A Late Night Thought About Operating Systems
It is late. The desk is quiet. I have been sitting with a thought for the last few hours and I am not ready to talk about it in detail yet, but I want to put a marker down while it is fresh.
Modern operating systems are built on assumptions made in the 1970s. Files in folders. A single machine owning its own resources. You log into a computer, not an environment.
Those assumptions made sense then. They do not make as much sense now.
Networking is fast enough that devices sitting a metre apart could share compute, memory, storage — seamlessly, transparently, without the user ever thinking about it. The hardware has been ready for a while. The software model has not caught up.
I have been thinking about what you would build if you started today with none of that baggage. No legacy to support. No decisions inherited from a world where a hard drive cost more than a house.
There are some interesting answers.
I am not ready to share the details yet. There is still a lot to think through and I want to let the ideas breathe before committing them to a post. But the direction feels right, and the timing feels right, and I have learned that when both of those things are true at the same time it is worth paying attention.
More soon.