I recently read and shared You’re Wasting Time in Java Without These 10 Libraries. I commented on it a bit in my newsletter, but given the amount and intensity of reactions, I think a full-blown post is in order. The referenced libraries are:  Project LombokMapStructJUnit 5 & MockitoSLF4J with LogbackApache Commons Lang & Google GuavaJacksonHibernate ValidatorSpring FrameworkApache HttpClient / OkHttpLiquibase or Flyway  Lombok     Project Lombok is a java library that automatically plugs

Frankel's platform  covers topics related to software development, technology trends, and coding tutorials. Through articles, tutorials, and personal anecdotes, Frankel offers insights into programming languages, development frameworks, and best practices in software engineering. Readers can learn about software design principles, development methodologies, and emerging technologies to enhance their coding skills and build high-quality software applications.

A Java geek

A critical analysis of 10 commonly recommended Java libraries, evaluating their current relevance and usefulness. The author argues that some popular tools like Lombok and MapStruct are no longer necessary given modern Java features and alternatives like Kotlin, while others like Liquibase/Flyway remain essential for production deployments. The piece emphasizes context-dependent decision-making over blindly following listicles, noting that tools like SLF4J, Jackson, JUnit, and Spring remain valuable, while Apache Commons Lang and Guava have diminished utility in modern Java versions.

Are you really wasting your time in Java without these 10 libraries?

Pretty shallow article?
I mean, yeah, Kotlin is nice, agreed. But what the hell is the few paragraphs about mapstruct? It just says “probably your different layers actually don’t have much difference, so why add the overhead?”.
Yeah sure, but are you basically saying Dtos make no sense? Bundling domain objects for specific network requests, making your rather fixed API layer independent of your domain design avoiding breaking the API?
Same for the relation between domain and persistence model. For more than “university project” (even there) complexity, it is unlikely to be the same all the time.
Now, WHEN you need to write mappers, code can easily become complex, error-prone, repetitive and boring. The next best alternative to mapstruct is probably use an AI to generate the code. But why bother if we have this annotation processor?
Discarding a library with picking some specific antipattern situation that is more or actually less common and taking a verdict based on that seems not really like it.

Very useful article, it opened my stubborn head (again) to (more than ever) not to fall to the most used or trendy library/feature/etc, without asking first if it is truly needed or not. Like a true senior sw developer or tech lead would do.
Thanks a lot :)

The verdict on MapStruct is incredibly simplified, the moment nested (lazy loaded) entities come into play a tool like MapStruct comes in very handy and even it is also null checks for serialization.

Very nice article, thanks for sharing! 😄👍