What do you think about this? It’s crazy but such a cool story! 💥

Not all companies are AMAZON or NETFLIX. There is a cost in managing microservices, they don't come for free. Every business has different needs and we have to analise the best solution and the best architectural design. It could be microservices, it could be not, it is usually a mix of the available patterns. It depends. I personally think that in the last 10 years the IT business became very good in justifing costs and employments. We are good in creating new jobs to do the same things we were doing 20 years ago with an exponential complexity. With a team of 5 people instead of 1. And we are all victim of trends and hypes. No tool can compensate for human intelligence. Function as a Service is very cool if you ask a developer. But it is not cool if you ask a CTO o CFO when the software gets implemented by 10 different people in 10 different laguages with 10 different standards. Building standards is the drive for cost containment and system robustness. Converging, not diverging. I am not against evoution and I know it goes throug a divergent-convergent dance, but we have too much today. We are all looking for the silver bullet, I personally keep looking for simplicity.

Looks like trends are shifting and people start noticing that unless you are Netflix you will have more issues with microservices than benefits. This should be a chance for Grails. It should be much easier for Java devs to rewrite Java microservices to Grails than Ruby On Rails...

This is why I choose Grails again after building several projects with Spring Boot and Spring Cloud, keep it simple, use the one-person framework to build from hello world to IPO

Exactly. This is even more true when you start a new project as a single developer or a small team.

Yeah, Grails is better than Rails when we are talking about the JVM, many languages, so many open source frameworks and libraries could be used, also can be used for Microservices or FaaS, Groovy is so powerful and multi-faceted language

My experience is the "Micro" in Microservice is taken too literally, and I myself have dropped it from my vocabulary when referring to "services". I think if you have 70 services that can be replaced by a single service, then your domain was likely very badly designed for the 70 smaller services in the first place. An OpenSource community project I did some work in recently has fallen into the same trap, by making every noun in the domain a service. There is a real tendency to jump onboard a pattern or technology before the requirements are properly understood and this is the real issue I've found.

It is a sin to over-design and optimize too early without validating user needs and business value. Most people tend to choose technology that is popular and complex but not the right one for them

Maybe 1% of apps should start with microservices and 3-5% should end up there. It takes bravery, as a CTO, to admit that and migrate to a monolith. Grails has made solid progress with 5 -> 6.2. 7 with the latest dependencies will be key to telling its story for the future.

Think big, but start small, then move fast.