Or stated differently: Most people are morons. Do not make majority-decisions if you want quality, reliability, etc.

Well said.

That may be intentional. In fact they confirm it is. What better way for an attacker to cover his tracks after a successful break-in then being able to credibly corrupt the logs.

The hallmark of a system that has gotten far more complex than it has any business being. I foresee that the standard answer to Linux system problems will be to "reinstall". I think I have heard that utterly primitive and anti-intellectual advice somewhere else before...

There is some suspicion that the main goal in systemd is to make the init-system vulnerable to well-funded attackers. Especially the binary logs are a huge red flag.

Sendmail is solved? Not from what I can see. I took a look at it again a few years back and decided that Postfix was a much more sane option.

Makes sense. As virtualization comes with its own limitations, they would probably not even notice what they are breaking.

I know somebody that tried this. At around 5000 threads he got no real progress whatsoever anymore.That was a while back, but at that time Java was already a few years old.

I certainly agree to your observation. The question is whether conformity is required to standards (whatever they may be) that have nothing to do with the skill and quality of work that an individual brings to the table.

I also noticed that people desperately wanting to be "hip" and "rebels" and "modern" and "informal" are often anything but and often also have rather low skills. These are basically using their "informality" as "formal attire", and fail just the same to focus on skill and capabilities instead of on conformity. Most of them also do not understand that enforced informality is just the same conformity they claim to reject.

Simple: cat /dev/zero > /dev/sda

Good hypothesis.

This is a typical thing: The "modern" solutions are all bells and whistles, but are missing basic things or do them wrong. More often than not, this is because they were designed by people that saw some real or perceived shortcomings in older tech, but completely missed its strengths and failed to reproduce them.

Example: Ever tried to do massive multi-threading in Java? That fails miserably and the byte-code interpreter does nothing but task-switching very soon. In an "enterprise" setting, these things are toys when the back-end is considered.

I think you are quite mistaken when you look at reliability, scalability and stability of the definition. These are all huge cost and risk factors. Of course, most "programmers" do not even understand what these terms mean.

My guess it is mostly from people that find learning languages really, really hard and that try to get by with one (or sometimes two) languages, like Java and JavaScript. When these people are told "why not do COBOL?", they immediately freak out, as they are reminded that they are not really programmers, but 1-trick-ponies and actually have really no business producing software because they suck at it. And then they feel that COBOL is not something new and it is safe to lash out at it. Unfortunately, these people represent the majority of all "programmers": http://blog.codinghorror.com/t...

Conversely, competent programmers will just see it as yet another language and learn it at need and without any issues in a few weeks. A friend of mine once had to learn APL (the one with the bizarre symbols) and even that did not take more than a couple of weeks to become productive.

The typical observation is that the more formalism (not only clothing), the less actual skill and insight is present. "Show" replaces "skill".

Very much this.