4 ms·

Yes, Spring is so ratched barely anyone starts a project with it these days [citation needed]. Today "Spring" really means "Spring Boot" which is "fine" until y

Yes, Spring is so ratched barely anyone starts a project with it these days [citation needed]. Today "Spring" really means "Spring Boot" which is "fine" until you start to live the sugar-water scene from Men in Black and realize is still Spring all the way. Some good concepts and cool ideas, not gonna lie, but when that rubber meets the real road things turn ugly or at least mushy. Too bad Play Framework made a strategic mistake in V2 ( IMO ) and now it's pretty much dead. Javalin it's pretty cool but I'm not sure it's for the Spring crowd, Quarkus seems to be the new hotness for them with the migration to micro-services - I think it will turn into another multi-year BDSM session just like Spring but only time will confirm that.

joshlemer 5y ago
What was Play's strategic mistake?
- PedroBatista 5y ago
  Scala. Play was a very successful Java framework, with Play2 they went full Scala while still maintain Java support, theoretically that's fine but in practice we all know one has to be the step-child. Of course we have to take into account the context of the time, back then Scala was the hot ticket and Java sucked because it wasn't "functional programming" capable, but in practical terms they ended up splitting their resources and a large part of their Java user-base left. Once the Scala hype died down, the framework also died down. In a more nuanced angle, Scala wasn't the root cause. The root cause was losing focus and do other things too differently from what made the framework successful and that included implementing it in a new language. There were many problems on the ground just like every other framework, 2 that come to mind: - Horrible compile times with Scala, if you happen to have a some big project with 100+ classes ( which means real World ) it would be comically slow. Yes incremental compilation, yes it got slightly better over the years but it's still slow. - Big and many breaking changes every time you upgraded, the "WebService/WS" part changed completely pretty much every update 4 or 5 times ( from 2.2 to 2.6 more or less ). In general it seems to me they were experimenting a lot and respecting the time and investment of their their users wasn't a priority. Of course there are even more reasons, like the fate of Typesafe/Lightbend, etc but IMO the Scala detour was a strategic mistake. Having said all that, Play! is STILL one of the best frameworks in existence for Java today.

joshlemer 5y ago

What was Play's strategic mistake?

PedroBatista 5y ago
Scala. Play was a very successful Java framework, with Play2 they went full Scala while still maintain Java support, theoretically that's fine but in practice we all know one has to be the step-child. Of course we have to take into account the context of the time, back then Scala was the hot ticket and Java sucked because it wasn't "functional programming" capable, but in practical terms they ended up splitting their resources and a large part of their Java user-base left. Once the Scala hype died down, the framework also died down. In a more nuanced angle, Scala wasn't the root cause. The root cause was losing focus and do other things too differently from what made the framework successful and that included implementing it in a new language. There were many problems on the ground just like every other framework, 2 that come to mind: - Horrible compile times with Scala, if you happen to have a some big project with 100+ classes ( which means real World ) it would be comically slow. Yes incremental compilation, yes it got slightly better over the years but it's still slow. - Big and many breaking changes every time you upgraded, the "WebService/WS" part changed completely pretty much every update 4 or 5 times ( from 2.2 to 2.6 more or less ). In general it seems to me they were experimenting a lot and respecting the time and investment of their their users wasn't a priority. Of course there are even more reasons, like the fate of Typesafe/Lightbend, etc but IMO the Scala detour was a strategic mistake. Having said all that, Play! is STILL one of the best frameworks in existence for Java today.