Totally agree. It’s great that the Linux desktop experience has come as far as it has, but it still suffers from compromises as a result of being general purpose at best and server oriented at worst. Haiku can make technical decisions that make sense for a desktop OS without also negatively impacting other uses.

> We need an open-source operating system 100% focused on the desktop.

I agree. In my opinion though, Haiku's insistence on becoming just another platform to run your UNIX software stack on means it will probably never succeed at that goal.

In case you haven't noticed, a majority of people on Windows or macOS use software that was originally part of the "UNIX software stack" -- between SSH, bash, GCC, FFmpeg, etc. I am pretty sure you could find at least one piece of "originally UNIX" software that at least a majority of programmers using any given OS use.

Anyway, I think you miss the point of Haiku if you think it'll never get anywhere because of how much UNIX software runs on it. It's not here just to "be different from everyone else." The POSIX API makes a lot of sense; anyone going against it has to have a particularly good reason to do so. Even Apple doesn't, for the most part.

Unix is indeed ubiquitous, but it would be nice to have a personal computing system whose core metaphor is something other than a teletype.

A majority of programmers on Windows perhaps, certainly not a majority of people on Windows.

Not even the majority, UNIX refugees, yes.

What do you propose as alternative? All of these UNIX tools provide backwards compatibility.

Plan9? But it's still ahead of its time. I think it would be quite hard to port something like Haiku UI to such a platform. POSIX is fine.

Actually Inferno, Plan 9 was the middle iteration.

We have virtualization nowadays to cover compatibility requirements.

How is it different than macOS?

I think macOS/Mac OS X was always known for fantastic Mac-specific software as opposed to just ports of user apps from, say, Linux — probably because Mac OS X was the successor to Mac OS 9.

So Mac OS X inherited one userbase that appreciated platform-specific software. Haiku doesn't really have anything to inherit like that so until people start making Haiku-specific software, that could be an issue for seeing great Haiku software rather than just ports.

In the beginning, there was also a good deal of people coming from NeXt and finally finding a target audience. Some of those applications fared even better, as they were basically Cocoa already, whereas some of the older Mac apps were more baroque, not supporting all the nice new features and having a more outdated look. So the "newcomers" served as important style guides. (Let's not even mention the backlash against further outliers like Java Swing apps or RealBasic)

For a while, the OS X app space was rather "pure". A minimalistic aesthethic, plenty of small opnionated productivity applications. Apple itself didn't wreck or disrupt it too much (looking at you, MS Office).

Then mobile happened.

That's a great point. A lot of the best macOS software came from, and still come from, developers from the NeXTSTEP era. Some have faded a bit, but they're still there setting the benchmark.