I think in the long run we will use a decentralised/federated system. Privacy concerns aside, it just doesn't make sense to rely on one service or expect one social network to meet everyone's needs/desires.
The massive collections of monolith data sets for financials, health/medical, credit history, employment records, taxes, census, etc are far too valuable to not be highly valuable to criminal orgs and/or government entities. Centralization into monolithic organizations will lead to irreversible issues of data non-privacy for a generation or more..
As an individual, I won’t lose my medical records because I have three backups and offsite and whatever. You probably do too.
But how do you remain really decentralized when the system has to work for a 76-year-old who can’t even remember what Google is, much less his password. Will this work in the golden hour after a stroke, when literally every minute matters?
So we would make a medical record “bank”, which is federated sensibly, and require everyone to use it. And eventually we end up with a similar system that we have now.
Except accessing your encrypted records require the presentation a doctor's key and your key, which might be a standardized health card with a chip. So not entirely similar to what we have now, it could be considerably safer.
If my card is lost or broken? The system you describe seems clearly at less risk of inappropriate disclosure. Whether it's safer is perhaps another question.
cf. the difference between "fail safe" and "fail secure", in a slightly different context
More relevant, what happens if you lose access to your data? Or it becomes corrupted? Now, because you owned your data, it's only your responsibility. You know who's not going to put up with that? Anyone who can afford to hire someone else (like a doctor) to keep track of their chart for them so that stupid stuff doesn't happen.
Most people don't realize that many of the connected health record problems have already had high quality solutions provided in the form of VistA (https://www.wikiwand.com/en/VistA). It's interface would make a startup designer cry, but medical professionals find it highly effective. Especially the network effect of being able to hand charts off and coordinate care, which would be very difficult in a decentralized health record world.
How feasible do you think it would be to use Mastodon for this?
If you have a "Medical Node" as you put it (which I find apt btw), you could share it with your doctor(s) and provide access (maybe using Keybase? Auth0?) for them to either add "their own posts" with the analysis results, their findings, etc, or just to read the relevant data (e.g. for a dentist).
Again, just thinking out loud, but they way you put it sounded interesting :)
https://www.vox.com/health-care/2017/10/30/16228054/american...
"It turns out there are strong economic incentives for doctors to keep patient information to themselves — and even stronger incentives for electronic medical records not to play nicely with each other."
"While patients might want one hospital to exchange information with another hospital, those institutions have little incentive to do so. A shared medical record, after all, makes it easier to see a different doctor. A walled garden — where records only get traded within one hospital system — can encourage patients to stick with those providers."
On the other hand, with old fashioned paper journals, you could get all copies and store them yourself. If you lose them they're gone too. If there was a cryptographic construct where the access of data would require a public action, i.e. "to assemble the key to decrypt this data, I need to broadcast this to the world", then I can think of a number of good ways to store things with a robust accountability feature.
https://gdpr-info.eu/art-20-gdpr/
In particular, there is this wording:
"In exercising his or her right to data portability pursuant to paragraph 1, the data subject shall have the right to have the personal data transmitted directly from one controller to another, where technically feasible."
This could, in principle, require Facebook to automatically broadcast your posts to your friends on third party social networks, once a standardised technology for that becomes widely implemented. Fortunately it seems that progress is being made on that:
It's what has always worked for me.
I feel like reasonably effective retro-active deletion is an important feature of a social network, as it gives users more control.
No, there are middle grounds, such as a decentralized/local authorities. E.g. you send a hyperlink to the message on your server, rather than the message itself. Theoretically, you've still ceded control to the receiver, but practically, in most cases, you haven't.
Links to photos, events, and group chat can all happen using email as transport. Like @esfandia suggested, the social-emails could be tagged as hidden so they are not displayed by your regular email reader.
Here is a old write up of how "events" could work over email: https://news.ycombinator.com/item?id=12671812
Does anyone know if something like this has been tried before? Surely someone has thought of this before...
I’m being ironic cause we did end up with a centralized system for email. Not a single center, but a few: Gmail, Yahoo, Microsoft, etc. Because of abuse (spam, primarily) and because of ease of use.
Having an easy to use and safe federated system is discovering-the-theory-of-relativity hard, in my opinion.
But also, those three are by far not the only ones, there are quite a few more significant email providers across the globe.
And also, there are tons of participants who use none of those services. Apart from some individuals running their own servers, lots and lots of companies run their own email infrastructure.
> Having an easy to use and safe federated system is discovering-the-theory-of-relativity hard, in my opinion.
The important thing is that it's way easier than a safe centralized system.