147 comments

[ 2.9 ms ] story [ 210 ms ] thread
This development should not be going on in github anyway.
That's rather a separate issue. Also, why? I'd certainly have a backup after the last takedown, but I doubt that any commercial host will be much better.
Maybe laws don't apply on gitee ? (Chinese GitHub)
Did I just read that correctly? WAT!

Try hosting code on gitee that mentions Winnie the Pooh, tank man, or _____________.

I think that comment was referring to Western IP laws.
It's funny that all of these type of comments mention Winnie the Pooh when I'm pretty sure this is a reddit injoke they've convinced themselves is important. Does anyone in China know they're supposed to be epically owned when you say that?
It doesn't really matter what the offense is, but if the CCP feels offended, say bye bye.
Sure, but has it offended the CCP so far? If not, celebratory posting as if you’ve been censored when you haven’t sort of makes you a poser.
On sr.ht at least the issues would stay distributed in contributor's email inboxes in the event of a takedown. Maybe they'd even keep it up despite questionable takedown notices, like youtube-dl.org's host did (not surprisingly unlike the MS-owned repo host, given it's an RIAA member).

Another solution to issue loss would be embedding the issues in the repo, as some already extant tools do.

Doesn't matter. Repos can be cloned and easily moved to another git service. The attempt to take down yt-dl itself showed that. A misguided attempt at it shutting spawned countless forks across multiple git hosting services that cannot possible all be taken down. A techy Streisand-effect if you will.
Yeah, I remember that. That was a day to remember, a victory for the open source. The explosion in forks across many other git services, was so widespread and loud that github (apparently under pressure from google) had to pull back on their decision to shut it down.
How did Googs put pressure on Microsoft to have Github change their mind? What concern was it of Goog's?
Google owns Youtube and youtube-dl doesn't serve google ads?
wow, total brain fart. of course. makes mental note to not be so quick to post right before bed
youtube-dl users probably are tiny drop among the adblock users who also do not see ads. And Google does not care about adblock.
I assume google cares about people who block ads because a month or so ago, google figured out how to show ads even though I am using a content blocker on iOS and macOS.
Will probably work on manifest v3 chrome then too
I don't think so. Disabling ads on youtube requires injecting JavaScript. It's not forbidden in manifest v3 (and it would be absurd to forbid it, as 99% extensions exist because of this).

For reference: I don't see ads on youtube using hand-crafted JS which was inspired by uBlock origin, so I think that uBlock origin should work as well.

Not really. The thing that makes youtube-dl useful is that it has one large dedicated community that can find ways to circumvent the changes youtube adds to prevent scraping. Basically, every few months or so youtube-dl becomes useless, and a huge distributed team has to kick in and find a way around it. When one of them does, they commit the change to one centralized place and it gets disbursed to everyone.

Without one centralized place, propagation of patches would take a lot longer, and youtube-dl would spend a larger percentage of time not working for most people. This is something the anti-piracy folks would be quite OK with, as it would lead to attrition among users even if not total eradication.

git is a decentralized vcs
the actual motivation of hitting people with legalese and takedowns isn't to literally take the software down, it's to drive up the cost of development, and to get people who maintain it to quit, which is probably why we're in this thread

given that software like youtube-dl basically needs to be permanently fixed to even work slowing down development is enough.

I think there should be new legislation to prevent such targeted nuisance takedowns, in a similar way to nuisance lawsuits.
Current legislation doesn't permit that specific takedown anyway, the procedure to remove a copyright infringing tool requires a court order. I still don't understand why anyone paid attention to it at all. The best I could find on it was that the RIAA was arbitrarily claiming they had the right.
> the actual motivation of hitting people with legalese and takedowns isn't to literally take the software down, it's to drive up the cost of development, and to get people who maintain it to quit, which is probably why we're in this thread

This only does anything useful to them if the software and all of the forks stop being maintained, regardless of whether the people doing it now are the same people who were doing it before. Which is one of the reasons why it's important to replace them, to deter this kind of attack. But they have been.

> given that software like youtube-dl basically needs to be permanently fixed to even work slowing down development is enough.

Not really. The demand for having youtube-dl or equivalent work is orders of magnitude higher than the amount of effort required to make it work.

Attacks like this also fortify development efforts because of the Streisand effect and the increased effort people are willing to put into something once it becomes a matter of pride to spite the attackers.

Sorry if I can't tell if you're telling the truth because I haven't been able to use yt-dl since the last non-github repo broke months ago.
(comment deleted)
For those unaware, there is an active fork called yt-dlp, that is (mostly) a drop-in replacement:

https://github.com/yt-dlp/yt-dlp

switched to it a couple of weeks back, working flawlessly for my needs.
Also bypasses throttling. Much faster, most of the time.
Sounds like a great recipe for getting banned from services. But, as long as it works, I'm all for it..
Apparently the trick is to pretend being an Android device, or something in that style...
You used to be able to bypass throttling by watching the video in browser during the download.
I have always wondered why video sites (who don't want downloaders for various reasons) serve their content faster than required for normal speed viewing. Well, I don't complain, but I wonder.
For buffering, since there might be a network hiccup sometime in the future. But generally also a better user experience if the user wants to skip a few seconds/minutes forward.

And I guess the bandwidth required also depends on the video bit rate/format/compression ratio, so you need to implement a lot of smarts to make sure you don't serve something too slow or too fast, and still people would be able to download the video, albeit at a slower speed. So, basically, it's not worth it.

The primary reason I've used ytdl is to work around youtube's crappy buffering.

In the past you used to be able to leave a youtube page open and it'd buffer the whole video but now it stops at like.. thirty seconds? Have fun watching when their servers can't serve.

Video streams are just a series of 2-second-long video files. It's no different from any regular CDN.

(The real money is in the compression processes.)

I was surprised by just how many small chunks (as you said 2 seconds, a few hundred kilobytes) the browser has to request and string together to make a video stream that play smoothly and that I can skip around. It's kinda amazing when you think about it.
They address this by having the video player stop buffering more than 30 seconds ahead. It's also less compute resources for the servers to send all the bytes for the next 30 seconds to a client at max speed, let the client disconnect and wait 3-5 seconds, then give them the next 3-5 seconds immediately when it's requested. If they only let you download video at a rate slightly faster than video playback rate than all clients would have a constantly open connection downloading video which would tie up more resources.
The work needed to bypass the rate limiting is pretty impressive: https://github.com/ytdl-org/youtube-dl/pull/30184#issuecomme...

An excerpt:

> ...the challenge algorithm is served in a mini-language within the minified player JS and therefore the specific algorithm could be extracted and executed by interpreting the mini-language without actually running the JS itself.

Quicker explanation of how Youtube is doing it is at the source code of pytube: https://github.com/tfdahlin/pytube/blob/bb890af3b271a616bbe1...

This is linked from the description of PR 30184 of youtube-dl but @varenc linked to a specific comment (talking about patching Windows builds, AFAICT) which might confuse some people.

I'll even save you the click:

> YouTube's strategy to restrict downloading videos is to send a ciphered version of the signature to the client, along with the decryption algorithm obfuscated in JavaScript. For the clients to play the videos, JavaScript must take the ciphered version, cycle it through a series of "transform functions," and then signs the media URL with the output.

There are still a lot of things I don't understand here but gotta give Youtube a slow-clap moment for their devious ingenuity.

> There are still a lot of things I don't understand here but gotta give Youtube a slow-clap moment for their devious ingenuity.

Well, sorta. They have basically come up with a DRM scheme by another name. They're always just speed bumps.

Whoops, didn’t mean to include the link to that comment... that’s a much better explanation!
As someone else said here on HN, in situations like this all it amounts to "mutually assured time wasting".

It reminded me ofthe old joke about two people entering an area infested by lion's, and one of them sits down on a log, and takes 5 minutes to replace his robust hiking shoes by a very expensive pair of running shows. The other says "why are you doing that - you can't our run a lion." The reply is "I don't have to be faster than the lion - just faster than you".

It makes prefect sense of course, when the lion has a choice. But in this case there is no choice. There is only one youtube. No matter how much time and expense youtube putting into their running shoes, youtube-dl / yt-dlp is always going to chase after them.

They are always going to win that race. While Google may be paying 5 or 10 of the cream of worlds programmers, those two open source projects ultimately how pools of 100's of equally talented people who occasionally want to download a video, and will happily contribute back a few hours of their time to do it. Which is probably how yt-dlp came about - youtube-dl slowed down, and someone said "fuck it - I'll do it myself". Worse from google's point of view, unlike splitinering a commercial endeavours, once those two projects are already contributing code to each other and may well merge again.

YouTube already specifically seeks to throttle youtuble-dl to an unusably slow speed, like 70kbps even if you only download one video a day.

What else are you supposed to do?

I always managed to get getter speeds using the aria2 parameter.
It's the only way to be sure I'll ever find certain videos again. And Google randomly bans users either way, they're not gonna stop if yt-dl disappears.
How would Google associate a YT download with a specific YouTube user?
Why would they do that when they can just make sure yt-dlp/youtube-dl doesn't work all together. :) No need to ban users who download content, just ban the service itself (ban as in build roadblocks to deny the downloader service access to your content).
They have been trying to break YouTube-dl for many years (and similar framework for iOS) and the code maintainers just need to update their code slightly to find another workaround and get it working again.
For now. I can see actual DRM coming some time in the not too far future.
No DRM needed. many sites just use hls.js type of chunk download tactic, it is already enough to prevent ytdl from downloading.
Is there any way to solve for that? I’ve noticed that and maybe that’s it or other chunked download stuff. That I haven’t been able to download. Never looked into this stuff enough to see what solutions are (besides brute force screen recording Ofc)
No one size fits all solution. You have to reverse engineer the way the script fetches and processes chunks and write a downloader or a plugin that instruments the running code and spits out the video data.

When wasm released I speculated that it will make this much more widespread and harder to reverse engineer due having more performace to waste on de-obfuscating at runtime. Feels nice to be wrong in this case.

Oh, I thought ytdl could deal with that! Do you know the specific issue and whether support is planned? Lack of HLS support would explain why there's no support for Zoom recordings.
I puzzled over this for a minute as well. Maybe the GP is referring to the user who uploaded the video.
> Sounds like a great recipe for getting banned from services.

If you are using your account for downloading, that is. I never use any valuable account when dealing with automated tools, it's just asking for trouble.

The best part is the native sponsorblock integration.

It adds chapter indicators to skip the sponsored sections, opening titles, begging viewers to inflate engagement metrics (please like, comment, and subscribe!), openings of music videos that aren't the actual music, opening and closing titles/credits, etc.

Basically cuts out all the bullshit you'd otherwise have to manually skip past by spamming the arrow keys.

by the way, out of curiosity, is this something content producers add to their videos or something youtube autogenerates? I've seen SmartTubeNext skipping ads and other bullshit and I was wondering where the feature came from.
Sponsorblock works by it's users marking parts of videos and then submitting them. Other users then can vote if these parts are right or not.
It also has a "skip to highlight" button that skips to the meat of the video and saves you 98% of its time.
How does it do this? Is this something the uploader of the video provides?
All the data is submitted by users of the browser extension (there's a thumbsup/thumbsdown system to flag bad submissions)
It can also be submitted via a Newpipe fork or mpv script.
yt-dlp works great. Worth to note that for macOS you don’t need their tap to install it, even though the yt-dlp readme says to use the tap.

yt-dlp is in the main package repository of homebrew, and the version in the main repo of homebrew is kept close to just as up to date as the tap.

So to install yt-dlp on a Mac that has homebrew installed, just run:

    brew install yt-dlp
Do you run it inside a sandbox?
I use mine via terminal the same way as YouTube-dl. The commands are the same too.
Not only have the yt-dl maintainers had to deal with the legal issues and the stress that comes with it, it appears they have a massive spam problem on GitHub which must make it difficult to efficiently deal with bugs. I was looking through the reported Issues recently due to a bug that caused yt-dl to stop working for a few days; a PR was quickly written to address the issue but it wasn't merged for several days. That's when I saw Issues like this: https://github.com/ytdl-org/youtube-dl/issues/29809
I think a good solution would be to make the bar higher to submit an issue. Disable Github Issues entirely and only allow true mailing list type issues that are significantly "uglier" and less "normie" friendly.
No. I contribute many useful GitHub issues, exactly because it’s easily navigable and comfortable to use GH issues.

If you want to raise the bar, I think a reputation system would be better.

Who know who is doing quite ok? Stack Overflow.
I think a system like how Slashdot moderation works. You get X points to rate issues or comments based on how many points you have. Comments get a value -2 to 5. Default view only shows 0 to 5. You really only focus your efforts on removing things with a negative rating.

I remember it working surprisingly well, especially browsing at 1 or higher, and I've never understood why no one has replicated it. 'Karma' systems must get better engagement.

If people can easily post spam issues what's stopping them from spamming upvotes on them too?
Slashdot use meta-moderation. Mod points weren't given to anybody; you only get a few once in a while. And then the way you use them is rated by meta-moderators. The result of that meta-moderation determines whether or not you get more mod points in the future.
(comment deleted)
Yes, the mailing list is the best regardless
In my experience maintaining software, raising the bar to report bugs is a really bad idea. 99.9% of your customers/users will just stop using your software when they hit a bad bug. Only a tiny minority will bother to report bugs.

Github issues is already a pretty high barrier (you need a github account to post).

I make software targeted at a technical audience, and only 10% of bugs are reported on Github. So by requiring Github you already lose 90% of potential bug reports.

(numbers in this comment are very rough estimates)

But youtube-dl is targeted at a technical audience (it's a command-line application; I know there are GUI wrappers but those aren't what that repo is for.) So by your numbers, youtube-dl would only miss 10% of legitimate bug reports.
What I was trying to say is that even for a technical audience having to log in on Github is an obstacle and will reduce the number of reported bugs.
That's not helping. Mailing lists are annoying to use for bug reports. It will discourage useful contributions.
When software makes it harder to submit issues I often just don't submit them. For example, the way VS Code's issue reporting system is configured means I can't use it. The result is that I haven't reported any of the bugs I've hit in the software - sure I could probably dig up someone's email address, but why would I waste the time?
>When software makes it harder to submit issues I often just don't submit them.

That's the goal.

Wow, I thought you are exaggerating and that it cannot be that bad... but the issue tracker is really, really bad.
The main problem with their issues seems to be that nobody has the ability to close anything. Sure, there are the likely accidental reports and the tutorial requests like the one you posted, but any large enough project is bound to get those.

Even that one you linked had someone from the community likely solve their problem and could be closed both for being irrelevant and for being solved. Several of the issues I checked were similar, probably bad reports but solved anyway.

When 4/8chan came to raid the audacity fork I created, we also had huge problems getting rid of spammy issues.

There's only a feature in github to forbid issues being created by "new accounts" which means if the account is older than 48 hours, it can still spam the shit out of your organization. And good luck trying to keep up with hundreds of accounts simultaneously spamming comments with racist and harrassment content. And this feature only exists for organizations, not for public repos owned by a person.

The only alternative is to make the github orga only accessible to triage members or maintainers, which would kind of defeat the point of issue tracking and bug reports.

Things would be so much easier if GitHub/GitLab/BitBucket had an automoderation bot that does this for you, even a simple blocklist of words or syllables would decrease most of the spam.

As most of the chans/spammers use lurking accounts that they reactivate for that purpose, they're very easy to identify. Their email addresses used in git are always from the same providers, and they always use the same kind of language, same kind of avatars, same kind of grammar and questions avoiding the main topic. So it's easy to spot them via a simple ML trained model (bert or tacotron models could easily classify this). Probably even a seq2seq model would be enough.

But yeah, guess this was just my two cents that GitHub has a moderation problem.

(comment deleted)
Are you the guy that people from 4chan attempted to kill over the fork?[1] Do you know when the case will be going to court?

[1] https://github.com/tenacityteam/tenacity/issues/99#issuecomm...

That whole story is debatable at best. Here's just some of the things that make you scratch your head and wonder how much of it is really true:

- Author of the fork claiming that he recieved 70+ calls on his pinephone and jokingly said it didn't crash. [1]

- He deletes said tweet moments after someone asked him how anyone might've found his pinephone number despite it not being on his website or social media. [2]

- Then he says he doesn't have any proof becuse he had to wipe/reflash his phone due to his pinephone crashing (this directly contradicts his previous tweet). After being called out, he immediately deleted his HackerNews post. [3]

- The author was posting f-word and r-words on his github bio. [4]

sources:

[1] http://web.archive.org/web/20210706000825/https://twitter.co...

[2] https://news.ycombinator.com/item?id=27746263

[3] https://postimg.cc/BPXX26W1

[4] https://archive.is/ZqK2T

(comment deleted)
The only answer I have to you: I learned not to give a damn about trolls. Maybe start something useful with your life instead of claiming things that aren't true?
The parent poster provided sources for all of their claims. It’s unclear to me what they said that was untrue. Care to elaborate?
None of it really has to be untrue to not draw the same conclusions. It’s possible that the phone crashed after some time, it’s possible they deleted messages and gave up because, as the parent comment suggests, they decided not to feed trolls.

Citing sources for specific claims does not mean that the overall conclusion is inherently true.

Auto-update and lock all issues with the text "Waiting for moderator approval", tag them with "approval queue". In a private repo for contributors, duplicate the issue with original text and relevant metadata. Approval in the private repo entails adding the "approved" tag and simply answers the question "Is this spam?". When that tag is applied, the original issue in the public repo is updated and unlocked. Any edits trigger a similar process, "Edit waiting for approval".

As issues are denied, a history of what type of spam or trolling is occurring is built up. That could presumably be aggregated and shared among the oss community, with concomitant spam arms race I guess.

Oh well, Youtube blocks Tor now so you need to use Invidious anyway.
Why did you previously need to use Tor?
To keep Google from maintaining a profile of my behavioral patterns and video viewing history based on my IP address.
Wouldn't be VPN a sufficient solution?
I think falsifying your user agent would be critical. Even with e.g. Mullvad how many requests will it take until one is deanonymized? Let's say I'm using a crowded VPN exit, still I can be profiled based on my user agent, the vpn IP and my requests. All it'd take them to connect the dots, would be me openning my apps/browser tabs on my phone/laptop/device without VPN for a second.
Not just user agent. Tons of other info is used for fingerprinting, from viewport and screen size to the list of installed fonts.
How would you go about them? So far I rely on Mullvad, privacy badger, ublock and random user agent.
The Tor browser employs a variety of anti-fingerprinting measures and there’s a project that’s upstreamed some of them back into vanilla Firefox.
(comment deleted)
To prevent predators from stalking me.
Why isn't the canonical fork of YouTube-DL hosted on a decentralized, anonymous system like Radicle[1]?

[1]: https://radicle.xyz/

That website is something else for sure. It really needs some help on explaining what it even does... (And it throws around the fancy buzzwords like Web3)
A web 3.0 website with a 1990’s style
More like early 2000's style, it almost wouldn't look out of place on geocities if not for the weird mobile-first-looking scaling.
I gotta say, this is the first website I've seen that shows less the larger the browser window is. When I maximize it one line says "code collab" and when the window is smaller it says "code collaborati". I think it wants to say collaboration, but it never actually shows up completely. Then there's "No more wall" and "explore the codela", whatever that's supposed to mean.

Whether this is on purpose or not, please try to decide whether you want to build a program or an art project.

On mobile, it looks readable but you are spot on about it looking more like an art project than a developer oriented website. Also the text is often unreadable due to contrast issues of gradient color in front of pictures.
probably because that looks like it's not good
All git repositories are decentralized and pseudonymous.
We're relying way too much on YouTube.
You can just use the other site that has millions and millions of videos. Oh wait... Well at least you can start another site aggregates millions and millions of videos. Dang it, then that site will run into the same problems.

Joking aside, I think the "competition is just a click away" applies here it's just that the competition isnt as good.

> the competition isnt as good.

The competition does not have Youtube's network effects.

The competition does not have Googles’s ability and/or willingness to burn that much cash.

Apple or Facebook or Amazon or Microsoft have the cash, but obviously none of them see a sufficient likelihood of sufficient ROI.

I cannot blame anyone who does not want to start a YouTube competitor, the moderation costs alone must be insane.

But for video creators interested in sharing their work, they can do it without Google, and it has never been easier in the history of the world.

> Apple or Facebook or Amazon or Microsoft have the cash, but obviously none of them see a sufficient likelihood of sufficient ROI.

Or they're back at their old tricks and are doing cartel stuff again; making secret agreements with each other about respecting each other's home turf.

I will go with the simpler explanation of video hosting and content moderation being expensive, and needing deep expertise in selling ads online in order to make any money on it.

Vimeo has been around since 2004 catering to businesses, who are far likelier to be paying customers requiring much smaller expenses for content moderation, and still does not earn a profit.

Doesn't Facebook have a video product?
A.K.A not as good? I mean couldn't you say that about any product? My little marketplace would be amazing if I had Tik Toks network effects.
The problem with creating a competitor for youtube is probably more challenging than most other web services.

You don't have to deal with network effects and technology, you also have to deal with a host of thorny copyright issues. There were times when youtube was in less than favorable legal waters. Without a Google-sized team of lawyers to deal with these things it would've probably been sued out of existence long ago.

We're not sure if YouTube even makes money
(comment deleted)
There are some content creators making stuff on other platforms but it goes from free to watch (youtube+ads) over to paid subscription and there is very little middle ground. From the youtubers I see doing alternate platforms they use the alternate as the "you'll find me here when I'm banned".
not just the viewers... many creators are concerned about yt shutting them down, demonetizing their videos, etc. thats why adam neely is trying to bootstrap an alternative platform called nebula.
Creators should also upload their content to YT competitors like Odysee or Rumble. They may not make much money there now, but those sites can only grow if creators start using them.

I am subscribed to Nebula too, but that is more of a curated site of already established creators. I'd rather see again the open and free YT of ten years ago.

How to get upvotes and look cool : We're relying way too much on <insert something that works really well>.
Perhaps an aside but recently using Google assistant to play music tracks from YouTube has mostly stopped working for me. Commands that previously worked such as "play $title by $author" now get a response that YouTube premium is needed to search tracks. some specific song titles in particular still get played but most do not. I also used to be able to say "play $album_title full album" but whereas it would previously start playing a video or playlist of the full album now I get a message advertising YouTube premium and then directing me to a mix based on the album title but not a playlist. Of course someone could pretty easily roll their own bot to provide this lost functionality but I think one day this too will be locked down via various means. Maybe the post scarcity era of YouTube and internet was never realistic to begin with. Was spoiled by the days of free Coursera, youtube, cheap Uber rides etc. I feel like in the future I'll be that old guy (or "boomer") telling the young people about the good old days of the internet and they'll just be rolling their eyes and dismissing such talk with some meme, incidentally to the benefit of Google/Alphabet
What do people use YTDL for? Is it to maintain personal records of videos you like?
It’s useful to download talks before jumping on a flight.

It’s also useful to pipe the output to mpv and have a native video playback experience.

This is especially useful on Linux where there is no hardware acceleration for video in the web browser.

>This is especially useful on Linux where there is no hardware acceleration for video in the web browser

Pretty sure there is some now, but I still would rather just watch media in my media player.

I cut chunks out of videos and put them into a home brewed speech therapy application for my son, to teach him to type.

(Bonus points is that my 5 year old appears to be getting the hang of bash)

I use youtube-dl for my cloud backup routine.

The only YouTube account feature I really use is playlists. I threw all of those in a bat file and ytdl downloads them for me into different folders whenever I start it.

Many videos I've downloaded this way aren't available on YouTube anymore. It's definitely worth it.

A lot of stuff will disappear from YouTube even after 12 months, let alone 5-10 years. If you want to archive your favourite videos then downloading is the only option.
Although there is now Premium LITE for 7 eur (which I finally subscribed to), listening to music mixes on YT is now atrocious.

You just can’t listen to a music mix anymore on YT. They play "ads" except the ads are complete music videos that run for 2-5 mins easily. It’s so dumb. Whatever happened to "soandso new album out now!". You don’t get an excerpt, you get an entire track of some random band, often times quite different genre that the mix you wanted to listen to.

It’s so fucking bad, it’s literally like YT says "hey you want to listen to ABC, why dontcha listen to XYZ instead".

I wouldn’t mind a short ad for some random album or artist when I listen to music, but now YT as made it almost impossible to listen to a good mix, when you get some OTHER music that starts every 5-7 mins, and runs for 2-3mins average…

EDIT : forgot to mention on one music mix (just progressive house dj mix), I’d keep getting an "ad" which is some electricity mini doc in Turkish that runs for 25+ mins. WTF !?

Youtube is actually making people want to skip and block ads because their "ad" program is atrocious for users.

Thankfully they do have a Premium LITE now but it’s still a bit expensive imho it should be just 5 eur.

And wtf I had to Google to find it. Within the apss I was never told that it was available, and clicking through the premium engagement was always showing me the 16+ EUR sub with Google Music etc.

But anyway even saying all that I still want to save videos that are invaluable to me, and just disappear randomly or turn "private" after few months/years.

What is the point of making a playlist when they show you unavailable items as the months go by?

AND, also you have sometimes vdeos that get edited. Like there was a really nice cafe ambiance and because some losers complained about being in the video, the uploader replaced it with a audio-only version.

So tldr when you see a video you REALLY like, you should DL it asap. You never know when it gets deleted.

I mostly use it to work around crappy youtube's buffering. Download a video and you can watch it without hiccups.

It's also handy for saving music that you want to listen to without being deliberately interrupted and paused by youtube.

I.e. it's all to work around youtube's crappiness.

Where I live for some reason YouTube Kids is not available, which means that YouTube is effectively not available on my kids device. A workaround for me is to grab the videos and put them on my Plex server, which at the same time allows me to curate what they can listen to. And since Plex allows me to locally sync content to the tablet, at works quite well for road trips too without requiring a data connection.
YouTube kids is useless anyway, especially in non-English speaking countries. It has extremely limited content.

And what it is completely missing are videos my kids watched that I was glad that they watched (as opposed to merely not minding) - crafts and ideas videos, not limited to but including minecraft inspiration videos, history for kids and other educational videos. Especially when not in English.

so that you can use it as a podcast offline.
My personal use case is syncing my music to my Garmin watch. I'm a paid Youtube Premium member but Garmin only syncs with Spotify and some other small-fry services, not Youtube Music. So I use ytdl on a daily scheduled job to pull down every mp3 from my Workout playlist. These get automatically added to my watch when I plug it in for charging.

I don't believe anything I'm doing is illegal, or if it is I doubt anyone would care. It's just a way for me to listen to the music I already pay for that works for me.

I think we should thank this maintainer for his great job, he made the internet a better place. I really hope he will now be working on some other project that he finds exciting, either that is painting or writing the next cool app :)

Thank you!