4 ms·
Every time I think about this issue, the closest I can get is a distributed world of APIs, where everyone hosts their own data [however], and for any service wh
by enobrev 7y ago
Every time I think about this issue, the closest I can get is a distributed world of APIs, where everyone hosts their own data [however], and for any service whether it be government, commercial, work, friendly, etc, the entity's agent makes an API request and gets a result. But that result isn't the data - it's a privacy-focused result that can be used in place of the data.
The simple example would be "I'd like to send indigochill a gift", which would send a request to your API (with some notification for you to approve or deny receiving said package) and return an id that could be written or printed on an envelope. That id is unique and the mail service would scan the id and the package would be routed. And the mail carrier would scan the id and get an actual address (or a blip on their map of the block or something). Nobody along the way has learned your address or name or anything - nor do they know my address, which is just another unique id.
There are some very interesting implications in that data can become stale so quickly. If you approved of receiving the package but were out of town for the week, the package could be routed directly to you without any interaction whatsoever. Nobody along the way would know the difference between your home or your hotel room.
The same could easily be done for lots of personal data. Phone numbers, email addresses, etc. All your shopping history could be stored on your own private data store and the local grocery stores would send out a mailer "Address this mailer to all people in the neighborhood who have purchased discounted burgers in the past month". The store doesn't know who is getting the flyer, but the people who would be likely to use a coupon for buying burgers would get a notification and likely go buy some burgers on sale - or reject the notification and not be bothered - to the savings of the store and the recipient.
Social Media could still exist with this sort of thing but the services would be querying engines rather than the creepy data warehouses that they've become.
This of course gets significantly harder with the queries this article is about, but the idea is the same. Instead of "give me everyone's location for the past month", the request would be "give me the anonymized ids of people who have been to x locations for x amount of time within x amount of space between one another". The response would provide data that could be used for the purpose in question, but the underlying data remains private.
I've been pondering this idea on random occasion for about 15 years, but I don't care enough about it to make it my life's work. But I sure would like to live in a world with that level of data ownership and privacy.
- thedudeabides5 7y agoSounds like an interesting possible world that might one day get built. In the meantime, sounds like most likely outcome is the government hits CTRL C, CTRL V on Facebook/google’s user database and starts a China/Tencent style SQL dB on every American. Or even more likely, this already exists and now the government can start using it more openly.
- thinkmassive 7y agoThis could be possible if the data is stored encrypted in a public ledger (blockchain), which has time-locked script functions that permit another key to decrypt certain pieces of data for a limited duration. I think something like this is a primary goal of most decentralized identity projects, some of which have been underway for around a decade. Of course none of this stops the person with temporary access from making a copy to save forever. I doubt that problem has a technical solution but would love to be proved wrong.
- blocko 7y agoI think the solution would be sending as little data as possible to answer the query, and choosing which/how many queries to respond to. Any large scale requests for data would have to be designed in a way that minimizes intrusiveness or risks being largely ignored. I don't think there's a great way to counter spoofing or sending garbage results without sacrificing user control, but the onus to de-noise should be on whatever institution is attempting to collate data. Additionally, if it's an opt-in system focused on public good there would generally not be much incentive to mess with it as long as users are somehow verified as being in the community (cert generated/granted once after initial identity check, etc). Sadly this is all very pie in the sky at the moment, the amount of cultural change + added infrastructure required for any system like this at the scale of a country would be insane
- thinkmassive 7y agoThat's a good point. A common example is how a bar doesn't need to know your birthday, let alone your home address or driver license number, but right now it's common to hand over your entire id. All they need is a yes/no answer to "is this person 21+ today?"