This is not true, actually. JDK1.1 1997 had JIT. MS had basically exposed the Variant and COM, the windows handles and part of WinAPI. Overriding paint(Graphic) on a native AWT component (say java.awt.Button) did nothing either and so on.
>>Instead of improving the JVM, Sun decided to make money by suing MS. No wonder that bloated piece of slow crap that looks like crap is dying a slow death and stopped being relevant.
JDK1.3 with hotspot compiler (May, 2000) had the fastest JIT by far.
> It was the fastest Windows-based implementation of a Java virtual machine for the first two years after its release.[1] Sun Microsystems, the creator of Java, sued Microsoft in October 1997 for incompletely implementing the Java 1.1 standard.[2]
Which is the fastest JVM after 1997 is essentially a moot point after the lawsuit was filed.
I also like how Apple completely sidestepped such kinds of situations in iOS by disallowing third party SDKs and alternate browser engines. If there can be no Netscape or Java, there is no threat and no need to kill something and potentially get into trouble. Same with Google and Chromebooks.
http://daringfireball.net/2010/04/iphone_agreement_bans_flas...
The clause was removed just 5 months later; there were rumors that potential antitrust issues were involved:
http://arstechnica.com/apple/2010/05/apples-compiler-policy-...
History repeats itself.
> The more important question is, was there anything that prevented Sun or anyone else from implementing those in the other OSes?
Yes, they made calls to native microsoft frameworks etc.
> I do remember the MS JVM was significantly faster than Sun's.
Lets dig out the binaries and do a perf test.
> So the solution was to slow it down or for Sun to make it's own faster?
This comment makes no sense. No idea what you mean by it.
> It's in Microsoft's interest to make things run faster on it's own OS.
This isn't about performance
> How is that a good thing and what MS did bad?
Who said it is any different?
> Instead of improving the JVM, Sun decided to make money by suing MS.
Yea they just abandoned the JVM in 2005. Hasn't changed since.
> I remember using a popular bittorrent client called Azureus that ran on the JVM.
Java for front end had requirements that most desktops of the time failed to meet. So what?
> Yes, they made calls to native microsoft frameworks etc.
Uhm, yes, that's how you run things on another platform. You have to call that platform's APIs. Sun's JVM won't be able to create an AWT window without calling WinAPI's CreateWindow. They have to call the native APIs at some point or another.
The simple facts of the case are that Sun license the Java brand to someone who wants to implement a JVM/API but you have to pass the TCK/JCK testsuite to keep the license. MS would have failed that.
The more important question is, was there anything that prevented Sun or anyone else from implementing those in the other OSes? Otherwise it looks like programs are limited to the most basic level everywhere. I do remember the MS JVM was significantly faster than Sun's. So the solution was to slow it down or for Sun to make it's own faster?
It's in Microsoft's interest to make things run faster on it's own OS. For example of a similar issue, see Flash on OS X and Jobs' memo regarding how it sucked and made OS X look bad and refusal to allow it in iOS. How is that a good thing and what MS did bad?
Instead of improving the JVM, Sun decided to make money by suing MS. No wonder that bloated piece of slow crap that looks like crap is dying a slow death and stopped being relevant.
I remember using a popular bittorrent client called Azureus that ran on the JVM. My laptop would get super hot and slow to a crawl. Also remember how J2ME made mobile Java development a complete hell before Windows Mobile and iPhone came on the scene?
http://www.zdnet.com/three-billion-devices-run-java-yeah-but...