I'm worried that while this sounds plausible at first it also seems to run directly counter to how many of the most popular channels on YouTube got their start. Most times when I've been a subscriber to a channel before it took off, the channel always begins with a small, tight knit group of fans that appreciate the content being created. The channel owner may not even run ads and sometimes doesn't have the sense or skill to edit their videos down. The channel will exist in this state for quite some time before it begins to attract more attention. How many promising channels are going to be strangled in the cradle by a poorly calibrated algorithm that thinks they aren't commercially viable?

I'm afraid that if a policy to terminate channels early and often is pursued then we will end up with a YouTube entirely populated by channels that just seek to maximize engagement at the expense of quality. This is sad but YouTube has been on the decline for some time, there is still a lot good about it but they seem to be intent on transforming it into the antithesis of what made it successful in the first place. Same old story.

> we will end up with a YouTube entirely populated by channels that just seek to maximize engagement at the expense of quality

Just like Medium did (and does). This will probably kill the plataform success...

How is this a good thing? Obscure and/or old archive footage which usually doesnt get that many views is one of the best things about YT.

I’d argue that YouTube is a for-profit business. They have every right, within legal bounds, to do whatever makes economic sense to them.

If you want an archive, go with archive.org. But even then, nothing is free. If somebody thinks a footage deserves to be archived and preserved, somebody needs to pay for the cost anyhow, e.g. via donations.

As a user I am not interested in YT's profits. However, what I am interested in is being able to access useful content. For all intents and purposes, YT has become the central place for people to upload such footage and of course the prospect of its removal is worrying to me. There are vast amounts of indispensable knowledge and information uploaded to YT that is far from being deemed as profitable to Google and as such are under risk of being permanently and irreversibly lost. Hence my confusion at the OP's description of the situation as "a good thing".

> As a user I am not interested in YT's profits. However, what I am interested in is being able to access useful content.

Choose two: a) free* b) reliable c) universal

* not counting ads.

Youtube being a monopoly on videos, is a de facto public service.

There are alternatives. We just need to use them if we care a little about it. e.g. your own web server (e.g. blog) / p2p network (e.g. zeronet / dtube) or even 'the modern social network' (e.g. facebook / twitter)

> Obscure and/or old archive footage which usually doesnt get that many views is one of the best things about YT.

and one of the costs youtube bears, since by definition they can't make money off those videos, but has to pay for storage.

I dont know if there's a solution - perhaps non-commercialized accounts can't upload more than X gigabytes of vidoes, or be forced to have ads? Or you (the channel owner) pay for the storage?

> and one of the costs youtube bears, since by definition they can't make money off those videos, but has to pay for storage.

Sure they can make money off them - they just need to be creative. For example, you can have a category of 'archived videos' to which you need to pay to be able to access. The payment is for the service, not for the video so it does not matter what are the specific settings on a specific video.

Even consumer SSD's are around 10 cents per gigabyte of storage. These videos wouldn't get many views, but they don't need a ton to recoup storage costs.

Indeed - going further, the long tail is the only advantage centralized systems hold over decentralized ones. The day YouTube only shows the top 99% of content is the day we can all switch to BitTorrent and not lose a thing.

I doubt these kind of users cost YouTube all that much. What I guess they are worried about are people who upload controversial videos who then run adverts alongside them, and those advertisers get very upset about the content that they are being associated with. That's the kind of thing that can really cost YouTube money & effort (they might have to have a real person take phone calls from the annoyed advertiser, shock horror!)

I think as everywhere 90% of money Youtube gets from 3-10% of content. That looks like cost optimization. They spend money for requests (displaying video) and storage. Even if storage is 30% of their cost and they can remove 10% of biggest never displayed videos they can save 3% of overall costs, which is huge improvement on very concurrent market. Video ads right now the most expensive and most fast growing market.

I doubt storage of extremely unpopular videos is a significant cost for YT. Their revenue is ~25 billion per year. Assuming with redundancy they are paying say ~1,000$ per 8TB added per year. People are uploading 500+ Hours of video every minute. Assuming that’s mostly 30fps 4K video we are talking 50 Mbps * 60 * 500 * 1 year ~= 0.75 Billion per year or ~3.3% of revenues. (50 Mbps seems like an overly generous average.)

Not free, but cutting off say 10% of uploads to save ~0.3% is not going to help their bottom line much as they would lose out on some viral videos and future stars in the process. Plus, there are minimum view requirements before revenue sharing. Worse they could easily lose their spot as the default location for video uploads long term.

I think that kind of policy would have a bad effect on YouTube usage. Not everyone uses it to upload videos that they want to go viral or get lots of hits. Plenty of users are storing videos on there that are just for family and friends. The users who are uploading this other ~90% of 'small time' content aren't going to keep using a site that would routinely delete their content because it wasn't popular enough.

It's also an arbitrary hammer to censor videos regardless of size etc...

Vs other replies, seems it's intended more toward genuinely wasteful content, like "10 hours of stupid on repeat".

It compresses very well

Where should people go for how-to videos on repairing x? Usually I've found what I needed on youtube, and while very helpful for my need, the popularity of such videos (often long for instructional purposes) is in the hundreds or low thousands of views.

I would argue that channels like Jim Sterling are a bigger problem. He's a pundent who criticizes the AAA gaming industry, and whose content is unmonetizable from Google's perspective.

Yet his videos get tens and hundreds of thousands of views (occasionally even millions), and are 15-20 minutes each (even Jim Sterling must play to the algorithm). 3-4 of such videos every week have to represent the very definition of commercially not viable for Google.

Option #4 take a small dent to their profits.

I think YouTube as a whole might already be a not-so-small dent on their profits.