> GitHub’s CEO suggested that YouTube-DL won’t be reinstated in its original form. But, the software may be able to return without the rolling cipher circumvention code and the examples of how to download copyrighted material.
Which is frustrating. Isn't the information needed to do this provided by youtube themselves? This isn't some private key that wasn't supposed to be public; it's literally given out every time you view a video using it.
Also, the "examples" were tests that used a given algorithm, as there may be different methods used on different videos.
I also find it laughable to claim that a session token is a "rolling cipher". If the claim is to be taken at face value, it would ban any form of scraping, as most sites invariably use some sort of session token.
Which would be advantageous to a lot of big players with walled gardens, which is why I kind of hope this goes to court. Assuming a win it would further enshrine the legality of scraping.
Scraping an unwilling system is already essentially illegal, there's a long case history on this. It's only recently starting to change (via pure judicial activism) in HiQ v. LinkedIn. As written, the CFAA allows anyone to make a felon out of a persistent scraper.
> A federal appeals court last week issued a “hugely important” decision with potentially major implications for data journalists when it held that using computer programs to collect publicly available information from the internet — or “scraping” — likely does not violate the Computer Fraud and Abuse Act, the main federal computer crime statute.
The CFAA has nothing to do with copyright and that finding would only be relevant if the youtube-dl authors (or users) were being criminally prosecuted under the CFAA.
...which many entities would love to do. If it were illegal to extract data any way but through an official API, with a TOS and fees and everything, that would make a bunch of dying business models so much more viable.
It already exists and it's used to kill startups every day (including mine). Google gets a pass because they're real special. See Perfect 10 v. Amazon.
afaik it is against Google TOS to scrape data from their services e.g. play store but I can't find the evidence now. Some mobile store analytics agencies do it very carefully seizing the risk of their google account being banned forever. Haven't heard the precedents, still.
You can download mp4 files directly from Youtube. I've found a way to download videos right from the domain googlevideo.com . No 3rd party tools, any requested clarity.
I thought I'll make a post to explain this easy method, then realized Google would take down that method within a week, so I'm keeping quiet.
Well that's sort of what youtube-dl does, no? Except for you don't have to inspect network tab or merge anything manually, and you can choose quality and download playlists etc.
I'm not gaining anything from your assurance. I've found the best way - A first party method. No shady downloaders, extensions. I'm going to keep using it.
When Handbrake first came around, it installed DeCSS since it was meant to convert DVD media. Because of that, I was not allowed to use it at corp job. At some point Handbrake changed to not include DeCSS, but if you had VLC installed it would use the library provided by the VLC install. This met our lawyer's approval and allowed us to use Handbrake vX.X and higher only.
Just an anecdotal story about how software that offends lawyers can become acceptable by making certain changes. I hope the suggested changes to YT-dl will allow it to pass lawyer's smell test.
Couldn't the software simply fetch the token and display it to the user for manual input? It seems like the software's automatic use of the token is somehow breaking part of the DMCA. Is it actually illegal to display returned data from a web request? Is the manual input of "keys" legitimately a possible loophole? Did any of the early DVD rippers require manually adding keys?
Luckily they don't need to consider what some rich capitalist CEO thinks and don't need to return to MicrosoftHub and hopefully they won't. Especially given there are plenty of easy options available: self host gitlab outside of US, use something non-US, like gitea, etc.
But in case they do something that stupid, there will be forks. I will probably even make one fork myself with removed code reinstated.
EDIT: what the downvotes are for? Do you really want youtube-dl to appease Microsoft, RIAA or something?
Too bad, because there was no false information. Plenty of countries don't have not just anti-circumvention laws for copyright, but also DMCA-like laws and may even require you to obtain a court order before you can get something taken down. Except for a couple of countries, most countries could make it significantly harder to take down youtube-dl under such obviously false pretenses. Some countries are even particularly hostile and ignorant towards copyright-related demands from other countries.
Which is frustrating. Isn't the information needed to do this provided by youtube themselves? This isn't some private key that wasn't supposed to be public; it's literally given out every time you view a video using it.
Also, the "examples" were tests that used a given algorithm, as there may be different methods used on different videos.