The following is an interview with Ola Jensen and Adam Chodorowski, two of the core developers of the Amiga Research Operating System (AROS), a project to develop a open source Amiga OS clone.What is the status of AROS today, and what are your aims and goals for the future?
How finished is AROS? Regarding functionality we are nearing AmigaOS3.1 state, however, there are still a lot of components lacking such as TCP/IP and sound support and other things people take for granted in an OS these days. Because of this, it’s not realistic to start using AROS as your main OS yet. However, developers are working on these issues and we are slowly getting there. Compatibility with Amiga OS is relatively good: most programs can be ported in a few hours with minor changes in the source code. I’m sorry to say at this time AROS has no integrated emulator, so you won’t be able to just run Amiga OS applications out of the box. However, we do have a port of UAE so its possible to run Amiga Applications in a virtual Amiga. As for aims and goals for the future, most of the developers have their own ideas for what AROS is to become. For some this is just a hobby and they find it interesting to work on a OS (they also get to learn a lot) while others want to continue where Commodore left it only to make it better and have it available on every platform one wishes to run it on and of course to have it free and open so that one will NEVER have to go through what Commodore had us doing for the last decade or so.
If I understand correctly, a lot of the current development is done in a hosted mode under Linux, but the goal is a completely independent OS. Will the hosted mode be further supported and developed into the future? It seems interesting for those of us in the Linux community that came to it as Amigans.
It depends on how many developers will be interested in the hosted flavor at that point. We won’t delete all sources for the hosted version when we get self-sufficient native version, but if there are no developers that are willing to maintain and work on the hosted version then it will stagnate and probably get broken. Time will tell.
A lot of people are interested in seeing memory protection (MP) support in an Amiga Compatible OS, will AROS support MP in the future?
This has been discussed many times by the developers, and the general consensus is that it is impossible to have traditional memory protection in AROS due to the API of the OS and the way message passing works. It may be possible to protect certain areas (executable code, kernel code) and impose restrictions (only priveledged tasks can talk to hardware), but full MP with the current API won’t happen. In the very long run (we’re talking *way* past AROS 1.0) it might be possible to change the API to allow MP, and relegate the old one to compatability libraries, but whether this will happen is to early to say now.
Interesting, what are the main obstacles in developing aros today, and will the development team branch into user apps at some point, like for instance will aros ship with its own browser in future, or a productivity suite?
The main obstacles are the lack of developers and sufficient time, which means the development pace is rather slow. There are still far to many things to finish in the core OS, and too few developers, to contemplate branching into user applications. Some larger projects we need to tackle before that can happen is a native (self-hosted) development environment (currently all development is done using a cross compiler under Linux or FreeBSD) and a TCP/IP stack, to name the biggest. Once these are finished and the OS is more or less complete, we think there will be an influx of users and developers,at which point development of applications probably will pickup. We definitely want a web browser and email client, at least.
For those reading that might be interested in aiding development, what have you to say?
AROS needs YOU! We need all the help we can get, so join the team you want to help!
Interview conducted by David Ford, a freelance writer. He is looking for work while also creating a chat empire over at P10Link.net.
Aros, is in my opinion a pretty neat Os, i’ve downloaded it, and used it for about 3wks now, the software that comes with it isnt great but it works, i’ve only had 6 Os crashes out of about over 50 boots, the menuing system is a bit fiddly to start with but easy to get used to, AROS is one of the best free Os projects i’ve seen on the net
AROS is one of the best free Os projects i’ve seen on the net
Agreed. It’s impressive. The UI is very FAST (although still slightly buggy), clean, and consistent.
put it on a bootable mini CD and played with it for a while. It was cool
I must also note that AROS is pretty slick. I burned the demo CD and gave it a whirl a while back. It does a lot- I figured it wouldn’t have been as full-featured, considering how it doesn’t really get talked about all that much. I mean, you don’t hear about people who run it, etc etc. Yet, it is quite complete in some areas. I can’t help but wonder why they’re not concentrating on things like TCP/IP/ to pull AROS into the realm of an OS folks could use for everyday things?
Keep up the good work, AROS team! You’ve got something special and cool on your hands, to be certain.
I can’t speak for the AROS developers, but I think TCP/IP gets in there when it’s convenient.
I like this project a lot. It’s a quiet point in the Amiga universe where people concentrate on development, rather than throwing too much mud at eachother. And it’s nice to see it picking up some speed in development.
Fabio Alemagna also works in the AWeb webbrowser team and when TCP/IP is done, I’m fairly sure there will be a port to AROS. Suddenly AROS becomes very usable. 🙂
Also keep an eye on:
http://www.chodorowski.com/aros/screenshots/
which gets updated regularly.
🙂
right now, we are not really focused on TCP/IP. It will come. A few people are working on it (out of the few developers), but don’t hold your breath. Let’s meet again in 3 years, and see if we will still be rocking. Current short term goals seem to be drivers, desktop/UI and devkit, and that’s fine by me, because these are the basic building blocks. Not to mention documentation.
Also yes, the site needs a status update, but we lack a status update writer
Very interesting.
I would love to be able to use AROS. I support it in the way that I can, but I am not a devloper. I am sorry to say, but without more devlopers I really don’t think it will get anywhere. It is amazing how far they have gotten with as few devlopers as they have. I wish them luck and hope some more devlopers join them. This could be really a really good OS, I think.
they clean up the UI somewhat. It looks out of date (fixable I’m sure) and from what I’ve seen, it seems like the UI for programs tend to be crowded with widgets. Also, the DOS thing is a turn-off.
It is a clone of an old OS. It is mean’t to be out of date. But I understand what you mean.
>>>>>>>>>>>>>>
Also, the DOS thing is a turn-off.
<<<<<<<<<<<<<<
You don’t like CLI?
Given the well established IPC interface with Amiga (and follows thru with Amiga Like OSs) …. typically seen in terms of the Arexx port and the heavy IPC use in the Hurd, the Hurd and AROS wil make for a nice combination with many user benefits.
Imagine extending the terms “dumb terminal” and “smart terminal” as analogies to ones personal interface to the computing power of the very large developer base of GNU. Where KDE and Gnome are considered “dumb terminals” to access in the user space of the Hurd. The “Smart Terminal” would be AROS interfacing with the Hurd thru IPC so as to access the GNU resources.
But AROS is also targeted for stand alone, and that too, is good, and further supports the “Smart Terminal” analogy.
Nothing like taking your smart terminal (OS + Personalized interface) with you (on something as simple and small as a business card CD or memory stick) that you can run it on what is the most popular and common hardware or PPC (both the Hurd and AROS are being ported to PPC), while also being able to interface it in an integrated way with the massive application functionality and multi-user security base of the FSF GNU works.
Many have said GNU/Linux is to complicated for the typical user (especially those who don’t need multi-user support/security complexity) and Many have said Amiga OS (and likewise AROS) is simple, easy, efficient to use. Nothing stopping the best of both finding the common IPC integration.
One needs to understand that there are three primary User Interface as explained at http://threeseas.net/vic/html/3uis.html
Perhaps zephc thinks that AmigaDOS is something related to MS-DOS. AmigaOS has had its own CLI for a while. DOS is a term which has floated around (Disk Operating System) before MS-DOS. It’s not like AmigaOS is based upon some gimpy single-tasking MS thang…
Just offering the chance for people to suggest by e-mail, articles or interviews theyd like for me to write, not neccessarily related to AROS.
Hi there,
I just wanted to know the status of the Pegasos port. In my opinion developping for x86 is not the right direction to go (there are already too many oses on x86). Now that we have some valid hardware (Amigaone or Pegasos) it would be great to have an os to go with it. It would be great.
Michal Schulz is already working on it, he has a Peg board and just got a Voodoo3 card so he can hopefully do more gfx stuff. So yes, PPC port is on sched … oh, nevermind.
However I for one would not afford a PPC board in the near future, but as I’m happy using Linux on x86, and don’t really expect to take over the world with AROS (but with Zune instead
I dont care 
Anyway, the choice x86/PPC is more determined by hardware and knowledge availability than any other reason, I guess. So send us even more PPC boards and PPC hardware hackers.
Sure, If any one want to send an AmigaONE or PegasOS to the AROS team, we would be happy to put more developer time in getting it up and running.
Right now very few of us can afford to buy a PPC machine, so we have to use the cheapest available hardware (Read: x86).
Our Thanks go to Genesi for providing a PegasOS board.
I believe zephc may be taking about the screen mode selector that appears at boot time…
If so, I quite agree with him. We will sort that out ASAP.
It’s a quiet point in the Amiga universe where people concentrate on development, rather than throwing too much mud at eachother.
A very good point!
Hi, I noticed AROS runs on PCs with 486 instructions or better. It’d be nice to see AROS continue to not require the most recent CPUs, as it’d be ideal to put on older systems.
I’ll bug {szuler} (hihi;-) to speed up with PPC port ;>
It is nothing like msdos, it actually support full multitasking!
on amiga.org forums? Maybe contact him to see if the offer is still in place..
Mr Buck did, and that is why we have actaully managed to get AROS booting on the PegasOS computer (and aparently Open Firmware based Macs?).
But Michal Shultz hasm quite a way to go before AROS can do anything useful on the PegasOS. These things take time.