[rodolphoarruda]

I've been catching myself thinking about this idea for the last two years. Maybe it's my old obsession with PKI and "personal digital infrastructures" that were both promised to us in the early days of the commercial Internet, but never turned concrete for various reasons.

IMHO, the best we could have today in terms of digital infrastructure is a personal/family level custom Mastodon node with basic Internet services like email, posts, tasks, chat, IM etc. but implemented in a way that all data would be portable to other services (open standards) and its storage would be "bottomless", meaning that users wouldn't have to worry about storage limitation for photos/videos for instance, as they would be sharing resources with other nodes worldwide. There would have to be some monetary incentive(s), of course, but they would be secondary to the bigger cause of keeping a true cyber interconnected community outside big tech.

[asim]

I agree with a lot of that. I think the hard part is, who runs the nodes. If you hand a piece of hardware that people run in their house that's one thing. But if you expect them to run it themselves in the cloud it never goes well unless you have an engineer in the family. Maybe automation can allow the ability to spin up these nodes but ultimately it might be easier to support multitenancy and let a group of people run it e.g like every other saas service. But I guess the difference is the values on which it's founded. Every commercial or VC funded product goes the same way. Whereas stuff like Let's Encrypt has gone in a different direction. I'm not saying I have all the answers but some of these things we always seem to struggle to overcome. One thing I will say, the people who run it matter. Their motivations and their morale flexibility affects direction e.g ChatGPT has led OpenAI in a very different direction than first intended. Why is that?