Stratocaster

> Thanks to the hard work and dedication of a big team of people both inside of Sun, and in the Free Java community working on projects as diverse as GNU Classpath, GCJ, and IcedTea, Sun's open source Java initiative has reached a new milestone. Both Ubuntu 8.04LTS (Hardy Heron) and the upcoming Fedora 9 releases have an OpenJDK-based implementation of the JDK in their free software repositories.

A year and a half of work so that two Linux distros have JDK built-in, allowing an easier download/install. How many people do you think this affects, and to what degree? I'd guess it's maybe 10-100 developers who use those distros but have refused to use non-Free software, or weren't able to accomplish the download/install from Sun. are you picturing more than that? If so, can you quantify it?

It's nice progress, and I'm not trying to be a party pooper, I'm just wondering what I'm missing - how many orders of magnitude are my guesses off by.

Also, I followed the "said" link you provide, and I don't see where you specifically said that you wanted to "get Java into GNU/Linux distros".

> Now developers who are inventing the next YouTube or Twitter, the next amazing web application that quickly becomes something we all can't live without - and who wouldn't imagine using anything but a completely FOSS stack on which to build - can rely on Java.

Hmmm, you seem to be implying that those who care deeply about licenses are more inventive than the rest of us. How inventive have these people been, working on just duplicating Java, rather than working on something new and inventive?

> Now the platform itself will evolve that much faster, driven by the needs of the most sophisticated developers on earth. Now Java can go wherever GNU/Linux goes.

Please try to quantify the evolution, if you can. Sun seems to be staking a lot on GPLing Java, it would be nice to see more than antecdotal progress. I have a hard time imagining that there will be a spike in Java adoption or "evolution/innovation", but unless someone gathers hard data, we'll never know.

> There is still work to be done: the implementations out there are not 100% compatible, though they are 100% Free software.

There have always been implementations that are not 100% compatible but are 100% Free. I think what you mean to say is that they're actually very close to 100% compatible.

> But there are only a few bits that are still encumbered: the sound engine, some SNMP code, and a few other odds and ends. And these encumbrances are on the verge of being cleared, thanks to the diligence and passion of Free Java developers. And then there's governance to be established, infrastructure to be built, projects to be hacked, more distros and platforms to be ported to, more code to make Free, more innovation and excitement to be had.

I'm sure you're excited about all this, but I do wonder whether the number of Java developers who are also excited about it is in the 10^2 range, rather than the 10^7 range.

Posted by Andy Tripp on May 05, 2008 at 02:09 PM PDT #