Hacker News Reader

Comment by armchairhacker | Hacker News Reader

We really need an alternative to YouTube. But

> When you publish with Streambus, you create a website, on your own domain name. Reach your audience over RSS on the podcast apps they already have on their phone.

This is the problem with most other current alternatives: no discoverability. YouTube has no serious competitors because it has significantly more viewers and creators. If I use Streambus, how will I get others to see my content? If I want to find new stuff on Streambus, where do I see others' streams?

IMO any serious contender for YouTube would need:

- People using it - Decent recommendation algorithm - for all the hate YouTube gets, I've seen worse - Decent moderation. YouTube is too severe, yes, but you really don't want hate speech and shady stuff on your site, and especially recommended. It will scare everyone away. And copyrighted stuff will get you into legal trouble

View on HN

(maker of this thing)

> If I use Streambus, how will I get others to see my content? If I want to find new stuff on Streambus, where do I see others' streams?

There's two answers to this.

The first is that the discoverability works through all the same channels as podcasts. You can put your feed on podcast listings, rank in the search and discoverability of the many different apps.

The second is that subscriptions don't need as large of an audience. If you're doing an ad model, you get a million people to watch your video and make maybe $5k in ads. But you could instead get 500 people to pay you $10/month and you make the same with a far smaller audience.

New platforms are a bit of a chicken-and-egg problem. Podcasts is our plan to solve it. If you publish a video podcast, you get to piggyback on the already large audience of podcast-listeners. Then subscriptions lower the audience size you actually need for the platform to be viable and drive more video-creators to RSS, which drives more viewers in a virtuous cycle.

That's the plan anyway. Hopefully it will work! Youtube ads and their enormous % cuts really suck.

If you find this as interesting as we do, or you'd like help launching your own show, send me an email at evan @ streambus . com.

Or if you're in the bay area and vaccinated, let's get a coffee or go for a bike ride!

It would be nice if your service automatically handled submitting the RSS feeds to duck duck go.

Their video search has started including a wider range of sources; I imagine they’d welcome your content (especially since you’re likely to have a better privacy policy than YouTube).

This is cool, we'll look into this

What’s your content policy like? YouTube is famous for aggressively censoring views that don’t align with progressive sentiments, which makes sense since they’re a part of Google, who has the same cultural traits. People have been pressuring Substack to censor/ban content as well, which would probably cause subscribers to abandon them (I know I would). What is your policy?

This is the narrative. But I haven't seen much censorship that wasn't well deserved.

Perhaps there are a lot of left wing hate speech personalities out there that are not getting taken down, but I have seen little empirical evidence that google really is targeting non-progressive views.

Would you expect evidence that isn’t anecdotal to be easily available to the public? It’s not like Google publishes a transparent public log of every act of moderation, so compiling evidence beyond a list of anecdotes is simply not possible. However, I would argue their publicly stated policies on disallowed content is enough to see the progressive bias.

To me and other moderates, and I imagine conservatives, the bias is clear. In this last year I’ve seen Google ban content relating to BLM riots, COVID (like the lab leak hypothesis), trans issues, firearms, and more. If you take an issue like whether transwomen should be allowed participation in women’s sports, enough content is taken down that what’s left as a result of Google’s censorship is a significant amplification of the views of dedicated trans activists relative to everyone else. Google doing this at their scale amounts to propaganda - after all, they are more influential than most governments given the large number of users they have.

Your definition of “well deserved” may fit your own worldview. But for me, I see a brazen suppression of free inquiry, and in many cases, the truth. These actions are framed as being positive, by defining censored content as “hateful” or “violent” or whatever, but they’re actually just damaging and dishonest, since they inhibit open discourse. It’s an important enough problem to me that I can’t support any new service that simply recreates the same oppressive mess - I just don’t see the point because whatever other features they might have, it will still feel mostly the same if this problem isn’t fixed.

I'm very doubtful as my own experience would lean strongly against it having worked at multiple similar companies (facebook, tiktok, snapchat) and having even worked on recommendation system area. Internally the typical employee tends to be liberal/libertarian and for facebook I remember conservative employees being so rare there was one person doing meetups with anyone interested to talk to an open conservative in the company. For tiktok I worked on the main team in the US about recommendation engine for videos and content like BLM was very supported and pretty frequently would appear as a hashtag in trending page.

I haven't worked at google specifically, but having heard plenty of people complain about tiktok/facebook recommendations with very incorrect guesses on what the system bias is, I'm doubtful google is biased against that type of content. Snapchat isn't known as much for recommendation so haven't heard much feedback on it's system.

This seems like the core issue.

Having a hard time imagining why you'd use StreamBus instead of Patreon + Youtube/Twitch etc... unless you are looking to evade the possibility of your content being censored/moderated/demonetized by the platform.

I’ve been thinking along the same lines, but regarding the games industry:

https://mobile.twitter.com/larsiusprime/status/1412881397375...

I think the challenges in my industry are greater, but I think what you’re doing is part of a growing trend I call the “digital crowbar” model. Supercast is an example of one in your space that has already had some major success stories, though their latest poster child (Breaking Points) piggybacks still off YouTube. And of course Substack and Ghost are doing well with newsletters and blogs.

Curious what you think of my framing and how much it applies to your space or not.

Wow Lars you're on Hacker News, that's awesome.

As an Indie who started way way back in the flash portal era, I think the current situation with Steam and consoles is a vast improvement over what I experienced back then. Sure I had direct contact with players in theory, and no barriers to publishing my game. Except players had zero interest in me and my poorly made games.

Every indie looks at their conversion rate and goes "I would do so much better if my game got more attention" missing on the reality that attention is the gold of the modern age. Given enough attention anything can succeed, hey you can even sell attention to others by putting up ads.

Thus any platform which is both open, and gives attention (discoverability) is creating an unsolveable market issue. The attention is valuable, so giving it to anyone will just push the market into a market for cheapest converter of attention into dollars: porn. We see this already with Steam where the top lists always contain some sort of porn-lite games.

Thus my own take is: I don't want an open market. I want a market which heavily selects for quality. Then I want to develop and polish my game to meet that quality bar: since in such a situation I know my little train game will not need to compete with fake japanese hentai.

> Thus my own take is: I don't want an open market. I want a market which heavily selects for quality. Then I want to develop and polish my game to meet that quality bar: since in such a situation I know my little train game will not need to compete with fake japanese hentai.

This actually makes the attention thing worse. Now you’re competing for the attention of a few people who are sitting under a deluge of games. Established players with an existing relationship get entrenched and you better find a way to build a relationship outside of the official channels if you have no name.

Plus some market wag will be along to tell you that the open market already heavily selects for quality. Which is fairly true and the issue for an individual developer isn’t really the result of the aggregate market.

The actual issue is that there is a very strong power law in terms of actual sales. Success begets larger success as the algorithms take notice, word of mouth hits scale and the developer can afford to reinvest with worthwhile advertising.

Secondary to that is that there is just too much information for players to make a fair assessment of all the games they could buy which further amplifies this issue with the market.

Basically discoverability sucks still

For the developers who would be automatically on a closed platform, yes less games would fix it but they’re probably already riding high in the curve. For developers who wouldn’t you want to be very sure you’ll definitely be one of the anointed. Be careful what you wish for.

I’d like to see instead the Steam open up a bit, curators exist but it’s hard to make that a job as it doesn’t pay. Tackling that and turning Steam into a collection of stores rather than a store might help flatten the curve by allowing niches room to breathe and paying curator businesses to mine for gems. The games press used to be this way but now outside of tiny outlets it’s all pretty homogenised and they all talk about exactly the same games.

oh hey, I've played the original defenders quest, thank you for making that.

I think this framing exactly applies to our space, down to the term "anti-platform".

Also I'd be deeply interested in how "Instant games" effects game-land, especially with WebGPU + faster internet on the horizon.

Aw thanks :) I promise to ship DQ2 before I die.

Happy to chat about instant games anytime, email me at Lars.doucet@gmail.com and in return you can tell me more about what you’re doing in video.

Specific questions for here on HN comments -

1) how do you feel you contrast with say, supercast?

2) Chicken or the egg problems are tough. What’s your plan for delivering a truckload of one or the other to break the deadlock? (In this case chickens = content creators, and eggs = viewers/subscribers)

> Happy to chat about instant games anytime, email me at Lars.doucet@gmail.com and in return you can tell me more about what you’re doing in video.

I'll absolutely take you up on this! :D Email coming soon!

> how do you feel you contrast with say, supercast?

I think we are using the trappings of supercast in order to advance the goal of an "open video platform". But the core goal is "open video platform", not necessarily to follow that exact playbook.

> Chicken or the egg problems are tough. What’s your plan for delivering a truckload of one or the other to break the deadlock? (In this case chickens = content creators, and eggs = viewers/subscribers)

Two things: - Shrink the eggs required to make an omelette. (forgive me for stretching this metaphor) If you make paid, subscription feeds; you need fewer individual people, and you can use existing platforms (Youtube / Twitch / Tiktok) as a marketing channel for your paid videos on the open platform where you control the business. - Piggyback on the existing channels that podcasts have in place (search rankings, recommendation rings).

If the balance is right, it'll work and we'll have an open video platform! And if not, then at least it's been fun to work on. Or maybe to put it another way, if it doesn't work, it will still have been the right thing to have done.

It sounds like both of you guys are wildly off the mark.

Steam, the App Store, the Android stores, now Roblox... they have paid out like hundreds of billions to game developers over the years, more money per year directly to IP creators than any other category ever in all of history. So anything that starts out by characterizing them as the antagonist will fail to attract successful game developers, regardless of how much money you spend acquiring an audience or paying up front for content. See Exhibit A, Epic Games Store.

What is the problem with Roblox? The games suck. So many utterly garbage shitty clones. The shitty stupid mindless microtransactions in their shitty clones. There are no good Roblox titles. I don't care how many 8 year olds play the games, I don't care how much money in spent, they're all complete garbage. If you can't say that, that you and no one you know with a mature brain plays these shitty games, you are going to struggle to communicate with game developers.

What is the greatest threat to Roblox? Apple could start refunding parents on IAP. You're talking about such sophisticated stuff that doesn't matter. A successful Roblox developer tried to convince me the money came from $5/mo Robux subs and look, that's delusional. Even if you ask the developers they are delusional. The money is obviously from whales, because it's always from whales, and like, what does a 8 year old whale look like? An utter menace. Spending money mom and dad would love to get back.

What is the closest platform to achieving the goals you describe? Apple Arcade. It doesn't extract the full value from whales, so it will never be the #1 source of revenue for game developers, but the games are good!

Who is the real antagonist? Ketchapp (https://venturebeat.com/2014/03/30/threes-vs-2048-when-rip-o...). Riot Games. Epic Games. Most of Tencent really. Even Blizzard's last title which I love, Overwatch, is more Team Fortress than it is not.

There is one thing that rings true. It has been characterized as "it is way easier and more personally lucrative to be a money manager than to be someone the money manager invests in." Epic and Roblox are really tales about the power of capital, that they reinvented and cloned and pivoted their way into completely accidental success, being led by chickens with their heads cut off with fundamentally cloned and shitty IP, because they had huge amounts of money. The greatest ingredient of their success is owning the rights to the dice when they roll six, and rolling many, many dice.

I think you've completely misunderstood my position, or else I've failed to communicate some of my background assumptions. I thought I referenced the "chicken or the egg" problem and fruitless red ocean fights throughout the twitter thread, but perhaps I was too subtle. The entire point is to establish a new category rather than go up against them directly.

I'm very aware of this and it's not something I'm ignoring. It's the central thesis of "So You Want to Compete With Steam" [1] which is background for everything I think about. How to solve the Chicken or the Egg Problem is absolutely fundamental to these issues. Just because I don't have an immediate solution for it in a new space (other than spend truckloads of money) doesn't mean it's not something I'm thinking of constantly.

My answer to "here's how to compete with Roblox" and "here's how to compete with Steam" is literally the word "Don't" in big bold letters right at the beginning of both articles.

Also, I think you give Roblox short shrift. 99% of everything is garbage, that's just Sturgeon's law.

[1] https://www.fortressofdoors.com/so-you-want-to-compete-with-...

Have you heard of twitch? Cause it sound a lot like twitch, and they already exist and have huge userbase.

This seems much more difficult to use than twitch and much fewer features too.

Have you interviewed both successful and unsuccessful YouTubers and asked them about their growth story? Questions like: What tactics worked and didn’t work for their own subscriber growth?

You could even ask them how they think they would recreate their growth story through your proposed platform.

The main problem with the current alternatives is that you can't even upload a video. For example, that is the problem with this website Streambus. You can't upload a video. It's some kind of "beta signup" manual approval thing.

On Youtube, you make an account, you click the "create" button, you select the video file you want, and it uploads. It works.

Another example: have you ever tried to upload a video to peertube? Typically you make an account and then you click "create" and it says "uploads are not allowed on this instance" or something like that. As if the person making videos has any idea what that means. This is trying to use a printer and it keeps saying "PC LOAD LETTER", of course some nerd knows what that means but 99.9% of people are never going to care.

If you want an audience, allow people to upload videos to the website. This website doesn't solve that problem. Peertube doesn't solve that problem except if you somehow manage to find the https://peervideo.club instance, which is apparently the biggest english-speaking instance, but if you go to the "about this instance page", apparently "video quota" is 20GB, and "User registration allowed" has a red X next to it. What the hell does that even mean? Why is it so hard for people to make a website that makes sense to the average person? It is astonishing to me the lack of urgency there is for creating an alternative to youtube. Isn't Peertube maintained by like 4 people? Isn't this a massive global problem? Why is no one taking this seriously? Do people not understand how powerful it is to be able to see videos of things? For example, footage of war crimes or atrocities or things happening around the world? All the proposed solutions look like cutesy little hobby projects. For example, I'm sure they are great people, but honestly I don't care about "Evan" and "Kevin" or their little cartoon portraits, which is the first thing you show me when I try to create an account on Steambus. It doesn't strike me as headed in the direction of being a genuine solution to this global problem. What's going on with youtube is a nightmare and to put it bluntly, lots of people across the world are suffering and dying because of it.

I didn't even realize you can potentially upload videos on Peertube instance which is not hosted by yourself. I'm totally not surprised this feature is not allowed on most instances.

Peertube's UX could be better, but it doesn't present itself as a Youtube competitor, does it? It's the cult of decentralization people that bring it up everytime somebody gets kicked off Youtube.

But it's not Youtube and never will be. Pretending that it can be defeats the purpose. As does an overly simplified UX that hides the system capacity underpinning the technology (20GB limit). It will feel like a bait and switch when a newly converted user runs up against the limits of decentralized solutions, because they were not educated about that before.

>Peertube's UX could be better, but it doesn't present itself as a Youtube competitor, does it?

That's a pretty stupid question, isn't it? If we replace your word "competitor" with the word I used, "alternative", then yes, it does.

>But it's not Youtube and never will be. Pretending that it can be defeats the purpose.

What purpose?

>As does an overly simplified UX that hides the system capacity underpinning the technology (20GB limit). It will feel like a bait and switch when a newly converted user runs up against the limits of decentralized solutions, because they were not educated about that before.

99.9% of people will never care about "the system capacity underpinning the technology".

> how will I get others to see my content?

This service gives you a website, so I'd imagine you'd get an audience the same way you normally should on a typical blog:

- Make good content

- Judiciously share that content and collect feedback

- Let members of the circles you shared with continue the cycle as they re-share, adapt, cite, and respond to your content.

- Notice other people hosting websites of their own that mention you; advise them to follow the same steps.

- Sit back as the recursion triggers logistical growth.

This strategy has worked out quite well for multiple online identities of mine. Truly good content doesn't need a silo or advertising to be recognized.

Could of observations:

Many people are making good content.

If you skip all of what you said and just upload to YouTube chances are you expose your content to X-hundred million people and you can then promote it just like you said.

You are correct. My point was that you can still build an audience without being locked into a silo or megacorp the exact same way you would for a normal blog. My point was not that doing so is easier.

If your content is good, it should succeed without a megacorp's help.

Incorrect. Uploading a video from a brand-new account on Youtube doesn't expose it to anyone. When someone searches "[your video topic]", what are the chances that yours come up, assuming it's a common enough topic?

The algorithm doesn't surface videos unless you've created a fairly substantial library that tells YT that you intend to post up for the long term.

This is a good description of Content Strategy. Nothing becomes popular on its own. You have to do the work of getting it in front of people's eyeballs, and improving the work after receiving feedback.

How bout Odysee?

- People using it (not that many, but there's a sizable community) - Decent recommendation algorithm (okay ) - Decent moderation (part of the whole appeal is "free speech", but it's pretty toned down compared to some others)

It's pretty okay on all fronts, but not the best. Still something to check out though!

Unfortunately it uses cryptocurrency/blockchain nonsense. A better alternative is PeerTube, which is part of the Fediverse; people can share videos, comment, and post across Mastodon, PeerTube, Pixelfed, and others seamlessly.

Plus, you can pick (or host!) an instance that agrees with your moderation preferences, and block instances that don't (e.g. alt-right and troll havens). The distributed/federated approach to moderation and hosting scales well with a limited budget.

It would be nice if Peertube wasn't developed using Nodejs.

Why does it matter (unless you think about contributing to the codebase). I'm a vocal opponent of unnecessary Javascript (TypeScript is fine) but I never care what other people use for their backend.

It's not nonsense, it works. But yeah, PeerTube is a good alternative too!

However, discoverability and ease of use are always factors. With Odysee, you just use it like YouTube, not much is different.

I haven't figured out how I can upload videos to Odysee without buying LBRY (which is also not that trivial).

I guess the easiest way is to ask someone to donate me some, but I don't know anyone on LBRY well enough for this.

I'd like to point out that one of the main reasons people publish content to YouTube in 2021 is because that content will be monetized (or, they want to be big enough to monetize their content).

I see zero competitors that offer anywhere near the revenue that YouTube offers. Until you have another platform that can offer anywhere near this level of revenue for content producers, there will only be YouTube. (Twitch is a different market and doesn't compete with non-streamed YouTube content, which I think we'd both agree on)

For that matter, perhaps it's a good opportunity to regulate YouTube's monetization policies, so that other companies can compete for content producers more evenly.

Streambus already offers a higher revenue model than Youtube.

A Youtuber with a million subscribers that can convert 1% of their audience to paid subscribers at $5 a month makes $540k a year on Streambus.

And they'd do it on their own domain name, with the full ability to leave at any time without losing their subscriptions or audience.

They don't need to leave Youtube, or turn off their existing monetization models either.

Meanwhile, their audience can use an app they already have on their phone, and don't have to watch ads from a middleman.

That's an insane price ($5 a month). HBO subscription costs $14.99 a month. Expecting indi artists to get that level of revenue is insane.

I currently pay this price for several of the Patreons I support, I'm not sure what makes this any different. I'd certainly rather throw money at a small handful of good content creators; quality over quantity, and my time is limited anyway.

And yet people do it already just for a bunch of emails on substack. It’s worth it for somebody, that’s all that matters.

Not to pile on, but people give small monthly sums to all kinds of individuals and organizations for any number of reasons. It’s a huge factor for small and large publishers and creators alike. In many cases it’s outpacing ad revenue entirely. It’s great because it allows more freedom and incentivize quality.

You only need a few; I have a Patreon for my comics and so far I've gathered about eighty people willing to pay $1-5 per page, times 6-8 pages most months. Which covers most of my rent and expenses; doing private commissions fills in the gaps. They mostly have IT jobs and like the stuff I'm putting out for free enough to support it, fifty bucks a month is a drop in the bucket compared to what they're paying for rent in SF or Seattle or whatever.

And I am absolutely nobody, I have all of like 1700 followers on Twitter, and am pretty much absent from other corporate social platforms.

i like your art style, digging the cools vs warms for character distinction

Thanks <3

Except for the fact that YouTube already has a membership model for creators on top of their ads model, so you can be a paid subscriber (which includes other benefits) on YouTube too to support your favorite creators.

How would this compare to Patreon? Which seems to be used already by many Youtubers.

From what I can tell, Patreon allows you to export customer contact info, but you can't take the financial relationship with you. They're not subscribing to you, they're subscribing to Patreon. If you want to leave for another service you have to awkwardly email everyone and ask them unsub and then resub somewhere else, which will probably not work.

Streambus seems more similar to substack/supercast -- you as the content creator own the stripe account, and streambus only makes money so long as you feel the service is providing a good value to you, otherwise you can leave any time.

The discoverability can come through community at some level. I have a large blog roll on my self-hosted TTRSS reader that I have curated over the years. I find new blogs when people mention them on a blog I follow or a friend sends me an article.

Agree with everything you said, especially on the challenge of recommendations and feel this is where those designing the recommendation engines have at least the same moral responsibility as a bio-engineer doing gain of function research on viruses.

IMHO the core problem is the recommendation engines are optimizing the wrong metrics and this leads to destabilizing feedback loops.

The problem is likely AI complete, but it should be possible to tune recommendations so they don't create self strengthening conspiracy theories. I suspect a reinforcement learning environment that mimics users interacting with a recommender system would be the best way of tuning the larger system behavior so the longer term properties aren't catastrophic for society.

OTOH if the platform was setup so "good faith" arguments were somehow incentivized, there might not be any need to optimize for any other metric than engagement.

Which that last part is hilarious because it's exactly what helped propel YouTube to what it is today. IIRC the Janet Jackson Superbowl fiasco and some specific copyrighted song were what helped them out the most.

What we need might be the decentralized search engine.

Probably just publish a few videos to Youtube and ask future subscribers to follow your non-Youtube channel instead.

I imagine that's against T&Cs? And if it becomes popular then YouTube are going after anyone successful for lost revenue?

I don't think it is, there are literally tons of YouTubers that advertise alternative platforms such as Nebula/CuriosityStream.

Why can’t we just add video to existing podcast aggregators?

The only stuff I watch on YouTube come out episodically and don’t have shitty clickbait titles. Some of them already have sponsor ads similar to what podcasts do.

I don't know, Substack and Locals and the other content creators that are starting their own subscription based websites are doing very well, although they build their audience with YouTube.

Surely some enterprising folks can create video RSS crawlers and put together a video podcast app, right?

Linking from social media is the discoverability mode here. And the decentralized nature will make it impossible for them to censor it unless they completely block 3rd party sites like AOL back in the day.

> YouTube is too severe, yes, but you really don't want hate speech and shady stuff on your site, and especially recommended.

It’s not clear to me why you’re advocating for censorship. Why do you care if someone else shares content that you’re offended by? Just don’t click or watch it. Vague policies like “hate speech” are always just ways to suppress political opinions, and end up reflecting the biases of the provider or loud vocal activists. There’s almost no point to seeking out an alternative platform if it cannot differentiate itself by upholding basic free speech principles. Moderation should just stop at explicitly illegal content.

Because when you want to go watch cat videos, having a "HOW THE JEWS ARE CONTROLLING THE WORLD WITH 5G" video next to it is:

* buzkilling and revolting

* makes you associate the platform with nonsense

* scares people of association with the platform.

Reply on news.ycombinator.com