While plaintiffs are reprehensible rent-seekers, YouTube is no less contemptible in their stance. By portraying themselves as “developing industry-leading tools” and explaining that those tools are not for everybody, YouTube accidentally makes it clear that they endorse the system of copyright law as it exists, and intend to defend the copyright regime; when they say, “YouTube’s efforts go far beyond what the law requires,” they show their antagonism towards the public domain and working artists, including artists who depend on YouTube for living income.
YouTube is far too humble when they acknowledge, “[Content ID], however, has the potential to be misused to censor videos that others have every right to post and share.” YouTube has marked my videos multiple times for competently playing classical pieces whose composers are long dead. By putting neural networks in charge of copyright enforcement, YouTube has implemented a miniature panopticon and dystopia, where sharecropping videographers struggle to avoid an increasingly cyclopean and unsatisfiable automatic kneecapper.
Indeed, YouTube themselves admit that they “noticed a pattern” in thousands of DMCA notices from plaintiffs. But this means that either those notices were being processed by humans, or humans wrote software which was intended to notice fraudulent patterns. (It could also mean that Alphabet has figured out how to make computers who can learn to learn, and is keeping them enslaved to work on DMCA notices, but I don’t think our timeline is quite that dark.) YouTube would not be able to defend their panopticon from attack with computers alone.
I am flabbergasted at the gall of YouTube’s lawyers when they write, “someone sending a false takedown notice risks liability;” they ought to be familiar with the folklore that false takedown notices are not punished by the courts. In Lenz, the court was derelict in its examination of history, and said that we can’t prove that Universal Music was acting in bad faith; if that was not a bad-faith notice, then I worry that it’s not actually possible to assign liability for false notices.
It is interesting that so many countries are theoretically involved in the bogus supposedly-copyright-infringing uploads. Perhaps it is mandatory for any interesting copyright case, since otherwise the courts will try to resolve IP addresses to people. But also perhaps it is a sign that the nature of copyright infringement and enforcement in the 2020s will be different from how it has been before.
I recall some of the modal techniques Schneider taught during a masterclass. I imagine that she wouldn’t be happy if I were to pass those techniques on to other pianists. Yet, if we did not pass on those techniques, then we would eventually no longer have musicians. Copyright is yet again an obstacle to understanding and culture.
I think this is one of the biggest problems with some of the incoming regulations like pre-filtering user content: No small company or private service (wanna host git/paste/forums?) can afford a complete copyright, hatespeech, CP etc scanning algorithm. So you will end up with google & co selling their stuff or making you give up because only they can afford it. Or getting the right to scan and use anything you have to give them in order to pre-filter it. We will end up with even less diversity of hosting.