The way to get there isn't the same, the impact on Java library authors it is.

Even Jake Wharton already complained about this publicly, and he is now working at Google, go figure.

> Happy Tuesday Android developers! Java 13 out today. Java 8 is now 5½ years old. Here's your bi-yearly reminder that Android is holding back the Java library and language ecosystem.

https://twitter.com/JakeWharton/status/1174020113466568704

> The Android platform is on the cusp of holding Kotlin back. New bytecodes not being available both in the Dalvik set but also on devices in the ecosystem means Kotlin is forced to target older Java bytecode versions which can hinder new language functionality.

https://twitter.com/JakeWharton/status/1017597264225857536

> Kotlin is great but there's no reason anyone should need to target Java 6 bytecode and APIs in 2019. IntelliJ platform switched to 8 in 2016.1 and 11 in 2019.1. Servers are mostly 8 trending towards 11.

https://twitter.com/JakeWharton/status/1174024747400777729

So we are just in the same position as if J++ would be kept being a product, any Java library author that wants to have market share on Android needs to constrain themselves to the Android's view of what means to be Java.

And I have to bow to Jake as he is doing a very good job adding support to R8 and D8, to at very least desugar modern Java features into Android Java.

But by that logic, any large organization using an old JVM version and requiring developers to target that version would be holding back the Java ecosystem.

Exactly, with the small detail that almost everyone else does provide support for more recent versions on their Java implementations.

> any Java library author that wants to have market share on Android

Had Google used a completely different language instead, this market wouldn't exist at all, and Java library authors wouldn't even have the option of choosing to target it.

I don't disagree that Google's handling of the language is bad, but that doesn't mean it should be illegal. Creating a bad legal precedent harms us all.

The same could be argued as defence for J++, but naturally Google gets free pass, because it isn't Microsoft.

Microsoft's actions weren't ruled illegal, they had to pay because they broke a contract.

> there's no reason anyone should need to target Java 6 bytecode and APIs in 2019

Java 7 byte code was a big mistake. People complained but Oracle didn’t listen. As always.

http://chrononsystems.com/blog/java-7-design-flaw-leads-to-h...

Yet everyone else on the Java world doesn't have an issue having up to date Java implementations.

IDE, debuggers, bytecode libraries, build systems, profilers, all of them had to deal with “up to date Java implementations”. Which hurts adoption.

Not just because of this issue, but because of tons of other mistakes made by Oracle: users shouted, Oracle ignored.

The net result is inertia is high, but Java is stable declining. Existing projects continue to be developed in Java, new projects often choose something else.

That's not true. BD-J is limited similarly for instance.

They forked the shit out of it too with tons of base functionality being in org.dvb. and javax.tv.