That's true that I put an explicit emphasis on C#

But let's face it, the whole CoreFX library is in C#, Roslyn Compiler is in C#... the line is already blurred

So while the CLR is not language specific (like the JVM is not Java specific), rewriting the runtime with the most prominent language used for the underlying platform sounds an acceptable implicit emphasis ;)

Fair enough :)

I don't think It's a problem of level of investment from Microsoft. They have quite a good chunk of top level engineers working across the board on .NET. They are making lots of progress in performance area in C#, including the work on AOT on CoreRT...

No, in my opinion, the critical issue that .NET has been facing is not to be able to enter the academic circles. I remember to work in Java 20 years ago in my engineering school, and Java was already there... going to the same school a few months ago to present .NET and Unity... and students were literally discovering .NET! The amount of complete misinformation is very scary there... That's the unfortunate backslash of years of Microsoft not being friendly with the Linux&OSS community (closed source, no OSS, proprietary APIs to lock users to their OS, trying even to play not nicely with OSS...etc).

Now a lot has changed at Microsoft for the past years to improve this situation, it is definitely not the same, a lot more open. But still, look, .NET is barely visible in the academic circles. It will take years of education to re-balance things. In the mean time, the Java community has been working so closely with academic circles for years that they have been able to bring many interesting breakthrough... not counting the super large ecosystem that Java has, directly and indirectly because of that. There are a lot more Java OSS friendly developers... while a large part of .NET developers are more corporate-closed-source developers that are not sharing anything... despite the .NET OSS movement that has been slowly growing in the past years... yet, not at the level of Java.

Though, luckily, the design process of the C# language has been a lot more streamlined than Java (faster evolution in many areas, better choices at the beginning - value types, a lot more control of memory layout, no generic type erasures...) and in my opinion, been able to bring more interesting features than Java has been able to bring. Also, performance wise, Java is today, in many tests, behind C# and .NET Core. So things are not that bad after all.

About the Jit Compiler, I'm really please to see this Graal initiative in Java, as it can help to further bump the idea that it would be very relevant for .NET as well!

I depended where in the world one was located.

The Portuguese university where I graduated from, had an agreement with Microsoft Education, I remember visiting it several years later, and getting positively surprised that all compiler design classes had migrated to CLR as target platform.

The same classes where the first ones several years earlier to adopt Java, alongside JavaCC, soon after they got released.

Very true, let me correct that my comment mostly applies to what I have seen in France ;)

IL2CPP is going to stay for sure, but imo, it should be reserved to platforms where we don't have a more efficient way of having GC aware codegen (because it has not been ported yet, typically for a new platform).

CoreRT could perfectly work at some point on iOS, and more efficiently than IL2CPP. But this will require lots of work, so indeed, this is not going to happen anytime soon, maybe never, but we should be aware of all the options and why they each matter.