Yesterday Jonathan Schwartz and Rich Green announced the availability of the Java technology under the GPL license.

From a J2ME developers point of view:
Most of the java code I wrote lately was for J2ME applications. The filing I have when I write J2ME code, is that I am not really writing java. I always have to be carefully not to write code that will have problems, because of a bug in the J2ME implementation of some mobile phones. A lot of mobile phones (and I mean A LOT!), have many strange bugs in their MIDP implementation. IMHO the J2ME applications are not cross-platform.
Some may say, “if those implementations are not fully tested and compatible to the specs, why Sun gave them the right to call this VM a JVM?”. Well, I think that Sun had to give ground. I think that Sun did that, because If they didn’t the J2ME would not be in almost any mobile phone in the market.
I hope that this will change. I hope that the companies that make mobile devices will at least be able to make better and really J2ME compatible JVMs. In the mobile phone market there are a lot of companies who make phones and most of them REALY like proprietary things. I hope that GPL license forbid them to make proprietary versions of J2ME.

From a J2SE and J2EE developers point of view:
A free implementation of Java means that Java could be preinstalled in a lot of operating systems, like the BSDs and also in many distributions of Linux. This means better adoption of Java. A lot of users and companies don’t use java applications, because they can’t install java on their system.
The open source java also means that the java is now free to evolve. I read in a comment, in the blog of Jonathan Schwartz, “I’ve avoided doing some cool things w/ Looking Glass since java wasn’t gpl’d. Well you’ve removed the obstacle. I guess it’s up to us now”.
At the other hand the GPL’d java just brings this two things to us a little bit earlier than project Harmony would. After less than a year, the apache project has it’s own implementation of most of the parts of a JDK. Maybe the existence of project Harmony pushed Sun to open source java earlier than the wanted. I think there is also the answer to “why GPL”. It’s obvious. Sun didn’t want another company to take their asset and make it their own. Someone asked Mr Schwartz and Mr Green, “why didn’t you just commit the source to Apache Harmony?”. The funny thing is that someone told that this was IBMs proposal.

I think Sun is a company that understands what FOSS means. The proof is that they have open-sourced a lot of their software (and hardware). They say they have committed the most to the free software community. I don’t know if this is true (IBM also gives away a lot of their IP and source) but the truth is that Sun gave to the free software community a lot of great software, like OpenOffice, OpenSolaris and now OpenJDK.

Sun Microsystems CEO Jonathan Schwartz said at the JavaOne conferance that they are planing to open source java. The only issue is how it will be open sources, under which license and which entity will continue the development of the language, to ensure that java language will be safe from forking.
It looks like the ASF project Harmony also did a good job here. Without releasing anything yet and while still in incubator, they gave a very strong message to Sun. Free java or we will make our own, fork to another way and leave you alone behind.

After about 6 months of (not exactly constant) work, my ObjectOriented TCP/IP stack is close to be finished. The Ethernet, ARP, IP, UDP and TCP protocols was already working since 2 months ago. The last thing to be implemented was the TCP Socket.

The first byte has already transfered over the Socket implementation. What’s next? Some more testing to be sure that we can send bigger amount of data over these sockets, the implementation of the TCPImplementation hook and finaly some polishing and performance improvements…

Why am I writing a TCP/IP stack? Let’s take it from the start, for those of you who don’t know about this TCP/IP stack. First of all, I am in the development team of JNode (http://www.jnode.org). Also I have one last obligation in my school, which is my student thesis. The above two, plus some good ideas about the design of a OO TCP/IP stack, gave me the idea to write a OO TCP/IP stack writen completely in Java.

p.s: I should apologise to my fellow developers in JNode about the delays :/ and to everyone else, about my bad english :p