Comment by simpaticoder

The big question for me is how to give a child an experience that builds an accurate mental model of what a computer is and what software is. This was easier back in the day (like, Apple IIe times) when you ran one program at a time and if you didn't interact with it, it sat there, inert. (Yes I realize there is always some steady-state physical process going on, e.g. the pulsing of the clock-crystal, refreshing DRAM, etc. But I'm talking about software.) There is something beautiful and....solid about running a single process running in immediate mode on a simple CPU.

Contrast this with a modern experience. There are hundreds of inscrutable processes running, constantly talking to the network, to disk, doing who-knows-what. What does "software" mean on a modern machine? Software runs the gamut from a command line one-shot tool, to an invisible daemon, to a desktop app with a window, or multiple windows, to something hosted in a browser, to something hosted by something hosted in a browser. How can you build a clear mental model amidst so much noise, clutter and (both UI and runtime) heterogeneity?

In the same why "phylogeny recapitulates ontogeny" I'd argue the correct computer for a kid is a single process box with either hard-coded programs or a floppy-like experience (e.g. SD card "floppies). Given all of this, the perfect kids computer system is an HDMI TV dongle that runs an Apple IIe emulator, Logo, and a handful of retro games, and an SD card slot for other programs (prepped by the parent). No wifi or internet, bluetooth for connecting a keyboard, and no mouse.