Data Pods

Traditionally, web applications have stored user data on their own servers, and thus, the server owners bears full responsibility for and ownership of that data. They can analyze it, sell it and manipulate it how they wish (which includes destroying it). Now, what if we flipped the script, and instead of web applications storing user data on their servers, they needed user permission to read data users own? The user would then have full ownership of that data, and could inspect, remove, change ownership and access as they wish. Applications would also store the data on the users server instead of their own.

This idea behind the SOLID project, in which users would own what they call data pods. In other words, users would store semantic data in a pod and grant applications access to read from that pod. That way we can move away from companies like Meta storing and keeping so much information about us. SOLID is not the only company exploring this concept. AT Protocol, developed by Bluesky, does something very similar. Users can store their tweets on Bluesky's server, or host their own PDS, or personal data server, retaining ownership of their data on the pod.

ICloud arguably can also function as a data pod. It enables apps to store data directly to the user account instead of a server, enabling the seamless synchronization that comes with having data stored on a server.

Lately, I have also explored Caldav servers, which share characteristics of a datapod. What I like about it though is that everyone is already signed up to some sort of email provider, and with that they most likely also have a calendar, which typically uses caldav as the backend. Caldav uses the iCalendar media-type, and it is surprisingly flexible; it supports events, todos and even journals. There are therefore a lot of potential for interoperable social productivity tools and note-writing tools that haven't really been explored.

- Marc