Spring-rs is a microservice framework in Rust inspired by Java's spring-boot

> I suppose Spring is still a popular, widely known framework. Other than desired popularity, there's no resemblance and is antithetical. Anything I see associating with Spring, I would assume to be outdated and inefficient.

Spring is a battle tested, robust, optimized framework, that has scaled the largest brands in the world. It is literally what fortune five hundred companies trust their fortunes with. Spring is far from being outdated or inefficient, and will likely continue to dominate the market for years if not decades to come.

> Spring is a battle tested, robust, optimized framework, that has scaled the largest brands in the world. It is literally what fortune five hundred companies trust their fortunes with. Spring is far from being outdated or inefficient, and will likely continue to dominate the market for years if not decades to come.

I spit my coffee out on part of this one!

I’ve been a Spring developer for the better part of the last 10 years, and it definitely has warts, bugs, undocumented behavior, and plenty of source code diving for half-implemented features.

It does allow corporate middleware to easily be inserted, making management feel better and devs lives’ harder.

Don’t get me wrong, it has a placed in “corporate” development, but there are much more streamlined frameworks out there. In today’s day and age where SDKs are readily available and APIs are mostly sane, I think it has less of a place.

> but there are much more streamlined frameworks out there

Which ones would you recommend instead of Spring?

> It is literally what fortune five hundred companies trust their fortunes with

You’re overstating what “trust their fortunes with” means to a corporation. A framework is just technical detail. Often it’s tech debt if the framework is no longer fashionable. As has happened with J2EE.

Also: robust, heh. Spring has plenty of footguns and bugs. If you’ve deployed Spring anywhere, we should count the number of CVEs in production for your code.

And optimised? No. God no.

For devs new to the industry seeing this: a framework isn’t a runtime and isn’t a language. Learn how to write decent code without the training wheels of a framework. Then when it’s warranted, use a framework when it fits your needs. But don’t get into a situation where you’re totally dependent on the framework.

> likely dominate the market

They said that about J2EE as well. Didn’t work out so well. Again, knowing good programming technique > knowing a framework and better for career growth.

Specifically for Spring, its new owners Broadcom will probably look to making money off it. It’s already begun in a sense, a bunch of Fortune 500s are realising that Spring 5.x goes end of life in August 2024 (yes, this month) so it’s now time to pay Broadcom for sec fixes or upgrade to Spring 6.

It doesn’t mean Spring is dead but its proponents will have to work just a bit harder to justify the cost.

> I spit my coffee out on part of this one!

I don’t know if the Spring team have dev evangelists, but I got a chuckle out of it too. Such a dev evangelist thing to say.

I do agree with your assessment, btw.

I refrained from replying as different folks have different experiences and views. I'll just leave this picture[0] of what I consider not optimized. Most of my beef is with JPA/JPQL/but-most-of-all-Hiberhate--it's a 2000's solution for a 1990's problem.

[0] https://ptrthomas.wordpress.com/2006/06/06/java-call-stack-f...

Wait isn't Spring Framework all under "Apache License 2.0", so anyone could fork it and use it? Why wouldn't there be a group other than Broadcom contributing updates and providing support?

You can absolutely try. There are companies which provide support for out of support frameworks like Angular 1.

In reality there’s not enough Spring committers who don’t work for Broadcom for it to happen widely. (Also Broadcom has other options, eg follow Elastic and Cockroach Labs and change license.)

If you’re a senior dev, you need to be commercially astute — like finance folk can read balance sheets, you should be able to glean commercial implications from a project’s commit history.

PS. I’m a big fan of ensuring engineers get paid. If you actively choose to use Spring, buy a darn support contract and support its development. Given how many of you are on Spring 5.x or earlier even now, you’ll have to do so anyway if you want sec fixes. It’s a bit late to be looking for options in the short to medium term.

The answer is “it depends,” but personally I like ktor/micronaut better or golang’s stdlib. I do a lot of REST work right now, and I try to keep my services small.

> For devs new to the industry seeing this: a framework isn’t a runtime and isn’t a language. Learn how to write decent code without the training wheels of a framework. Then when it’s warranted, use a framework when it fits your needs. But don’t get into a situation where you’re totally dependent on the framework.

For devs new to the industry seeing this: Master Spring, Java, and the JVM, then coast and have fun, while counting your ever growing bank account. Don’t get caught in the vicious cycle of endless hype, “up and coming” languages, microservices, or K8s. Fads, like skinny jeans and man buns, will eventually just disappear back into the void that they came from, but at the end of the day, you’ll have a great career, work life balance, and no problem finding a job. You’re welcome.

I've also used Go without ORM-like libraries. For Java/Kotlin I've been trying out Javalin + JDBI which isn't too bad. You just have to roll your own things (or do the integration with what's available) sometimes. My apps start in 1/10s of seconds.

I wouldn't recommend it for a team that's used to copy/pasting from StackOverflow to get things done.