Five things operating systems should have

Remember when operating systems were cool? One day we were trying to just keep our computers running messing with himem.sys and imagining being able to format a floppy, download a file, and play a gmae at the same time.

In the early 90s, we went from primitive operating systems to ones that offered preemptive multitasking, memory protection, object-oriented environments. Things were progressing at break-neck speeds. And then..things..started..to…slow down.

What the heck happened? Windows, Linux, and Mac borrow/trade/steal from one another each new version. But there’s been no fundamental OS improvement on the level of preemptive multitasking or memory protection added in years.

I can think of five things operating systems need:
1) Distributed Computing Features
2) Distributed File Systems
3) Global User Accounts
4) Universal desktop environments
5) Component based designs.

These things would not just greatly improve the usability of our computers, provide convenience to casual users, and improve productivity. They would also create incredible opportunities for new forms of innovation.

My article here goes into much more detail. But I suspect I’m not the only one a little frustrated that OS vendors seem content to generally just bundle more software into the OSes rather than add fundamental new features.