Fixing problems, or finding solutions for your problems? Every advanced system has a learning curve, and Unix is an advanced system without a well-defined domain. It's great that you invested the time to make it work for you, but that you feel that it is inappropriate for "most non-programmers" is not necessarily a failure of the system.
As a sibling post said, "there is not much interest in empowering non-technical users". I think that's close, but I would add that it's a problem of scale: non-technical users require domain-specific solutions. If there were one single solution to empower non-technical users, there would surely be commercial interest.
I'm sorry, but I can't help feeling that the article has a high dose of "lo! I've seen the light and my light will solve every problem on the planet". It ignores the many good things Unix has enabled, and is very light on the details how a better system would look (or how it would be secured -- befriend, really?).
(You could implement this different ways, but I'm imagining that once you friend someone they can see a select subset of posts on your computer. For me that would be status posts, some photos, some WIP code, etc. This would be a huge deal. Think about how many webapps it would make unnecessary.)
EDIT: Less flowery sentences.
For example, your code as shown implies that you can reach your friend's computer through an Internet name (through dns even?). That already assumes that these non-technical friends can register their own Internet names, and care enough to do so. But that assumption is well outside the bounds of "unix sucks".
Perhaps a charitable reading of it would be that instead of reading "unix sucks" (sadly I use almost these exact words in the article) instead read "the mismatch between the tools provided by unix and the needs of a normal user, combined with decisions made with good intentions at the time and kept for backwards compatibility, make the unix user environment a bad user environment for normal people today".
But of course I didn't write that, so I'm in no position to complain too much.
I agree that a unix command line is not a good interface for non-programmers, but it is good for people who have to work with large quantities of textual information, and source code IS text.
I think that the main problem is that there is not much (commercial or academic) interest in empowering non-technical users. There are some specialty tools for non-programmers like excel macros, or task automation in photoshop, but general purpose system that let the average user program the computer without actually programming it never took of.
I still think it is a very big stretch to go from web
browsers (which suck on many levels) to unix bashing.
This is a totally legit criticism. I think this connection is probably the most interesting idea I have in the post. Interesting doesn't make it right though.I just know from personal experience when I try to get people to switch away from webapps and they ask "what should I use instead?" I don't have a good answer.
I could say "use desktop GUI apps", but that's not really convincing. What would be convincing would be "use this awesome command line/REPL environment!", but I have trouble honestly suggesting that (even though I love my setup) because I know they won't have a good experience with current command line environments.
I am also a programmer, know UNIX before Linux was a thing, as my introduction was with Xenix. But I was already versed in systems like the Amiga.
The CLI might be interesting, but a REPL environment is so much more powerful.
POSIX is stuck with UNIX System V. It hardly changed anything in terms of OS architecture and they don't seem to be adding anything new, in terms of modern systems.
I wanted other people to share this experience, but couldn't make any converts. It turned out that it was only working for me because I was willing to plow hours into fixing problems. For most non-programmers, command line Unix isn't a good enough user environment to be a competitor to the web. And we really need that competitor.