App developer Elias Saba has had some bad luck with Digital Millennium Copyright Act (DMCA) takedowns. His Android TV app Downloader, which combines a web browser with a file manager, was suspended by Google Play in May after several Israeli TV companies complained that the app could be used to load a pirate website.
Google reversed that suspension after three weeks. But Downloader has been suspended by Google Play again, and this time the reason is even harder to understand. Based on a vague DMCA notice, it appears that Downloader was suspended simply because it can load the Warner Bros. website.
Application stores are basically random number generators. The worst possible applications, from non-functional garbage to ad-ridden gambling games designed to prey on children, make up the bulk of what’s on offer, but functional, useful applications spiral into Kafkaesque bureaucratic dark holes. Being a mobile developer in 2023 is a nightmare.
Github isn’t any better, see the youtube-dl fiasco for more information. The thing is that the DMCA’s anti-circumvention provisions don’t precisely define what is a “technical protection measure”, which means that what is a circumvention tool is also not precisely defined. So, service providers have adopted a doctrine of “in the presence of a DMCA complaint for circumvention of technological measures, takedown the software first to avoid any potential billion-dollar fines and investigate later”.
The FSF made a big mistake by not investing serious resources in one-click compilation (I should be able to double-click a Makefile or single .c file to open it with a hypothetical GUI-driven GNUCompile tool and get a compiled exe file instead of having to collect tools like make and GCC separately and type CLI commands to make them work), because source-code is typically protected by freedom of speech protections. For example, a link to the source code of DeCSS can be found in Wikipedia, but good luck finding a compiled version of it. This also applies to patented formats, you can distribute decoders and encoders in source code form because source code describes how a decoder or encoder could be implemented but doesn’t count as an implementation itself.
kurkosdr,
The rulings were so inconsistent from one courtroom to the next and under different technicalities decss could have very easily gone differently.
https://supreme.findlaw.com/legal-commentary/the-california-supreme-courts-recent-decision-on-decss-the-dvd-encryption-cracking-code.html
https://www.cnet.com/culture/court-decss-ban-violated-free-speech/
IIRC a federal appeals court rulings not only the source code bans, but actually barred providing links as well! Back in the day they were trying to trying to sue everyone in contempt of court. Eventually the whole saga ended because it became impossible to consider decss “private”.
Fortunately for decss, there was safety in numbers as millions of us nerds shared the code, effectively nullifying the court rulings and changing the course of history. Even so, I have little faith this view that source code is fair game would prevail consistently in front of every court for arbitrary cases. So although everyone withdrew and decss is settled, the matter more generally probably isn’t. If a case were brought against a small team by a multinational corporation and they have neither the lawyers nor a soap box on which to fight the case in the court of public opinion, unfortunately I think their odds of winning go way down.
IMHO the ban in the article is ridiculous and far fetched, but our sad reality is that we don’t have equal justice. Even if we strongly believe in the merits of one side, courts are always a gamble for the little guys.
Distributing as source code already has the big advantage that you can distribute software-patented stuff in countries where software patents are enforced. This is settled btw.
For the case of circumvention tools, it’s a fight worth fighting.
kurkosdr,
Obviously patents not the same as copyright/DMCA though.
Patents are supposed to be available to everyone to see how they work. That patent holders aren’t obligated to provide source code makes no sense to me, but even assuming we can independently implement patented algorithms and distribute them as FOSS, it doesn’t mean we’re allowed to use them without a license from the patent holder.
I guess you may be thinking users are better able to go “under the radar” when they have source code, which is probably true. But as a legal mater the fact that we have source code doesn’t give us the right to use those algorithms. There have been so many ridiculous patents issued and upheld by courts.
For what it’s worth, I find this whole thing with software patents allowing individuals to monopolize algorithms idiotic and extremely regressive for innovation.
Yes… yes it is (mischievous grin).
Sure, but it creates lots of work for lawyers, and I am talking about the paid-by-the-hour kind of work here, and if you you look at the lawmaking bodies of most countries they are packed with lawyers. This is how the absurdity of software patents became a thing.