>>952434
>I think that there is mainly one person who is posting all this anti-unix stuff.
I never pretended to be more than one person.
>Maybe they are actually suggesting that we don't use any computers at all?
UNIX is extremely bloated and buggy but we shouldn't give up using computers just because some AT&T employees weren't good programmers. Lisp machines show that you can have a higher quality OS and better programming environment with much less code. Software can be faster, smaller, more reliable, and simpler than it is now and do more at the same time.
>Unix has its problems, just like everything else. But so far I can't find any other OS that I would rather use.
The number of problems in UNIX is much higher compared to other operating systems. UNIX commands are so inconsistent they look like they came from different OSes.
>1) Text interface
That's because of the PDP-11 hardware. Most computers in the early 70s used a text interface.
2) Shell and pipes
Pipes are virtual PDP-11 tape drives which are based around moving groups of single bytes at a time and seeking to different portions of the tape. This is another example of how UNIX makes things less efficient, less secure, more prone to errors and vulnerabilities, more bloated, and worse for users, all at the same time, like having to serialize and parse JSON or XML instead of sharing the data like on Lisp machines and Multics.
3) Modularity
UNIX is less modular than other operating systems. UNIX weenies like to point to the "tools" as if having to start separate programs to "cut" and "grep" is better than being able to use a single language for text processing (like Perl, Python, or Lisp). Marketers were able to turn a workaround for the shell sucking at text processing into a "philosophy." UNIX doesn't help to make newer programs modular since they're not based on sequential tape processing.
>Anybody who deals with complex systems or ideas will probably agree that breaking a problem down into small and simple elements is the best approach.
The design of UNIX tools is based on low-level byte at a time bullshit instead of what you want to do with data. That's one reason XML and JSON are so popular, because they let programs interpret data on a higher level than individual bytes. When I break a problem down, I want to do it in the way that's best for the problem at hand, not in some artificial way forced by the OS because it wants me to pretend I'm writing to a PDP-11 tape.
>What I would really like would be something very bare bones, like CP/M, with the features of unix that I really like.
That would be a better choice than UNIX.
>>952438
>Unix is not perfect, but Unix and its derivatives are the best we have right now.
Intel x86 is "the best we have right now" too, but it sucks. AMD64 killed the best parts of x86, Intel ME is a nightmare, and there are all these exploits coming out.
http://www.loper-os.org/?p=42
>My standard of comparison for any technology will always be everything previously achieved by mankind, rather than what is available on the market today.
Hmm. I used to think the strength of lisp machine tools
came from the fact that the developers actually used them
regularly in their work and depended on them in order to
develop everything they were going to need in the next
generation system. That is, I though that there was a
causal link between using your own tools and making them
better.
But maybe it's not whether you use your own tools that
makes them good, but rather that the goodness or badness of
your tools is just magnified over time by continuing to use
them. That would explain a lot of things about Unix...