Tim Berners-Lee is leaving W3C and has started a company called Solid. This is really interesting stuff. I spent a couple of hours the other night reading up on it. It amounts to a new architecture for database apps. Well, not completely new but if Berners-Lee popularizes it then it will be a big deal. I haven’t found an article that explains it well. Here’s my take:
Think of your contacts list. We can set up the database architecture a few ways. (a) I can keep all my contacts on my computer. If I want to share my info with you then I send it to you and you store it on your computer. (b) I can do the same thing but store that data in the cloud. (c) We both can store our contacts on the same service in the cloud and I give you permission to see mine and vice-versa. Plaxo did this; LinkedIn has a fancier version of it; Facebook is essentially an expanded contact info form. (Downside: someone is mining that data,).
(d) Berners-Lee is promoting another way. I would store my info on my computer (or server, local or cloud). You do yours. When I scroll to your name in my contacts app and click, my computer makes a request to your computer (or server) for your info. If you’ve given me permission then I get to see it. My computer doesn’t store it. This works as a simple web URL request. Each time I look up your name in my contacts list, my computer makes another request. Each piece of our contact info would have a unique URL. I could let you see some or all of it and revoke permission whenever I want. If I update my info you would see the updated info automatically.
Berners-Lee doesn’t say it but there would have to be a personal data license agreement that goes along with the data, to prohibit the recipient from copying the data or forwarding it to someone else. When I ask for permission to see your contact info the first time, your computer would respond “if you agree to the standard PDLA, click here.” Maybe people will respect these agreements, maybe they won’t.
The key is this is the opposite of the current structure, where data is aggregated with (typically) no restrictions at all. The other key is that there is no need for a central server for low-transaction services like contacts, or photo-sharing, or feeds among a personal network like Facebook, LinkedIn, Snapchat, Instagram, etc.
I wonder whether this concept could be utilized with Filemaker perhaps as an adjunct to the usual FMP database architecture.