Live data from Hacker News

Personal data storage is an idea whose time has come

blog.muni.town

21–30 of 287 posts

Re: Personal data storage is an idea whose time has come

#21

Earlier quoted context omitted.

The creator/consumer divide is still 90/10. Your example just doesn't matter.

If I don't create anything, and just consume creators, what do I need a personal data store for?

Just your existence itself already create a lot of data ;)

Re: Personal data storage is an idea whose time has come

#22
> Rather than being in countless separate places on the internet in the hands of whomever it had been resold to, your data is in one place, controlled by you.

I don't see how this follows. The moment you create/share data with a site, what's to prevent them from reselling it?

The only thing this seems to attempt to solve is portability/interop (and moving control of and responsibility for blocking/moderation/spam to users rather than sites).

I don't see how it helps at all with privacy or you "controlling" who gets your data. If you give it to site A but not data collector B, what's preventing A from selling it to B? As far as I can tell, the situation will remain identical to how it is today.

Your data will never be in one place unless you never share it. The moment you use it with other sites or services, it is stored there too, out of your control.

Re: Personal data storage is an idea whose time has come

#23
post #11

It was an idea that never went away. Many people have wanted to self host everything. Sadly companies have found it easier to centralise, and then as a bonus can monetise that data.

It wasn’t the companies but the users that found it easier. There’s a reason why everyone’s on Facebook, instagram, and gmail instead of running their own hosts—because it’s vastly easier for the majority of people to do so, and because everyone else is there.

We have not solved decentralisation in an accessible and useful way yet, and the incentives won’t change until we do. If ever.

Re: Personal data storage is an idea whose time has come

#25
For those of us who've been around for some time and still value privacy, this sort of paradigm is obvious.

The trouble isn't a lack of the right technologies - I'd argue it's a problem in the go-to-market strategy of those building these products/technologies.

Ideas flow along lines carved out by power/influence. Facebook's early strategy was to start with restricting its usage to people at Harvard University - arguably a highly influential institution - and then expand outwards to other highly influential institutions. Only once the "who's who" from those institutions were already onboard did they let down the walls to allow us plebs in, and we all rushed in head-first.

X's current strategy leverages Musk's visibility and influence (for better or worse).

Get the most prominent influencers onboard with your decentralized social network, and others will follow (dramatically easier said than done, of course). But without a significant contingent of influencers/powerful people, your network's DoA.

Re: Personal data storage is an idea whose time has come

#26
Among the first page and 2nd page (top 60) there is always atleast 1 post about how we're gonnna "take back the web" or make it back into some form of our 90s millenial nostalgia memories, self hosting, federated this or that, etc etc.

Meanwhile - Nothing changes, everything generally gets worse and younger generations come into the world with no memories of the 90s internet or the world before mobile devices or surveillence everywhere.

Applying for a job or apartment or anything today means creating endless pointless copies of your pesonal information in databases across the world that will eventually be neglected, hacked, exploited, sold off etc

I dont know the way out if there is one, I guess we can keep fantasizing and thinking about it. It just feels like it would be easier to get the earth to start spinning the other way sometimes.

Re: Personal data storage is an idea whose time has come

#27

How do I post a message on Discord/Twitter/Instagram from my personal data storage? If this is not supported, this idea is born-dead. Very few will use it, for the regular person the conversation goes like this: - Who can see my personal data storage posts? Can someone with Twitter see them? - No, but you'll own your data - Bye So maybe start with something which backs-up what you post on Twitter/Instagram/Discord to…

At this point distributed protocols are getting good enough that for a large class of social applications, network effects are the only thing keeping the incumbents in place.

The irony of ad supported free services is that if you just let the advertisers pay you directly for eyeball time then paid for your services, it'd be better for you financially while keeping the web pure outside of the "paid to consume ads" app.

Re: Personal data storage is an idea whose time has come

#28

How do I post a message on Discord/Twitter/Instagram from my personal data storage? If this is not supported, this idea is born-dead. Very few will use it, for the regular person the conversation goes like this: - Who can see my personal data storage posts? Can someone with Twitter see them? - No, but you'll own your data - Bye So maybe start with something which backs-up what you post on Twitter/Instagram/Discord to…

You just wait. The closed services will close down or become hostile enough that people will migrate. Not everyone will, but over a longer period - enough.

People getting into Solid and ATproto today are like people using own XMPP servers decades ago, or Mastodon years ago, or Matrix. Some projects like that will succeed, others will fade. But one day, you won't be able to post to Discord due to some policy changes and you'll have to reevaluate options.

Also, you can't backup from Twitter anymore. Or Discord. Or google photos. Or many others - they cut off that option once they're big enough.

Re: Personal data storage is an idea whose time has come

#29
post #6

The fact that the AT Protocol relies on everyone having a domain name, which is a centralized system over which few people have control, and about whose workings most people have no clue about, is problematic. Also impractical, once we consider that - as far as I can understand - 8 billion people should have their own domain name.

What's impractical about everyone having a domain name? It surely isn't due to lack of domain names, because foo.bar.baz.bim.bim.bap.com is a valid domain name.

It is true that full data sovereignty isn't something most people are interested in, but this is more about a cooperative model for data ownership and access. Having your data identifier be JackDaniels@yahoo.com isn't particularly different from it being jackdaniels.is.technically.bourbon.com. In both cases another organization owns some of the path to your identifier and could potentially lock you out of it. In both cases, verizon is near the top of that list (.com).

As far as the domain name system being centralized, I'm not sure I agree. DNS is like a feudal system with hundreds of kings (top level domains) who all work together with one pope (ICANN), and various lords and ladies occupying positions under those kings. If ICANN goes completely bonkers the kings can get a new pope, some of them are literally sovereign because they are nation states. Just for fun, some of those states are ruled by literal kings, too. There are experiments to run a TLD by Decentralized Autonomous Organization (DAO), but I think for the most part nobody really cares because the current system happens to work pretty OK. If you have an idea for a more decentralized way to organize a namespace that doesn't involve your grandmother typing in a massive UUID or onion address, and doesn't result in someone being able to domain squat literally everything; I would love to hear about it.

Re: Personal data storage is an idea whose time has come

#30
post #6

The fact that the AT Protocol relies on everyone having a domain name, which is a centralized system over which few people have control, and about whose workings most people have no clue about, is problematic. Also impractical, once we consider that - as far as I can understand - 8 billion people should have their own domain name.

It doesn’t really rely absolutely on domain names; at the very root there’s just a DID. DNS happens to be the best we’ve got right now as a human-readable username and address in-one goes.

We can work to make DNS /ICANN et.al. more democratically operated and people-owned while at the same time devising wholly alternate paradigms like Handshake and similar: https://blog.webb.page/2025-08-21-dap-the-handshake-successo...

Post reply on HN