Hacker News Reader

Comment by egypturnash | Hacker News Reader

I don't think this monopoly is really mysterious. Storage costs money. Bandwidth costs money. Someone needs to pay for it and the only way to cover those costs at any meaningful fraction of Youtube's scale is to have a money printing machine like Google's ads.

View on HN

In 2005, sure.

Its a bit more mysterious now a days though. Video compression got way better (albeit video quality also went way up), hard drives got way cheaper. Bandwidth is really cheap at scale. People are way better selling ads now then they used to be. A lot of video serving infrastructure got standardized.

Don't get me wrong, its still hard and expensive, but i don't feel that is the moat it once was. Network effects is also a whole other conversation.

Why do you think every single streaming provider and platform does checks every so often to see if you're still watching? It's not because they are being nice; it's because it is costing them money, even the huge companies.

> Bandwidth is really cheap at scale.

This is extremely false, where are you getting this information? Bandwidth is ludicrously expensive, no matter what your scale.

Why do you think Netflix gives ISPs server racks filled with the entire Netflix catalog, or Microsoft/Google/Meta spend billions on their own private submarine cables? Nobody would do that if bandwidth was cheap, but it isn't.

The problem is that video quality increased to meet availability.

In the past 480p would be okay. Now everyone wants 4k.

In fact, in the past IMAGES were normal. Imgur was an image website. Now everything is about short videos. Even memes are now videos.

I'm pretty sure if we make Internet faster and storage cheaper, we'll also invent a new sort of media to waste that speed and storage.

I think the idea is that they operate as a black box and work in mysterious ways, not that it's mysterious how they became a monopoly.

I don't thinks it's quite that simple, there are other factors as well:

There are significant network effects. Content creators use youtube because there are a lot of viewers watching content there, and viewers use it because there is lots of content there. Since YouTube already dominates the market, it is extremely difficult for another platform to compete, even if it was better in every way.

Google can promote YouTube using its other monopolies/oligopolies. Most notably, google search prioritizes videos on YouTube over other videos. Also, being able to pay for video ads and search ads with a single vendor is probably actractive for ad space buyers.

Google also already has its own CDN, which probably reduces the cost of distributing the content.

I think this network effect and the discoverability aspect are the main reasons why it's extremely hard to compete with YouTube. Why would people use another site if the content they want to see is not there, or is too difficult to find? Why would creators put videos there if they can't find a large audience?

The YouTube algorithm is problematic in many ways but it does succeed in viewers being suggested videos they want to see, even if the signal-to-noise ratio is not very good. That's hard to replicate when starting a new service.

Also youtube is big enough that they can get cache servers in isp datacenters for the popular content - it saves the isp the cost of a bigger pipe so deals not offered in general exist. (Netflix also has this with some - or at least they were working on it years ago)

At the rates I use video, my CDN doesn't care I'm distributing video bits, so at my end of the use spectrum, video bandwidth costs no more than the CDN fees I'm already paying. But yes, that won't work for Netflix or Disney+.

If the quantities of money paid to all the YouTube freelance advertisers is anything to go on, a video platform having their own ad network would itself be highly profitable.

That's a big assumption.

It's very possible that it's only that profitable at Youtube-sized scale.

you don't need youtube scale - but you need to be a lot bigger than most others are. You need to be big enough that ads pay for a full time ad salesperson as well as your other overhead.

There's definitely some very large critical mass necessary for the network effects to kick in. Is that mass such that there can only be one? Or could there be three or four, like other social media? Arguably Instagram is as large a video platform, though obviously very different. Same with tiktok; the individual videos might be shorter but the bandwidth costs scale with watch time.

> If the quantities of money paid to all the YouTube freelance advertisers is anything to go on, a video platform having their own ad network would itself be highly profitable.

Then why isnt everyone jumping at the opportunity to make a competitor? If it is soooo easy, we should have competitors. Nobody is stopping you from launching margalabargalatube.com and win the market.

Nobody, including Jeff Geerling, has an exclusive deal with YouTube to distribute the videos. Make it happen!

I mean, if that's how we're going to talk to each other, I can engage on your level...

You are soooo right! There's no such thing as a network effect or a first mover advantage! If something exists and is self sustaining (my original point), therefore creating that thing is trivial and anyone can do it (your invaluble contribution)! Your logic is flawless. Have you considered going into freelance consulting? Someone with such good and original ideas should be charging money for them.

Reply on news.ycombinator.com