Google has killed Discord's best music bot
pcgamer.com
pcgamer.com
It's not surprising that this was shut down like most "freeload until we get caught" hobby crafts.
But is there a way to build this service in a legit way on top of any existing music services?
For example, if you were able to give Groovy your Youtube Music API key at a subscription level premium enough for your Discord server, but that doesn't exist on Youtube.
I can imagine there isn't much upside for the big players in these hard spaces to support "glue app" value like Groovy, but how can you build Groovy without building your own music service?
Ehh, I'll be the devil's advocate, google makes money via tracking and ads. Both were avoided by using the bot. big brother is just protecting their business model.
Probably not because it would count as a performance and no consumer music services are licensed for that. You'd need something like whatever pubs and shops use.
So the way to go is by public advocacy and lobbying, not by looking for e.g. some technical solution.
I think the best way forward for this kind of thing is to not broadcast the music but instead have each user use their own local player with their personal Spotify/YouTube account and sync the playlist through a shared service.
[1] https://www.billboard.com/articles/news/9529030/twitch-metal...
My god we live in a strange world.
One can make a great painting and then sell the painting and the right to make copies of the painting to someone else. In this case, they would not be able to redraw that painting anymore than anyone else can.
Just because a creator can sell the rights does not automatically mean that every creator has. Let's assume good faith on the part of someone as proud of their work (and deservedly so) as Mike Shinoda.
One could always build an AI that train on the youtube data to provide background music that take a suggestion from discord and then fill in music that is generated with portions that are recognizable from copyrighted music but then does not do provide the whole song. User can then rate the background music and provide input to the AI.
Discord surely have enough lawyers and, you'd hope with a $10bn valuation, enough financial clout to wave at the record companies to get the kind of license they'd need to give their users background tunes.
But that would cost them time, money, and put them on the hook for shenanigans...
10 years ago, back when Windows Phone was a thing, Google wouldn't make an official WP app for Youtube. So somebody created MetroTube, a native YT client. Incredible app, with a completely seamless experience.
Then Google changed its API rules and it was dead in under a year.
Discord bots generally use a library like lavaplayer[0] to scrape YouTube for music streaming purposes; no API access per se. It was always at best a grey area, especially since many large bots rotate through IPv6 blocks to avoid YouTube's IP bans.
If you want a music bot that won't succumb to the same fate, you are better off self hosting something on a small VPS.
For anyone looking for a fun, quick project to do I recommend checkout out discord bots, discord.py is a great library that allows for some pretty cool stuff!
The most fun I've had is hooking it up to OpenAI so that typing "!ai some prompt" here will get the bot to respond back with OpenAI's response
Seems like Discord could estimate the value of Groovy to its users, and if non-negligible, could come up with their own license deal. Perhaps with Spotify or another company.
The people to be angry at are the recording industry.
It seems stupid that Google didn’t work something out as the main problem was that it didn’t play ads. So just play ads.
I dislike the trend of shutting down APIs as groovy should be able to use any user’s license to YouTube to play music into a channel (or Spotify or Apple Music or whatever).
This is the equivalent of playing a record over the phones to my friends and should cost me any extra money.
I’ve never used groovy in a channel with more than 5 people anyway.
They want to learn what you watch and listen to.
Bot usage anonymises your preferences and skips ads.
They don’t want APIs and only want listeners through their apps. Kind of lame they aren’t building technical solutions for actual user problems (like I’d like to listen to YouTube in my channel), they would be able to charge advertisers more if there are 5 people listening to an ad vs just one.
Yeah, that's against copyright law too. It's just that if you were doing that yourself, it would be on too small a scale for the record industry to fight against.
I believe the difference is the 'broadcasting' part. Same with streaming movies over a Zoom call.
Those don't sound the same to me.
I can’t deliver public performances. But I can certainly play songs for private settings. Usually explicitly, and if not, under fair use.
It doesn't fall under fair use, as that only covers the use of small portions of a work for a very limited set of uses.
But you might be right. It really depends on whether playing something over the phone is considered a retransmission of a broadcast, whether it's considered the creation of a broadcast, or whether it's considered a single performance.
Section 110 subsection 4:
(4)performance of a nondramatic literary or musical work otherwise than in a transmission to the public, without any purpose of direct or indirect commercial advantage and without payment of any fee or other compensation for the performance to any of its performers, promoters, or organizers, if—
(A)there is no direct or indirect admission charge; or...
And under section 101, 'Definitions':
To perform or display a work “publicly” means—
(1) to perform or display it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or
(2) to transmit or otherwise communicate a performance or display of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance or display receive it in the same place or in separate places and at the same time or at different times.
So - you're all good if you're just playing music for your friends at home.
But I can't find anything at all regarding taking a work and turning it into a limited transmission embodying that work, other than the code generally saying that broadcasts to the public must be licensed.
Anyway, while this was a fascinating exercise, I haven't actually come up with a solid answer. I take back my initial statement of "yeah that's against the law" and replace it with "...that might be against the law?" :)
What is the source is upstairs and the speakers are downstairs, etc.
I think the key factor is non commercial and to a small audience of friends. And that was why I thought fair use.
It’s not clearly defined, but Wikipedia describes it as “ flexible proportionality test that examines the purpose of the use, the amount used, and the impact on the market of the original work.”
So the whole song is played but the purpose is limited and there’s zero or negligible impact on the market.
it seems YT-dl has had its fight on the hill with google and won out for the time being, its possible a discord bot using YT-dl might wake a sleeping bear
youtube-dl + ffmpeg was the "traditional" way of doing it, but it didn't scale very well. These days, the majority of bots use lavaplayer[0] based solutions like lavalink[1] to stream audio.
There's also YouTube IP bans to be concerned about. Most bots get around it by rotating through blocks of IPv6 addresses.
The music content industry is very aggressive with using legal pressure to make sure they get their dime.
Copyright around music and film is insane and absolutely detached from reality. Having a copyright for 70 years after the last involved person's death? Because shoveling money into a grave is useful? Why isn't it like patents, 25 years and done?
the copyright & control world is rearing up very heavily here, with twitch, in so many places so many ways. unhosting the stream, simply providing metadata about what you are listening to, is a liberty these large powers that be will have a much harder time wrenching back from humanity.
scrobbling: https://en.wikipedia.org/wiki/Last.fm#Scrobbling
edit: I had to go just a bit further into this rabbit hole. Apparently the term "scrobble" is still a registered trademark of Last.fm, a company owned by ViacomCBS.
There's the open source ListenBrainz project, which uses "listen" instead:
https://www.w3.org/TR/activitystreams-vocabulary/#dfn-listen
I wonder if a "bot factory" would be more effective, so then you're in the same boat as torrent trackers (we just make the gun, we don't use it).