transitive relation.
             | 
             | I would expect most AI companies to be more than happy to
             | throw consumers under the bus if it affords them a carve-
             | out serving their own narrow interests.
        
       | lxgr wrote:
       | Very curious (and nervous, as I can imagine more bad outcomes
       | than good ones) as somebody that frequently lends ebooks from
       | libraries supporting LCP. (The only thing worse than "controlled"
       | digital lending would be no digital lending at all.)
       | 
       | LCP is as close to the platonic ideal of DRM as it gets:
       | Essentially no obfuscation; cryptography largely something to
       | point at when filing DMCA takedown requests. For better or worse,
       | I suspect we're about to get some new case law for what
       | constitutes an _effective_ technical measure.
        
       | captn3m0 wrote:
       | Readium was also responsible for the takedown on the noDRM repo,
       | a few years ago: https://news.ycombinator.com/item?id=29870151
       | 
       | The repo came back, but without the readium DRM code:
       | https://github.com/noDRM/DeDRM_tools/blob/master/DeDRM_plugi...
        
         | damnitbuilds wrote:
         | To further Streisand this, I think the restored repo is here:
         | https://repo.or.cz/DeDRM_tools-LCP.git
        
           | joecool1029 wrote:
           | It works. Though I've found it only works for LCP epubs, not
           | pdf's.
        
             | MyOutfitIsVague wrote:
             | It doesn't work for the newest encryption scheme for epubs.
             | Current epubs downloaded from NetGalley won't decrypt with
             | it, unfortunately.
        
       | baruchel wrote:
       | A more straightforward way to do it, IMHO, is to use Thorium in
       | conjunction with a Python script called lcpdedrm. If I remember
       | well, that script isn't available directly any longer, but it is
       | easy to look around and find copies of it. Then use Thorium for
       | locally saving the file and use the Python script for removing
       | the DRM.
        
         | miki123211 wrote:
         | Why do you even need Thorium for this?
         | 
         | If the files are just AES-encrypted and the links are there, as
         | the linked article suggest, it seems that it would be easy to
         | create a standalone script to download and decrypt these.
        
           | edent wrote:
           | (OP here) I discussed how the DRM works in an earlier blog
           | post - https://shkspr.mobi/blog/2025/03/some-thoughts-on-lcp-
           | ebook-...
           | 
           | Essentially, the key for decrypting the files is made up of
           | the book owner's passphrase _and_ the super-secret key
           | embedded in the closed-source binary.
           | 
           | I wasn't able to reverse engineer the binary or extract that
           | key.
        
             | ShrimpHawk wrote:
             | The key is b3a07c4d42880e69398e05392405050efeea0664c0b638b7
             | c986556fa9b58d77b31a40eb6a4fdba1e4537229d9f779daad1cc41ee96
             | 8153cb71f27dc9696d40f
        
               | edent wrote:
               | No, it isn't. That's the key for LCP Profile 1.0 - which
               | was deprecated some time ago - https://readium.org/lcp-
               | specs/notes/lcp-profile-upgrade.html
        
         | edent wrote:
         | The lcpdedrm script is, if I remember correctly, only suitable
         | for Profile 1.0, which has since been replaced with the newer
         | 2.0 version.
        
       | damnitbuilds wrote:
       | One should also note that Thorium is a remarkably crappy app to
       | read Epubs on.
        
         | lxgr wrote:
         | I've never used it myself, but I'd be really surprised if it
         | somehow managed to be worse than Adobe Digital Editions.
        
           | damnitbuilds wrote:
           | I lol'ed.
           | 
           | Indeed, the bar is low.
        
           | Finnucane wrote:
           | Well, it does have the advantage of being actively
           | maintained. If not necessarily very well. Recently, I had the
           | fun of noting that their app was no longer behaving correctly
           | with endnote links, and they suggested that my links were
           | coded wrong. I pointed out that they were not, were
           | standards-compliant and worked fine in _every other epub
           | reader_ I had. Even ADE! So I don 't know what they were
           | trying to do.
        
       | dvngnt_ wrote:
       | Do book drms even make sens? i can understand games, but how do
       | you encrypt words that are meant to be read. People used to
       | record music on the radio. It seems easier to ocr a book and
       | generate text that way.
        
         | snailmailman wrote:
         | A big place where this gets used is to make Kindle ebooks only
         | able to be read on a kindle.
         | 
         | Any time they update and change the DRM there's a brief period
         | where newly-released amazon kindle books essentially cannot be
         | read anywhere except kindle hardware and official kindle apps.
         | People have pretty consistently found ways around the DRM (for
         | now). But amazon is always trying to crack down on this.
        
           | myaccountonhn wrote:
           | It's also employed by digital libraries to enable lending of
           | books.
        
         | lxgr wrote:
         | You can always do OCR on the paper book, so if the easiest way
         | to circumvent some ebook DRM were OCR, the vendor would
         | probably consider that a resounding success.
        
           | goosedragons wrote:
           | Which is dumb because that still takes very little time and
           | effort. You can pirate any paper book in like 10 minutes with
           | a decent sheet fed scanner if you don't care about keeping it
           | bound. What a hurdle.
           | 
           | But hey, Grandma Martha can't read her Kindle version on the
           | new Kobo her grandson got her without buying a new copy.
           | Fantastic.
        
         | Zak wrote:
         | DRM on media almost never makes sense from an anti-piracy
         | perspective. Any reasonably popular book, movie, or TV show is
         | on The Pirate Bay within a single-digit number of hours of its
         | release.
         | 
         | It makes a lot of sense from a lock-in perspective, though I'm
         | not certain why that leads to publishers insisting on it.
        
       | easterncalculus wrote:
       | .mobi as a TLD for a book blog on the removal of DRM is
       | especially appropriate.
        
       | op00to wrote:
       | The only time to directly threaten legal action is when you don't
       | think legal action will achieve your goal, and you've got nothing
       | else. Otherwise, there is no benefit to threatening legal action.
       | Sure, you can imply it, but you gain nothing by showing those
       | cards.
        
         | NoMoreNicksLeft wrote:
         | But the goals that these people pursue are difficult to
         | understand or anticipate. You assume that because these people
         | are employed by some business entity that their goals are "make
         | more money" and such. That's rarely the case, even if we allow
         | for the unenlightened version of "make more money".
         | 
         | Instead, assume that many, possibly even most, are more
         | interested in being punitive out of some warped sense of
         | justice. Though it might cost them a million in legal fees to
         | discourage efforts to fix their DRM which in turn might only
         | preserve a few tens of thousands in sales or other earning
         | potential, they still doggedly chase these "goals". Sometimes,
         | I think, the lawyers even point this out, only to be overruled
         | by top management. This means that legal action can almost
         | always _achieve the goal_ , being that the goal is nothing more
         | than to make their targets miserable.
        
           | thisisabore wrote:
           | That's a pretty impressive way to imagine people's
           | motivations.
           | 
           | The likelihood of a non-profit outfit building open source
           | software being driven by a warped sense of justice and the
           | goal of making people miserable rather than, say, trying to
           | make sure they can continue their stated mission and ensure
           | publishers keep agreeing to digital book lending is pretty
           | bold. But sure, why not.
        
         | Mindwipe wrote:
         | This isn't true at all.
         | 
         | Legal action costs lots of money, no matter who you are, and
         | it's slow. Threatening legal action is usually the most
         | successful mechanism to make things happen, even if you have a
         | one hundred percent chance of winning your case.
        
         | lxgr wrote:
         | If this were true, i.e. in an ideal legal system, frivolous
         | lawsuits wouldn't be a problem at all.
         | 
         | In many actual US jurisdictions (and presumably beyond), it's
         | such a pervasive problem that there are now explicit "anti-
         | SLAPP" laws on the book against just that.
        
       | grayhatter wrote:
       | > KOReader never contacted us: I don't think they know how low
       | the certification fee would be
       | 
       | It's between 350USD (per platform) and 1,700USD per year. So the
       | possible range is between 1700USD, and 3k USD... Yeah, that's
       | totally reasonable for a FOSS project where the lower yearly cost
       | is 110% of the amount they make in donations every year.
        
       | favorited wrote:
       | > We were planning to now focus on new accessibility features on
       | our open-source Thorium Reader, better access to annotations for
       | blind users and an advanced reading mode for dyslexic people. Too
       | bad; disturbances around LCP will force us to focus on a new
       | round of security measures
       | 
       | This is so funny to me. "We might have gotten around to making
       | our software accessible, if it weren't for you meddling kids!"
        
         | Y_Y wrote:
         | Accessibility? Like with plain text? What a crazy world that
         | would be.
        
       | notpushkin wrote:
       | Yeah, I wouldn't share this so openly. Instead - like they said -
       | build a one-click downloader, then go fill the shadow libraries
       | with a bunch of trusted accomplices. (This is slightly less
       | legal, though.)
        
       ___________________________________________________________________
       (page generated 2025-03-17 23:00 UTC)