OSDev.org

More important than the questions of "how can I support a Java VM as early as possible so I can write the rest of the OS in Java" is the question...

...why would you want to do that? What is the "business case"? What is the advantage of doing it this way?

If your answer starts with "I" (as in, "I know Java best", or "I like Java better"), rethink your decision. I would accept "because I can do it", but "because it can be done" is a border case because that probably continues as "...by people with enough know-how, like, if I keep asking enough questions on some forums...".

Don't get me wrong. I'm actually eager to hear the reasons for a Java OS, because I can't wrap my mind around it.

Solar: Just out of interest I believe you said what I wanted to say, but in a far more tactful way (and a way which signifies you weren't just at the end of 8 hours solid debugging in a project with about 4GB of code and object data...)

JamesM

It's a good question. For me it would be for the challange. But, talking to the bare metal and doing a C/ASM kernel is also a challange and why I came accross this forum and started OS coding in the first place

Enough Java for me.

http://en.wikipedia.org/wiki/JNode
http://www4.informatik.uni-erlangen.de/ ... index.html
http://trac.devlime.com/neptune/ ( 64bit os in D )

binutils wrote:http://en.wikipedia.org/wiki/JNode
http://www4.informatik.uni-erlangen.de/ ... index.html
http://trac.devlime.com/neptune/ ( 64bit os in D )

Thanks for the promo binutils (I'm the main developer for Neptune), but I should add a disclaimer: this project is in its early stages. I've gotten the hard bits of D working, but the kernel is far from finished.

As a Java programmer, I have to say I was very comfortable moving over to D for this project. You may want to consider it, although implementing a Java VM as your kernel would be an interesting way to develop an OS for an existing software base...