Comment by shafyy | Hacker News Reader

Comment by shafyy | Hacker News Reader

Parent Full thread

Idea-wise, this seems similar to Tim Berners-Lee's Solid (https://solidproject.org/).

The problem with these approaches is that no significant company is going to do this voluntarily. And this is not going to make a dent until Meta, Apple and friends do it. They are not going to do it because it doesn't give them an advantage, it only makes their lifes harder. It makes it harder even if they don't use the data for anything, because of increased technical complexity etc.

So, this is not a technology problem, but a legal/political problem. Just as the EU is forcing big messaging apps to ensure interoperability now, the only thing that will make a difference here is a law that disallows companies from saving customer data in their own databases. A GDPR 2.0. If you want to make a difference in this space, you need to spend your time pushing new legislation.

I'm sure that in time the EU will get there. And then we will need good tech like Schluss or Solid. But before then, it's not going to do anything.

Does Solid have a concrete working example? I've tried getting into it, but it seems very academic and with the nitty-gritty details conveniently left out, it gets frustrating trying to work out what exactly Solid _is_.

I don't think Tim Berners-Lee is trying to sell us some vaporware, so has anyone got any pointer?

https://solidproject.org/apps

Solid is a specification that allows for decentralized data storage in so-called pods. The idea is that consumer applications don't store a copy of your information in their own silo - thereby having many copies of your data floating around with 3rd parties - but instead query your pod. That's the ownership / self-control aspect of the idea.

The crux is that you can self-host your own solid server with your own pods, or choose to rely on a hosting provider. That's the decentralized aspect of the idea.

The big question is how this plugs into an economic model with real world incentives, interests, distribution of power / leverage, trust / credit, and so on.

It would work with legislation if the loss of personal data of third parties induces extremely high fines that it simply isn't worth it anymore. Fines that could almost ruin even large players. Exception has to be business data since there is simply no alternative but the distinction isn't as trivial as it seems at first.

Since many would not be able to store user information they are reliant on people hosting themselves or at a service. I know it is difficult to explain that having control about your data is important and probably the most significant barrier. There are non-technical people that value privacy and data protection but most of them are highly educated.

It would be much easier just to mandate that storing customer info is not allowed as a company.

Would many companies pay for user data if they weren't allowed to store it?

Tim Berners-Lee's startup Inrupt [0] is essentially trying to partner with companies to sell an enterprise version of the Solid Project software. Their website suggests they've partnered with some large-ish companies already[1]. I could see a webpage prompting you for a release of your certain parts of your 'pod' rather than use cookies perhaps.

[0] https://inrupt.com/ [1] https://inrupt.com/solid-enterprise-natwest-bbc

I agree that this won’t change anything significant until the law makes it mandatory, however a huge step towards getting the lawmakers to do so is tho have the tech already out there and working well with hundreds of smaller players. That will save us years of lobbyists claiming it’s unworkable, to expensive, technically impossible, etc.

> That will save us years of lobbyists claiming it’s unworkable, to expensive, technically impossible, etc.

Lobbyists will do this regardless if the tech exists or not. They currently doing it for messaging app interoperability, even though open protocols and APIs have been around since the internet started.

Don't get me wrong, I'm not saying let's not work on this tech. Just saying that this is not some hard tech problem that needs to be solved. If a law would be passed today that forbid companies from saving customer data, a dozen startups would launch tomorrow with this tech.

> however a huge step towards getting the lawmakers to do so is tho have the tech already out there and working well with hundreds of smaller players.

Not only this, but it will also prevent a scenario where the political will is given, but the tech does not exist yet and then the lowest bidder will implement it, instead of a genuinely interested group of players.

Reply on news.ycombinator.com