That's the inevitable evolution of any popular open-source product/company. Light and enterprise-ready is hardly compatible, and you can expect this from any project that has a cloud offering.

Not only open source.

I don't have servers like this but the author makes it easy to understand, and it applies to a lot of other things.

The Comparative Costs picture does tell it all.

The purpose of so much software is to reduce human and machine costs through time, and this apparently turned out to do just the opposite, apparently after long-term testing under real-world conditions.

Could be an unsurmountable fundamental structure of technical debt or something like that which metastasizes.

Then this, anyone could say about anything:

>I’m not going to run a piece of software that requires 16GB of RAM, has a complex installation script, and is known to be a pain to maintain. I’m not going to recommend it to anyone else either.

Easier said than done, sometimes it's the only option.

It's a "script". Maybe that's why you have to "rehearse" it more than once before you barely get it right, and then you might have to really go the extra mile (and have a bit of good fortune) before you can achieve a "command performance".

How do you think it feels these days to have to settle for this kind of thing in expensive proprietary software too?