Ask HN: Would you watch a P2P YouTube replacement?
Imagine there is an app where you host a stream or upload a video and peers on the network automatically use a small amount of their bandwidth to mirror it.
There are also no ads. No tracking. Traffic is anonymous by default with onion routing. Has a good working GUI instead of a buggy slow JS page. And no censorship. And you can subscribe to crowdsourced filters for each type of content you don't want to see (such as spam, misinformation, low quality, etc). And there are desktop and phone versions that integrate fluidly.
Say it actually took off and had as much content as Youtube.
Would you actually use it? As a consumer, not a creator.
To make this fun, let's ignore technicalities like, "well all programs are crap now, you can't make one better", or "well the filter won't work for this reason". Just imagine it works. This is a thought experiment.
The question is simply one question: Would you use it? As a consumer, not a creator. Why or why not? Would you view the cryptographically verifiable news outlets? Would you use it to download or watch music and movies? Would you use it to watch tutorials? Would you use it to get angry about political happenings? Would you use it to work out? Would you use it to watch pro game streams?
I think you are describing PeerTube? [1] I've heard mixed results on HN about adoption and viewership but I don't know if there are any public stats to back any of it up. I like the idea but I've heard it's very slow.
[1] - https://joinpeertube.org/
Yeah, we're supposed to be imagining something fast here, and that works.
That doesn't look anonymous (people can see what content an IP is viewing), it has zero uptake, it's probably script kiddie tier software like every other p2p thing ever made, it appears to use federation which severely limits liquidity of content, and possibly even just does moderation by letting each member (server) of the federation be the moderator.
That doesn't look anonymous
Absolutely agree, anything p2p is not anonymous. The only semi-anonymous things I have seen are incredibly slow however Tor, I2P, Freenet and even those are not entirely anonymous and adoption is even lower with those protocols and platforms.
What architectural concepts do you propose that are both fast and anonymous?
It's not at all "script kiddie tier software". It's very actively maintained, has a large following and a surprising amount of instances.
Thats a lot of "ifs", but assuming it took off, from my point of view as a consumer, it does not matter if it runs on google's infrastructure, a raspberry pi, someone's basement server or on a potato. Your average consumer does not care about cryptographically verifiable news outlets either, they just want to watch videos.
I would watch a service with content that I want to consume.
It would be helpful if I could find that content.
Some content I want to consume like pirate movies and steamy romance audiobooks cannot find hosting on conventional services, but must content can still be hosted on YouTube. Any free speech oriented service is inundated with fake news propagators and people with really shitty opinions and a much as I would defend their right to speech I am not interested in thier content. Therefore I tend not to use existing free speech platforms because all the content is uninteresting. There is no need to host a regular tech tutorial on peer tube so that content I want isn't going to be there. If the platform mostly hosted stuff I liked, even if that stuff was lewd romance books and hooky hbo shows then sure. But it won't be.
People hate algorithms but live discovering the stuff they want easily. Any high privacy platform will have much lower doscoverability.
As a result, the things people find on here are going to be stuff the creator has built a marketing funnel for. People only do that if it is thing to make them money in a way they can't do on censored platforms. Therefore it would be full of crypto scammers and evil gurus.
A lot of existing alternative platforms have actually been high censorship just on a different political basis. If this imagined platform was genuinely high free speech then it would be full of illegal content like i2p and I'd avoid it for that reason too.
> Say it actually took off and had as much content as Youtube.
This is a huge assumption. There are two moats around YouTube right now:
1. Discovery processes and sharing. You claim an assumption that these exist, and specifically don't want to talk their details, so I'll skip details on this as well.
2. Ad revenue share. For good, bad, and sometimes very ugly, a lot of content just doesn't exist without YouTube's monetization. You assume "no ads" and don't mention a business model, how do you expect to pay for content?
Restricting to just the consumer standpoint, the answer is easy: if an app had the content I wanted but a better ad experience than YouTube I'd be very easy to convince to use it. I don't even need "no ads", I don't mind advertising paying for or subsidizing content I watch. (I grew up with broadcast TV.) I do wish there was less tracking involved (and find targeted advertising creepy/terrible) and I do wish for some sort of curation of advertising again. (A broadcast TV station would never interrupt the middle of a 5 minute music video to play an hour long infomercial. A broadcast TV station would also have a lot to answer from, including the FCC, if they allowed nazis and other hate groups to buy hour long infomercials.)
All the other stuff mostly doesn't matter to me as a consumer. I don't care about traffic anonymization via onion routing or if it is peer-to-peer or centralized. I don't care about cryptographically verified channels or censorship resistance. I care about filters obviously, but I only care if they are crowdsourced in so much as I distrust a lot of the biases of ML-based recommendation algorithms and would maybe prefer something with more paid humans (labor) involved.
All I care about as a consumer is does it have the content I want to watch. So indirectly I care a lot about the business model of whatever it is. If it is monetized well and attracts the creators that I care about, then I use it, it's as simple as that. The tech doesn't matter. You just need to be very careful in assuming you can beat today's YouTube on attracting creators given their moats.
I don't think what he meant with "no ads" would translate to no monetization.
I do not have a straight answer to this but there could be an alternate monetization.
Youtube's official monetization (and any ads in general) is an old concept that is prone to abuse. People dislike it, creators dislike it.
Youtubers now include sponsored segments in their videos. It's the main alternative that developed on the platform. Why? Because monetization just isn't good. It pays little and it's unreliable. Youtube arbitrarily demonetizes without warning.
And what's in it for the viewer? Both ads and sponsored segments are known to push scams on consumers. Because neither google nor the youtubers really investigate what they push on users. They just get the paycheck to display what advertisers pay for.
On a final note
>You just need to be very careful in assuming you can beat today's YouTube on attracting creators given their moats.
No service is eternal. Everything dies eventually. Something huge that took years to build can go down in flames overnight. You never know. Then there's a vacuum for something better to fill the hole. Will the next generation embrace P2P? Possible. New generations are tech savvy. They are aware of their digital presence. They are aware of censorship.
> I don't think what he meant with "no ads" would translate to no monetization.
I know. I wasn't assuming one way or another that they had a monetization plan. I was making what I thought were two simple points:
1. Hypothetical monetization plan matters a great deal more than hypothetical tech stack.
2. The raw concept of "ads" are not the problem. Right now, there is a confidence crisis in advertising, but I don't think that's a problem with advertising as a concept or a revenue stream. I think that's a problem with adtech since the 1990s. Tech companies have hyper-focused so much on the "tech problems" of advertising over the people and the needs of consumers and advertisers and have been pissing away centuries of general goodwill to advertising while doing that. (I think YouTube is a key contributor to the problem, so extremely relevant in this context of discussing a hypothetical YouTube competitor.) Consumers have started to route around the damage that is modern adtech by looking for more "no ads" services. In some ways this trying to avoid the disease by treating only one symptom.
I think "Sponsored Content" and "Sponsor Segments" are another case in point with reference to that: Advertisers have also started to route around the damage that is modern adtech. Sponsored Content is one of the oldest forms of advertising. Radio started out with mostly just sponsored content for advertising and it was years before they invented something resembling the modern ad break. I don't think it is a coincidence we've seen such a massive return of sponsored content in recent years, and I personally think I can point a finger at modern adtech as a root cause.
> They just get the paycheck to display what advertisers pay for.
Newspapers, radio stations, and TV stations all knew that advertising was content and subject to other content rules including moderation. They refused paychecks all the time. For a brief golden era "Truth in Advertising" was considered the law in the United States and it was a hard-fought law, but it required everyone accepting ads to be vigilant and refuse paychecks for scams and from bad actors, if for no other reason (like ethics or integrity) than that they could be sued for simply running bad ads.
Modern adtech is acting like advertising content moderation is a technically unsolvable problem and it's just too bad that they have to cash all the checks and ignore what it means to their brand image, their integrity, anything but their bottom line. (Either "Truth in Advertising" is dead or it has been having much too hard and slow of a fight against major adtech companies and their enablers. YouTube being both.)
> No service is eternal.
No duh. That wasn't the assignment, was it? If you start a hypothetical question assuming your major competitor no longer exists and has created a vacuum then it doesn't matter a lick of beans what tech stack you are thinking about. You can use 60 year old tech for all it matters if you've got First Mover Advantage in a Vacuum with known demand.
"Be careful of the current biggest competitor that you yourself mentioned in your hypothetical question as a known threat" shouldn't be a surprise answer here, by any stretch of the imagination.
> Will the next generation embrace P2P? Possible. New generations are tech savvy. They are aware of their digital presence. They are aware of censorship.
We're dangerously verging into anecdata categories that aren't useful to the hypothetical question assigned here, but this reads to me like you possibly have some interesting bubbles. Anecdotally, I've never met anyone in real life who cares about P2P other than as the means to some end and the only people I've met who "are aware of censorship", or have strong opinions on censorship at all, I've cut all ties with. Not because they "were aware of censorship", but simply because their perceived "threat awareness" of censorship was being fed to them along with the brain worms of terrible conspiracy theories and worse. Not that I'm implying you are at risk of similar brain worms, that just as likely correlation as it is causation and I try to make no assumptions about that without more than just anecdata.
YouTube's key offering for me isn't providing me a place to host videos or watch videos I already know I want to see.
YouTube offers the discovery of new videos I didn't know I wanted to see. My watch history is likely 80/20 videos suggested vs videos I used search to find. YT's algorithm provides me a list of new content from creators I like, related content to subjects I like, and new content that is relevant to my interests.
The primary focus of an alternative that I'd be willing to use is one that focuses on showing me relevant videos over focusing on mere hosting.
Even granting all those things (which I think are pretty unrealistic), there are a couple issues:
1. An uncensored onion-based video sharing network would very quickly become filled with the kind of content that gets the FBI knocking on your door
2. Much smaller issue, but I actually really like Youtube's recommendation algorithm. It's how I find a lot of the content I watch. A platform without that wouldn't be as enjoyable to me.
>Just imagine it works.
I can easily imagine that, there is already similar things out there.
>Would you actually use it? As a consumer, not a creator.
Yes as a consumer absolutely. As a creator too, why not asking as a creator? You mentioned in the first sentence :
>where you host a stream or upload a video
I would happily do both consume the content and upload my own content.
>Why or why not?
Well in my case, "why", simply because :
1. Youtube eventually runs out of daily content.
2. P2P in my experience offers a lot more 'niche' content. The big outlets for any media (video music or text) usually offer mostly popular and america-centric content. When you are looking for uncommon content that isn't in english you have to look for underground channels to get it.
3. geolocalized P2P would offer local creator more visibility in within their communities. It could greatly help for local culture.
4. Youtube or any other centralized sharing services (meta, twitter, ...) simply have too much power nowadays. P2P allows to share content freely and makes censorship very difficult.
I would use any decentralized service that offered identical functionality, UX, and performance of a centralized service. It's just that there's usually significant overhead to maintaining the decentralized system and, unless the centralized system is acting in bad faith which directly affects me, it's tough to get me to want to put in the effort to switch.
So, yes. On paper, if the content was readily available, I would use it. In practice, everything is too entrenched and I'm not about to start searching multiple websites for video when I'm already successful with just one.
Yeah, sure, why not? If it has the same amount of content, or more, as YouTube, and I could watch whatever I wanted, just like on YT. Especially if it's decentralized, and could get around a few copyright laws - I'd love to be able to watch full episodes of shows like you could back when YT was in its infancy.
The no ads is also huge, I understand that YT needs to make money, but the fact that the ads are so abundant and sometimes the content barely loads while the ad loads in HD quality immediately is maddening and seems like YT is just a platform to push ads.
So yeah, I'd use that site in a heartbeat.
Nope.
Here is the thing that people don't understand about the big tech - people actually like and use their stuff. Nobody really cares that much about ads - maybe we have become desensitized to it, or it was never too big of an actual problem. And as for tracking, people actively want it even if they don't know it - take aways peoples front page of youtube without specific recommended videos and they are going to be bored as hell on the website.
To compete with that, you are never going to build an alternative that does things effectively worse.
Youtube ultimately took off at the start because of copyright infringement. To me, it is why they have an impenetrable moat now.
The scale of copyright infringement needed to compete with youtube to spark a new fire is a non-starter because it is so intractable now compared to 15 years ago.
A site to watch videos to intentionally make people angry sounds like some brilliant marketing. Could open up the algorithm and have settings for more anger, more pointlessness, more stupidity.
>Youtube ultimately took off at the start because of copyright infringement. To me, it is why they have an impenetrable moat now.
This is absolutely correct.
I read somewhere this is what tech companies like Uber used as a strategy. Implement yourself regardless of the laws, and when the legal system catches up act in damage control. It's Blitzkrieg in a way.
I recall several places in the world were taxi services were regulated, when Uber showed up with better services and low prices, they won over the consumers in a matter of weeks. It was game over for the old services.
Napster did something similar but failed in the short run. If they had held off like Uber did and kept their ground everything would be very different as of today. But i think it was just too early.
Since you said to ignore all the downsides that would come with a realistic implementation, yes, I'd obviously watch it.
Depends on the content. For now alternative video streaming sites mostly feature Techbros and conspiracy theorists and that's not the bubble I want to be in and enjoy watching and the mass dgaf for neat backends
Say the homepage has a short url to type, the homepage loads fast, and the recommendations or searching is good, yeah I’d use it too.
One reason I stopped using Nebula is because it became hard to separate the signal from the noise.
> One reason I stopped using Nebula is because it became hard to separate the signal from the noise.
I'm still fairly new to Nebula but I am enjoying it. However I use the RSS feeds to subscribe to only the channels that I am interested in. I agree that their discovery isn't great. But honestly that isn't too important to me. I can discovery new channels other ways.
As a consumer, not a creator.
Any barrier or gate-keeping to creating and uploading is a non-starter.
sure