Microsoft GitHub, DRM Enforcer, Bans Free Software
- Dr. Roy Schestowitz
- 2022-01-11 12:08:42 UTC
- Modified: 2022-01-11 12:09:21 UTC
Reproduced from Mobileread, as can be seen
here:
Initially, I didn't want create an account on this site and keep all discussion on GitHub (the fewer accounts one has, the easier it is to stay anonymous ...), but I guess with the GitHub being gone, it's about time to answer some of the questions here ...
Maybe GitHub hasn't been the best choice for a platform, but I didn't expect there to be DMCA claims when there have been none over the recent years in Apprentice Harper's repository. I guess, in the long term, I should move to another platform.
I received the first message from Github about the DMCA claim on January 4th in the late evening, with a time line of 24h to remove the "offending" content. Of course that deadline is rather short - I am obviously not using my "main" mail address for stuff like this, so I didn't check this account every single day, and only found out the repository was blocked (some time on January 6th) when I checked this forum thread and saw the discussion on January 7th.
The GitHub FAQ states that when one misses the 1-day window to make requested changes, one can request an additional time of 1 day to perform the changes. I requested that by mail on January 7th, but so far the GitHub support hasn't gotten back to me yet. Right now, I see the same page that you all see - repo unavailable due to DMCA. They could have at least given the repo owner access to update the code, but they didn't.
Rather disturbing that they are allowed to block a repo after just one day of no response (they could have given me a notification on the Github page itself, in addition to the mail, then I would have seen it before the deadline was over ...), and then don't respond to the topic for multiple days, but maybe their support doesn't work on weekends and they don't consider stuff like this urgent now that the repo's down and they did what they legally have to do ...
The goal is to hopefully get Github to restore the repository once they finally read my mail, then remove the offending code from the repository, and have the plugin no longer contain the offending LCP code on Github to comply with the DMCA request.
The DMCA request mentions nothing about the difference between library books and bought books. The request states that the original repositories (apprentice harper and so on) are not part of the takedown - not because they have blocks for library books, but because they don't support LCP at all. So I doubt adding a block for library books would have prevented this takedown (or, would be an acceptable solution to get the repository back). The guys behind LCP know how easy it is to edit Python code to remove such blocks, and I think with this plugin being the first public solution for LCP DRM removal, I guess they are more concerned with people knowing the algorithm, and they think that with a DMCA request for this repo they can remove that from the entire internet.
I don't want to piss off GitHub (and Readium?) even more by now creating a new account or repository. Even though it's probably fine as far as the DMCA goes (if there's no LCP code in the new repo), it certainly violates Github TOS to just make a new repo when there's a pending takedown. So I'm going to wait for the support to respond, which they are supposed to according to their own FAQ. If they don't, I guess the plugin moves to another platform.
As for the other topics being discussed here in the last couple days:
- Someone mentioned that based on the description of LCP in the takedown notice, this DRM doesn't sound so bad - maybe it doesn't, but there's one thing they are purposefully omitting in that description, and one they either deliberately or accidentally explained wrong. They are claiming that LCP is oh-so-open and doesn't lock the user into a proprietary environment.
Yeah, LCP is not as proprietary as Adobe or Amazon, but it's still proprietary. Yes, they have the source code available on their GitHub, but still require you to pay huge amounts for licenses if you want to use the code. A critical piece of source code for the project is missing on their Github, and you only get this code (with a very restrictive license) if you pay them. So, the code on the GitHub is useless, as if you forked it and built the code yourself, it wouldn't work.
And the other thing they omitted is the fact that there's (almost) no reader support. They claim the DRM doesn't hurt content accessibility, it lets users share content with friends, and so on. But that's only true if you're reading on a phone or computer, or if you have a very new eReader from particular vendors.
If the codebase would have *really* been open-source (meaning, I take the source code, built it, and get a 1:1 100% identical binary to the one they give to users, without paying for a license), AND Readium had support on all eReaders, I doubt I would have deemed it necessary to add LCP support. The main reason I added this support was not to "crack" books and share them with the world, it was getting them to work on MY readers ...
- The latest release of the plugin (10.0.2) does not yet support QT 6 / Calibre 6, but the latest commit on master already does. I doubt there's many people that have that downloaded, with the repo now gone. Though, even if GitHub decides to block the repo permanently and I don't find any other useful hosting, the plugin only required very small changes in two or three places that became apparent when reading the error messages, so it should be easy for others to fix that, if needed.
As noted
in Mastodon:
"RIAA showed that it was acceptable to use GitHub's (legally mandated) DMCA process for DMCA section 1201... Any tool like this should probably self-host their code repository at this point... Can we please kill these anticircumvention laws? Maybe then we won't need tools like DeDRM?"
Related:
