Ask HN: Is Accidental Complexity Growing Faster Than Essential Complexity?

You could argue for equivalence under a broader definition of optimization and engineering, but in the context of software engineering the more specific definitions used above seem to be fairly well established.

The kind of over-engineering for the CV polishing almost always comes from our college hires. It's not really their fault, as I understand the attraction to the shiniest object when your a newb. When their monstrosities are allowed to make it into production, however, is when you have a dev manager that doesn't really understand software engineering.

For my part, I used to scream from the mountain tops that we were over-engineering everything, but now I just let them fuck shit up cause I'm tired of trying to save people from themselves.

There is a parallel to this where people were trying to where people write c++ programs in visual studio realize they often can do so without needing the microsoft visual studio runtime and its routines and their program drops from 68k to 2k and a dash of code to bring back what is missing in most cases, if it is even needed. https://hero.handmade.network/forums/code-discussion/t/94

I love looking at things and thinking about this, people in this field often find themselves over-implementing(or using software that is a source of over-implementation), setting up a tremendous amount of foundation to get a little thing done, getting carried away, and finding out it got too involved and over-engineered when there is a tight and simple solution that matches exactly what is needed.

Of course there are situations where microservices are still needed: https://semaphoreci.com/blog/2017/03/21/cracking-monolith-fo... Everything should be thought out and joining the latest craze isn't necessary. If you really need it, you will find yourself looking for or implementing the solution, regardless of what it is currently called, microservices have been around in one form or another before the hype, and used as needed when needed, instead of getting being used to wax a resume.

Your accidental vs essential comparison has real world monetary consequences. There are many examples that while it might take 2-3 months to stop and think and analyze what is going on instead of adding to the pile, you can save 100k-300k a month for even smaller scale deployments.

Anything beyond HTML forms, a SQL database, and maybe a bit of Javascript and CSS is likely to be more complexity than you need for something like this.