> I don’t think it is an exaggeration to say the amount required to learn to distribute software exceeds the amount I needed to know to write the first beta of HyperEdit! I wonder if it would have gotten off the ground if I started today.
This mirrors my feeling. It might be easier to learn to program say TypeScript or Go today than it was to learn php or C in the late 90s, but actually creating a complete program, distributing it, having it run or accessed by users is much harder than when I learned to program back then.
Same story in cloud. It takes less effort to write the program than to properly set up CI/CD, permissions, auth, auto-scaling, backup, monitoring, and so many more.
To be fair, bare-metal dedicated servers are still a thing, end up cheaper than the equivalent cloud offering on bandwidth alone (as they usually come with unmetered bandwidth), and you can just SFTP your project in there and run it with nohup (not saying it's a good idea, but if you're just playing around then it's perfectly fine).
I think that's doing it wrong because you don't need those features until you do... At which point you should invest in them. Otherwise, I don't see the benefit of having them. If all you end up with is one user.
You can blame the rise of authoritarianism (for lack of a better term) and centralised control, which Apple has been very good at. The increasing bureaucracy is ultimately beneficial to Apple's bottom line.
This mirrors my feeling. It might be easier to learn to program say TypeScript or Go today than it was to learn php or C in the late 90s, but actually creating a complete program, distributing it, having it run or accessed by users is much harder than when I learned to program back then.