Post by Doug BissettOn Fri, 26 Feb 2016 23:07:08 UTC, "Rick C. Hodgin"
Post by Rick C. HodginI got a reply from Arca Noae. They're pretty much dead set against the idea, and they know of no one currently working on an open source kernel.
Looks like the idea dies here.
Best regards,
Rick C. Hodgin
Hi Rick,
What I am seeing here appear to be a re-run of what happened many
years ago. Then a young guy was not satisfied with what OS he had to
work with so he decided to improve the kernel.
People said he couldn't do it, others said he would have copyright and
patent problems, yet more asked why go to all the effort.
He persisted and now the children of the kernel he wrote ate the basis
of the OS that run most of the worlds cloud and internet servers.
Don't give up. If you feel you have the ability run with it. You
will most probably find support along the way as well as people
willing to help.
When you get to a point where you people to test what you have on real
world equipment I will offer my companies servers and workstations for
those tests all of which are running on 64bit multi core processors.
We've had a few more emails since the first exchange. I've noted that
what I'd like to do here with OS/2 is what GNU did with Unix. They
began working on the GNU project and from 1984 through 1993/94, they
had written every proprietary piece of software as "free software,"
which means it was protected under the GNU General Public License,
which used the copyright system to enforce that the software remained
open and accessible to developers. Since they use the copyright system
in this way (to protect the receivers of the software, rather than the
authors), they often call it a copyleft license, it's like "copyright"
flipped over.
There's a movie on Netflix and YouTube called "Revolution OS" which
describes the differences between free software and open source. It's
well worth seeing as it outlines the vision Richard Stallman had when
he set out with the GNU project.
Revolution OS:
GNU is a full system, except the kernel. For the kernel they use the
Linux kernel mostly, though there are a few alternatives including their
own GNU home-grown kernel (called the "HURD"), but their kernel is
buggy and none of the kernels have as much hardware driver support as
Linux, so most use Linux.
I want to pursue one of two paths:
(1) Write a fully compatible open source OS/2 kernel that will run all of the existing OS/2 software, being a literal drop-in replacement, supporting all of its drivers, etc.
Or (2) might be the better path: To create a very OS/2-like system that is nearly 100% source code compatible with existing software, but does a few things differently under the hood, being a full ground-up creation from scratch.
I would like to support i386 at first, then extend to AMD64, and then finally to ARM 32-bit and ultimately ARM 64-bit.
I have plans to do this with my own operating system, but am thinking I would rather devote that effort to something with an existing user base, software, base, and be able to draw people who have an interest in the open source version coming to light. However, I am prepared to go it alone on my own design.
Here is my OS from 1996-2002 development. It was done after-hours in my evening and free time. It is a 32-bit kernel with limited driver support, but has every basic feature of a kernel except paging:
"Exodus" operating system (named as it was because I wanted it to be
a mass departure from the evil that was Microsoft and their proprietary
software model):
http://www.visual-freepro.org/videos/2014_02_13__exodus_debi_debugger.ogv
[If you can't view the Exodus video, use VLC: www.videolan.org]
Here's the source code for my current kernel:
https://github.com/RickCHodgin/libsf/tree/master/exodus
And my current software development plans follow a progression:
(1) Complete Visual FreePro, Jr. and Visual FreePro:
https://github.com/RickCHodgin/libsf/tree/master/source/vjr
http://www.visual-freepro.org/wiki/index.php/VXB
http://www.visual-freepro.org/wiki/index.php/VXB%2B%2B
This includes effort into (2):
(2) Complete my RDC (Rapid Development Compiler) framework, which
is a platform that enables people to build compilers, which I
will use for my CAlive programming language (an extension of
C heading toward C++ with some of my own creations, but is far
less cluttered than C++):
https://github.com/RickCHodgin/libsf/tree/master/books/rdc
(3) Continue development of my Exodus 32-bit operating system
(4) Port a version to ARM 32-bit
(5) Port a version to AMD64 64-bit
(6) Port a version to ARM 64-bit
Long-term goals. I've been working on this project since July, 2012. I've
had a couple developers come on board to help me, but that's been mostly
with Visual FreePro, Jr. only. The rest has been all me ... my blood, soul,
sweat, and tears. :-)
The desire to alter my plans to support the OS/2 kernel instead of my own
Exodus kernel came from a post on Facebook from an OS/2 user's group. I
was able to go back and re-install OS/2 Warp 4, which I haven't used since
the early 2000s. It was like an old friend, smooth and wonderful. It was
then that I had the thought to migrate my efforts to create an open source
version of OS/2.
We'll see what comes of it. My best-laid plans are that it will be May or
after in 2017 before I have the rest of my tools completed, but life things
may extend that out further, or bring them in closer. It's a lot of work,
and it's been a very solid bit of long distance running. I'm onto my second
wind now, but there's still a long way to go.
Best regards,
Rick C. Hodgin