"For the last couple of years the community has been working to ensure that developers can create applications using Java without having to depend on proprietary software. Today, the Free implementations are already very capable and support a vast amount of functionality that developers expect from a Java-like environment. Important large applications like JOnAS, OpenOffice.org 2, Eclipse 3 and Tomcat 5 are known to work and now included in distributions like Ubuntu and Fedora Core. This document provides a road map of the various projects; how they work together, where they are, where they're going, and how we make sure that they work well and are compatible."
To view parent comment, click here.
To read all comments associated with this story, please click here.
Member since:2005-10-08
> Oh absolutely. Just like how Unix-on-Intel
> killed Linux.
But until very recently, Unix on Intel was not something the average hobbiest could afford. And lets not forget that initial birth and growth of Linux was entirely due to hobbiests around the world.
The basic issue when it comes to Java is that most people don't care enough about the fact that it is not GPL to dedicate a lot of effort to producing a GPL version. It's free (as in free beer), it allows me to mofify the source code for internal use, or research purposes, and I can even contribute code back to Sun for inclusion in Java if I fix a bug or something. In addition, Sun is very good at conducting informal surveys and such basically asking the community "What do you want in this area of Java? How important are A, B, and C to you? How do you think we should handle issue D?"
In short, Java is open enough for the vast majority people. There is only a relatively small faction of people who are extreme enough to say "GPL or nothing".
The vast majority of people are quite happy with the Sun JRE / JDK. For an open source JRE / JDK to be accepted by most Java developers and the corporate Java users who have millions of dollars of business depending on their Java applications, it is going to have to be able to pass Sun's compatibility testing suite--a very tough trial indeed. Sun does offer free compatibility certification and testing to recognized open source organizations like the Apache project. So Harmony has a chance. But will it be able to attract a strong enough interest base of people who care enough to not just stick with the Sun JRE / JDK? That remains to be seen.
Member since:You seem to be bashing a strawman, here. GNU Classpath, Apache Harmony, Kaffe, and various other projects in that niche are far from being "GPL or nothing". GNU Classpath VMs exist in the whole wide licensing range going from zlib (IKVM), apache 2.0 (JCHVM), lgpl (sable vm), cpl (jikesrvm), to good old gpl (jamvm, cacao), and that's great, as it allows people to chose a VM for whatever licensing option they need.
As for passing the test suite, that should not be that hard, actually, as more and more applications work out of the box, meaning that the functionality provided by GNU Classpath meets the specs in those areas where it's implemented and extensively tested. Once any GNU Classpath using VM passes those tests, the floodgates are open for the other dozens of VMs using the same GNU Classpath class libraries to do the same, and earn the certification mark, should they so desire. Tweaking the VM to meet the specs is much easier than writing the whole huge class library from scratch, in my experience.
Regarding attracting huge corporate users, I think IBM's and Intel's contributions to Apache Harmony are a good sign of times to come.
I would not expect free runtimes to thrive in the same areas in which proprietary runtimes are available gratis as long as they are not obviously technically superior, but I'd expect them to thrive, prosper, and help create a market for runtime technology in those areas where current proprietary solutions are inadequate for various reasons (price, licensing, quality, availability, ...). That's the picture I see out there in the field, at least. The world is full of niches that proprietary runtime vendors can not fill, and may not have a business reason to do so.
cheers,
dalibor topic
Member since:
>>It would be like Microsoft offering .NET for linux - rendering Mono mostly useless.
Oh absolutely. Just like how Unix-on-Intel killed Linux.