Individuals or small organisations simply can't afford to take on the potential liability that comes with letting arbitrary people upload arbitrary stuff to their servers.

If some hosts carry the controversial stuff and others don't, then those who do will be targeted, censored, persecuted and/or sued out of existence, and they don't have money to defend themselves because all the money making stuff is hosted by others.

On youtube? Your video is just taken down because it was flagged, or someone made three bogus copyright claims on your channel, or some algorithm thought your video of art is nudity. In these cases, you have zero legal recourse at all, even if you wanted to pursue it. Youtube's solution is not to protect creators, but to simplify their own legal logistics.

But you're not wrong about liability, hosts will need some solution. One part will be that PeerTube will probably build in features to help hosts moderate and manage legal documents like DMCA. Another common solution to dealing with liability and risk is insurance.

But for individuals acting in a personal capacity that doesn't matter. For them youtube is the no recourse, no financial risk option. People upload random stuff, not everything completely legal but often not interesting for copyright holders either (e.g very old episodes of TV shows). Sometimes it gets taken down. No harm done.

For professionals, relying on youtube is risky though. They could lose their entire audience over night if they get suspended.

So it sounds like decentralisation could work for professionals, but of course the catch is that without the huge volume of random stuff uploaded by non-professionals the platform doesn't have mass appeal because it's not the go-to place for video search, and that means professinals won't find the large audience they are looking for.

Insurance doesn't work in this case. It only works if the likelihood of getting sued is low and you cannot be accused of having broken the law knowingly, neither of which is compatible with working in a grey area.

Let me try to make a case for how insurance could fit in here. Insurance like this is often predicated on the insured meeting various conditions. I could see a forward-thinking insurance company offering terms like these:

1. Require the producer to pre-declare exact claims of all source material for both music and video. For example, "1:02-3:55; Public domain music. Sourced from ...", "0:00-0:10; Licensed stock intro. License details ...", "5:07-5:15; Fair use news clip. Source at ...", "0:15-15:35; Original footage. Taken with X camera at Y location on Z date.".

2. Perform random audits on videos including tracking down all claimed sources to make sure they're documenting their sources correctly.

3. Offer running video through a ContentID system that can automatically check claimed sources with an internal database of known content. By opting-in, you can reduce your rate.

I think these bring the risk down to similar levels in other industries that insurance companies currently serve. Figuring out the exact rates, plans, kinds and amounts and limits of coverage, rates for different content etc etc etc is all the job of an adjuster. For example, fair use claims might have a higher rate of being contested, so maybe using them would be restricted and/or more expensive. It's in the insurance company's best interest to reduce the number of lawsuits and claims, so they will pursue anti-SLAPP judgements (and lobby for them!) to recoup legal costs of frivolous suits, thus reducing frivolous claims. The producer will have a well-known coverage and be able to manage their risk.

If the pre-declared claims sounds onerous to you, I could see video production tools being able to read specially formatted claim metadata embedded in any source and automatically outputting a properly formatted list of claims with timestamps embedded as metadata in the output. As long as the producer uses properly tagged sources it can be automatic. In fact, this could even be backed up with cryptographic signatures to allow in-place verification that you've purchased a particular license.

Another benefit: My apartment requires me to maintain renters insurance. Similarly, you could provide your cloud provider with proof that you maintain this kind of insurance, in exchange for them promising they won't disconnect your account due to copyright claims (and forward them to your insurance company instead). This combats the "walking up the stack" problem discussed in other threads.

Non-professionals are not just uploading home videos (which often contain copyrighted music by the way), they are also uploading stuff they have downloaded elsewhere (music, TV shows, etc). That's the key issue.

Your insurance idea sounds a lot like what Google is already doing and it brings back a lot of the downsides of centralisation. It's very onerous/expensive to run and therefore requires significant monetisation to cover the cost. It also carries the risk of false positives as everything would have to be automated.

The insurance company would disproportionately "punish" hosts that carry the more controversial/risky content unless the platform as a whole is the insured party or all hosts are forced to pay everyone elses insurance premiums.

This would create a disincentive for hosts carrying uncontroversial content to join the platform in the first place (not just for financial reasons but also political ones). They would create their own far more restrictive platform with much cheaper insurance premiums or simply upload to a centrialised provider like youtube.

The incentive to subsidise controversial/risky content is to be the go-to place for video search and monitise the heck out of it. It's simply a centralised business model.

Free speech is a completely different topic. If you're conflating copyright infringement and free speech when you're thinking about insurance, I can agree there's no way it could work. Let me clarify: coverage would be limited to copyright infringement claims only. Covering free speech with insurance is completely a non-starter, they would be opening themselves up to a new universe of risk, even suits directed directly at themselves. That said, the insurance company would be highly incentivized to discourage the abuse of copyright infringement laws to quell speech: these are separate concerns (from their perspective), so reducing such abuse would be good for them (and us!).

Think about what happens to controversial speech on centralized platforms now: some faceless moderator unilaterally decides -- without option to appeal -- whether to take your speech down from their platform permanently (it's their platform after all) and possibly ban you. In a decentralized platform you have the option to defend it with whatever resources you decide.

> Non-professionals are not just uploading home videos (which often contain copyrighted music by the way), they are also uploading stuff they have downloaded elsewhere (music, TV shows, etc). That's the key issue.

With a decentralized model like PeerTube you have to change your conception of who accepts and hosts uploads. You can't just upload to "The PeerTube" like you can YouTube, you have to upload to a federated instance. PeerTube makes hosting your own instance much easier and cheaper. Many will create their own, some might use one hosted by a family member, others might sign up for a more public instance. In each case, the instance will be more and more like the centralized model. But that's ok, because now there's a gradient. You have the choice to manage your own instance (or any available middle ground), with all the benefits and downsides that come with publishing your own content (except hosting costs of course, that's the point).

My point is that with this model, you have the choice to moderate it and defend it manually yourself (professionals, activists), or defer it entirely to a third party (basically the current centralized model). But there's a large gap between these, where some kind of insurance-like service could help bridge.

Two big blockers that I rarely see discussed for the decentralized future coming to pass:

- (D)DNS for the 1+ persistent devices per/user (IPV6 support/rollout basically)

- Lowering the barrier to entry for running a server on a users' behalf

If we really want the internet to be a web, we need to make it easier to run servers that act on behalf of users/hold their data, that they control.

If you'll indulge me, I've had in the back of my mind a project for a long time, a <$30 home-server (I'd call it "the tube" or "pipe") that has a cheaper but good-enough set of functionality that you get from the internet giants today:

- Document management and storage

- Backups (connect your local "pipe" to a backup storage provider)

- Chat/Messaging, SIP + Video conferencing

- Local networked/mesh services spanning from blogging, social networking, ecommerce, whatever.

I'm very much of the I-just-want-to-write-software so I preoccupy myself thinking of what tech I'd use to make it happen, but I think this is the kind of project that people need to really embrace the internet as a sparse group of interconnected nodes.

I also think that humans are ready to accept this kind of structure -- we already have it! People can understand that if you want a storefront to be up all the time, you need to have something run it all the time, the power to run it just isn't in their hands yet.

Of course, startups are hard, hardware startups are harder, and consumer product hardware+software startups are pretty hard, so it's much easier said than done. Decentralized systems absolutely can work, but the amount of work, fore-thought, and careful construction with consideration for generating network effects and making it easier for people to run and use -- basically good product design -- is just missing in the open source world, because it's hard.