Thanks for your insights M. Darrin. Could you give me some insights into where Docker is less mature than JVM today?

I also don't think the size and effort put into making an ecosystem should necessarily be used as an argument for why it is good. Sometimes you can go enterely in the wrong direction making most of that effort entirely obsolete.

One example of this was the arrival of RISC processors in the 1980s. Long time CPU designers in the industry with decades of experience and tools on their back got beaten by a small group of students making a simple RISC processor.

A far simpler design completely outperformed the complex designs made by industry insiders. That shows that if you got the fundamental idea wrong, it doesn't matter how many centuries of work you have put on top of that idea.

Now this field is not mine, so I got to be humble abou this. But read articles as an interested bystander, and I cannot help but notice quite a lot of guys from the JVM world which has discovered Elexir, Erlang and BEAM and concluded this is superior solution to monitoring and tuning your application.

https://mrjoelkemp.medium.com/jvm-struggles-and-the-beam-4d9c58547410

This also seems to be to be more of a real world advantage of using a virtual machine. Java and C# just seems to use the VMs to abstract away the hardware underneath.

BEAM uses it to produce an entirely novel environment with virtual processes which communicate and whose communications you can easily monitor.

When people who have spent their whole career on the JVM and it has a really mature ecosystem and you got these guys finding that they can solve their problem on a much less well known platform like BEAM, that tells me that maybe the JVM isn't all that it is cracked up to be.

You can solve a lot of problems by throwing a lot of manpower on a bad idea, but there are diminishing returns. Eventually you hit the wall.