-----Original Message-----
From: Ryan Heise <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Sunday, July 18, 1999 11:01 PM
Subject: Re: [JOS-Kernel] Construction of a new Java Operating System
>Cliff Baeseman wrote:
>
>> >Cool. How are you handling the graphics side? What sort of changes have
>> >you made to the kernel? etc.?
>>
>> Grapics is implemented by a bare minimum X implementation running all
native
>> code.
>
>Are you running a native or Java window manager?
Native for now! I plan on changing that...
>
>> >JOS consists of a small amount of native code at the core and a large
>> >chunk of Java code providing most operating system facilities. These
>> >Java libraries could run on top of different kernels, and in fact, the
>> >JOS project has at least two kernels in development. Your kernel is yet
>> >another kernel that could support the JOS libraries. You are free to use
>> >them.
>>
>> It would be nice to do this but my kernel is all native linux. I run the
>> linux drivers none of my own.
>> There is little benifit to running java drivers it just kills
performance.
>
>Although I would prefer Java drivers, it matters little. An operating
>system consists of much more than the kernel and its drivers. You need
>to provide libraries for multi-user support, system security, processes,
>windowing environment, control panels, etc. Since all of this code will
>be used extensively in Java applications, it makes sense for them to be
>written in Java. Besides Java is a better language than C. Libraries
>written in Java will be more robust, more flexible (components can be
>swapped in/out via class loading), and they will run on top of any
>architecture that provides an underlying Java interpreter.
>
I write code every day in ASM C C++ Pascal Java and VB. I never restrict
myself by excluding a languge that can just plain get the job done.
I agree it would be nice if it where all java but if it takes a couple other
languages to get there and it will than so be it.
>> hmm. It seems most of the JOS project developers are not interested in my
>> implementation. The JOS team lacks the one thing it needs to be
successful.
>> "Development Velocity!"
>>
>> They have been at it for almost two years and have nothing running, this
>> will not attract developers. Open Source developers like progress.
>
>I think a kernel that works "now" would be a good addition to JOS. It
>would be great if we could cooperate rather than exclude eachother.
>Also, I think you will need some sort of Java library on top of the
>kernel that provides operating system facilities. Rather than duplicate
>our efforts, you could more easily help us.
>
>BTW, it is false that we have nothing running. The JJOS team have a
>kernel that boots and can execute Java bytecodes.
>
I would have to see the code! I have not been able to see anything on the
site yet.
>> Not sure not to many people are interested I guess. I will gear up the
>> project in the very near future if
>> there is enough interest. Unfortunately I will do it seperate from the
JOS
>> project. It lacks all of the things necessary for success. "Developers"
>
That could be but maybe not. I just drove another Open Source project into
the ground. All it took was a little development speed and I constantly are
in communications with all the developers.
In order for me to jump onto the JOS project I would have to see one hell of
a lot more interest than I can see in it now.
I havn't gotten any mail on this list for days, my other projects list sees
at least 50 to 60 a day "Now that is communicating"
>If you attack and exclude JOS, I think you've just blocked out potential
>contributors.
>
>--
>Ryan Heise
>
>http://www.progsoc.uts.edu.au/~rheise/
>
>_______________________________________________
>Kernel maillist - [EMAIL PROTECTED]
>http://jos.org/mailman/listinfo/kernel
>
>
_______________________________________________
Kernel maillist - [EMAIL PROTECTED]
http://jos.org/mailman/listinfo/kernel
