Index

Project Status

Project Policy

Project Directions

Development Tools

Research & Documentation

Project Status

What is the status of this project? How much has been done already?

The project was started late November 2001. After experiencing several setbacks, the project was restarted in its current incarnation in January 2004.

I know a couple of people who would be interested...

Bring 'em on, so they can make up their mind about Pro-POS. You should note, however, that we no longer do public discussion of our goals, concepts, and designs. You can contribute to the Wiki (you need not ask for explicit permission, it's world-writeable!) or ContactUs by e-mail.

I want to help, what do you want me to do?

The subprojects PDCLib and/or FirstStep would welcome your code submissions. If you feel you can contribute in a significant way other than coding, feel free to ContactUs.

POS Design is done by the CoreTeam only (not open for the public), but if you feel you can provide valuable contribution, feel free to ContactUs. You might end up in the CoreTeam. :-)

"Foundation" and "Perfection" are dormant for now.

Project Policy

So this is a project to resurrect AmigaOS?

No. Pro-POS is not about resurrecting anything, except perhaps the fun in computing. AmigaOS and Amiga-like designs keeps popping up in discussions and examples because several project members used Amiga computers for many years, and AmigaOS had certain characteristics unlike most other operating systems.

Why reinvent the wheel when you could simply use the open-source Linux and create your own "perfect" Linux distribution?

Well, as explained in our MissionStatement, Linux and other Unix spinoffs are based on Unix, which we consider to be inherently so full of legacy, demanding so much up-front knowledge, that it could be considered the antithesis of Pro-POS.

But, perfection is impossible to achieve, isn't it?

That is correct. We are aiming at something that is per definition impossible to achieve and we are also very aware of it. Why? Well, William Ellery Channing once said:

"Fix your eyes on perfection and you make almost everything speed towards it."

So, we are not saying that our OS will be "perfect", just that we are aiming for it. That is why we use "Perfection" as a moniker for our OS: it is a constant reminder of what we are aiming for. The end-user version will have another name to remind us that the goal has not really been achieved so that we will work even harder towards that goal for the next version.

Is your OS going to be POSIX compliant?

By design, no. Be aware that POSIX is a standard for Unix based operating systems.

We are open for solutions such as AmigaOS' IXEmul.library or the Cygwin.dll - providing a POSIX layer on top of the operating system for ease of portability - developed by third parties. But since we are not sure too strong an embrace of existing POSIX software will really benefit our project, we will not officially embrace such a solution.

Please be aware that our main goal is to create a better operating system. If we were happy with POSIX, we'd be happy with Linux.

Project Directions

Why not support Java and/or other programming languages than C++?

Compilers and interpreters are not exactly components of an OS, although we would really like to see them added at some later point of time. Java is definitely a must, and is targeted for "Perfection" v2.0 (i.e., it will not be part of the first public release since we need something that the Java VM can run on). Other languages are likely to follow.

"Perfection" will come with a scripting / macro / application interfacing language included, in the tradition of AmigaOS ARexx.

I have a lot of ideas about an architecture, where is a good place to put them?

Read up on POS Design. If you think your ideas fit ours, or are ready to advocate a major change in design because you believe your ideas to be better than ours, ContactUs.

Why don't you use a virtual machine, it would be a much safer / more portable system?

A virtual machine is never faster, never takes less memory, and never has less bugs than a classic "process engine" since the VM itself has to run on such a process engine.

We would prefer to leave the decision pro/contra virtual machine - and what virtual machine - in the hands of the user. The system will enable a system administrator to install a VM on top of "Perfection" and disallow any other applications to be installed and run, effectively making "Perfection" into a VM-only system. Other people might have other demands.

Development Tools

I am using the <something> IDE for a while now and it is useful, stable, and portable. Should we make it the "official" standard for this project?

If you have an IDE you feel comfortable with, use it. Since there will be other people using kdevelop, Visual Studio, EMacs, or whatever, we will not make a specific environment "official", much less a "requirement". The "official" compiler is gcc, and the assembler / linker is the binutils package, until the day comes when we will bootstrap a native development environment. What you use as an IDE is up to you.

Why don't you provide a dev package for download, e.g. GCC / Bochs / ...?

Tastes differ. The "barebones" GNU toolset is either included in Linux, or comes with an installer in Windows (Cygwin). We discussed the option of providing a "package" of our own, but decided against it: It would be large, downloads would quickly blow the traffic limits on pro-pos.org. It would have to be updated / re-downloaded whenever an individual part of the package gets updated. It would require someone to periodically check for changes, making and uploading the package. Thus, we merely provide links to such tools, and leave it to the individual contributer what to use.

Research & Documentation

What should go into an evaluation of XYZ, in order to suggest it to you?

A brief introduction to what it is and what it is for. Any issues with installation or documentation? Is it easy to use or do you need to be a guru? Can it be made to work, theoretically, in our cross-platform environment, or is it tied to a specific OS / operating environment? Are there licensing issues? Can we put it to good use? And so forth.

Basically a good evaluation should provide us with enough information to make an educated decision, without having to plough through one or more websites full of documentation. Ending your otherwise objective evaluation with a personal recommendation is welcome.

The FAQ didn't answer my question, who should I turn to and how?

Just ContactUs. We don't bite.