This is a REALLY outstanding piece!

The one thing I think really needs some calling out, is that the percentage of apps that grow to the point where this discussion becomes super relevant is much smaller than people think it is.

I've worked in professional services for much (most?) of my career, and what I see time and time and time again are customers who are convinced that their project needs some sort of Netflix-like scale. Like... this is the app to record that you have calibrated the machines in your factory to the proper specs. Are you really going to experience such hockey-stick like growth in the number of machines in your factory and the number of calibrations done that this needs microservices, distributed architecture, etc. etc. etc.? Really? REALLY? Can we accept that a modern SQL DB can query millions of rows in a table by an index in a few ms, and no, you don't need this hyperoptimized structure for this?

As the folks who were greybeard wizards when I learned this business taught me, "premature optimization is the root of all evils" and that very much applies here. Creating a scalable architecture and infrastructure where it will never be needed is a disaster.

And I really loved how this article shows... all over the place... that at a small size, the scalable approach is less productive, and often less performant too.

Like I said, I really liked this. Thank you!

J.Ja