The best YouTube downloaders, and how Google silenced the press

The claim that Google secretly wants YouTube downloaders to work doesn't hold up. Their focus is on delivering videos across a vast range of devices without breaking playback(and even that is blurring[0]), not enabling downloads.

If you dive into the yt-dlp source code, you see the insane complexity of calculations needed to download a video. There is code to handle nsig checks, internal YouTube API quirks, and constant obfuscation that makes it a nightmare(and the maintainers heroes) to keep up. Google frequently rejects download attempts, blocks certain devices or access methods, and breaks techniques that yt-dlp relies on.

Half the battle is working around attempts by Google to make ads unblockable, and the other half is working around their attempts to shut down downloaders. The idea of a "gray market ecosystem" they tacitly approve ignores how aggressively they tweak their systems to make downloading as unreliable as possible. If Google wanted downloaders to thrive, they wouldn't make developers jump through these hoops. Just look at the yt-dlp issue tracker overflowing with reports of broken functionality. There are no secret nods, handshakes, or other winks, as Google begins to care less and less about compatibility, the doors will close. For example, there is already a secret header used for authenticating that you are using the Google version of Chrome browser [1] [2] that will probably be expanded.

[0] Ask HN: Does anyone else notice YouTube causing 100% CPU usage and stattering? https://news.ycombinator.com/item?id=45301499

[1] Chrome's hidden X-Browser-Validation header reverse engineered https://news.ycombinator.com/item?id=44527739

[2] https://github.com/dsekz/chrome-x-browser-validation-header

I also don't buy this argument about YouTube depending on downloaders:

> They perform a valuable role: If it were impossible to download YouTube videos, many organizations would abandon hosting their videos on YouTube for a platform that offered more user flexibility. Or they’d need to host a separate download link and put it in their YouTube descriptions. But organizations don’t need to jump through hoops -- they just let people use YouTube downloaders.

No, organizations simply use YouTube because it's free, extremely convenient, has been very stable enough over the past couple decades to depend on, and the organization does not have the resources to setup an alternative.

Also, I'm guessing such organizations represent a vanishly small segment of YouTube's uploaders.

I don't think people appreciate how much YouTube has created a market. "Youtuber" is a valid (if often derided) job these days, where creators can earn a living wage and maintain whole media companies. Preserving that monetization portal is key to YouTube and its content creators.

> and the organization does not have the resources to setup an alternative.

Can confirm at least one tech news website argued this point and tore down their own video hosting servers in favor of using Youtube links/embeds. Old videos on tweakers.net are simply not accessible anymore, that content is gone now

This was well after HTML5 was widely supported. As a website owner myself, I don't understand what's so hard now that we can write 1 HTML tag and have an embedded video on the page. They made it sound like they need to employ an expensive developer to continuously work on improving this and fixing bugs whereas from my POV you're pretty much there with running ffmpeg at a few quality settings upon uploading (there are maybe 3 articles with a video per day, so any old server can handle this) and having a quality selector below the video. Can't imagine what about this would have changed in the past decade in a way that requires extra development work. At most you re-evaluate every 5 years which quality levels ffmpeg should generate and change an integer in a config file...

Alas, little as I understand it, this tiny amount of extra effort, even when the development and setup work is already in the past(!), is apparently indeed a driving force in centralizing to Youtube for for-profits

> As a website owner myself, I don't understand what's so hard now that we can write 1 HTML tag and have an embedded video on the page.

You acknowledge that it's not that simple:

> running ffmpeg at a few quality settings upon uploading (there are maybe 3 articles with a video per day, so any old server can handle this)

Can any old server really handle that? And can it handle the resulting storage of not only the highest-quality copy but also all the other copies added on top? My $5 Linode ("any old server") does not have the storage space for that. You can switch your argument to "storage is cheap these days," but now you're telling people to upgrade their servers and not actually claiming it's a one-click process anymore.

I use Vimeo as a CDN and pay $240 per year for it ($20/month, 4x more than I spend on the Linode that hosts a dozen different websites). If Vimeo were to shut down tomorrow, I'd be pretty out of luck finding anyone offering pricing even close to that-- for example, ScaleEngine charges a minimum of $25 per month and doesn't even include storage and bandwidth in their account fee. Dailymotion Pro offers a similar service to Vimeo these days, but their $9/month plan wouldn't have enough storage for my catalog, and their next cheapest price is $84/month. If you actually go to build out your own solution with professional hosting, it's not gonna be a whole lot cheaper.

Obviously, large corporations can probably afford to do their own hosting-- and if push came to shove, many of them probably would, or would find one of those more expensive partner options. But again, you're no longer arguing "it's just an HTML tag." You're now arguing they should spend hundreds or thousands per year on something that may be incidental to their business.

Here's me hosting a bunch of different bitrates of a high quality video, which I encoded on a 2016 laptop. http://lelandbatey.com/projects/REDLINE-intro/

The server is $30/month hosted by OVH, which comes with 2TB of storage. The throughout on the dedicated server is 1gbps. Unlimited transfer is included (and I've gone through many dozens of TB of traffic in a month).

People paying for managed services have no concept of bandwidth costs, so they probably think what you just described is impossible.

Bandwidth these days can be less than .25/m at a 100g commit in US/EU, and OVH is pushing dozens of tb/s.

Big ups on keeping independent.