[Mask of Destiny]

A while back it became clear to me that Palm OS was dying. There weren't any Palm OS 6 devices on the market despite the OS being available for a while and PalmSource was floundering around without any real coherent strategy to fix the problem. The imminent death of Palm OS became all the more certain when Access announced ALP. Being a fan of PalmOS and not being terribly fond of Unix-like systems, several months ago I started work on a Palm OS clone with the intention of releasing it as an open source project.

The goal of my little project is to develop a portable OS that is binary and source compatible with Palm OS 5 wherever possible while at the same time moving the system forward. So far I've implemented sizeable chunks of the Memory Manager and Data and Resource Manager. MemHandles are just pointers since between MMUs and better allocation algorithms there are more elegant solutions to the fragmentation problem (though currently I'm using a non-MMU aware and generally lame algorithm). Memory can be allocated in chunks larger than 64KB as can database records. Databases now support 32-bit indexes so you can have more than 65,536 records if you use the new 'extended' versions of the appropriate syscalls. Currently it's more akin to a library than an actual OS, but that should change very soon. I need to write some process management code and some 68K assembly to call the appropriate functions from the TRAP #15 exception and then it should be able to run simple applications (with no UI or user interaction since that's not done yet) on 68K devices.

Does this project interest anyone here either as a potential user when it's mature enough or as a developer wanting to help out? Any input on what you'd like to see in this little OS?

[mrpropre]

It sound very interesting...

good luck in your project!

[Omniyukai]

If it is stable then I will definitely give it a try! One thing I would be interested in is if it added multi tasking, which is the only thing I dislike about the Palm OS. If I could code I would be more then willing to help out, but currently all I can say is I am a potential user. It sounds like your project could go far and beyond what palm will ever do, then again I am a firm believer in open source.

[Mask of Destiny]

Multi-tasking will probably be one of the first "enhancements" added. Might even do that next after I get programs loading as it's not terribly difficult to do at this stage.

[Tarran]

So how do you plan on implementing this OS? Are you going to work on flashing the PDA's ROM, work on a loader that persists in memory after a reboot, or run it as an extra layer on top of Palm OS? Sorry for the annoying(?) questions but you're WAY beyond my level of coding experience here. So the most that I'll be able to offer is "good luck in your task".

P.S.
Since I just lurked on Tapland, I just wanted to also say that I've enjoyed your work for years. (Ever since I found ConvSCD and your Sega CD dev pages.)

[Mask of Destiny]

Tarran, on Jun 22 2006, 06:58 PM, said:

So how do you plan on implementing this OS?

Is this in reference to the questions that follow or a more general question about the implementation of the OS as a whole?

Quote

Are you going to work on flashing the PDA's ROM, work on a loader that persists in memory after a reboot, or run it as an extra layer on top of Palm OS?

The very first releases will probably be for the Sega CD and/or 68K Amiga for the novelty factor (since in the early stages it won't be good for much other than novelty anyway) and because I'm pretty familiar with the hardware. Those have reasonably standard boot processes so for those it won't be much of an issue.

Once I move on to realy PDA hardware I'll probably go with a bootloader approach initially. Basically you'd load needed file(s) onto an external card and run some kind of standard PalmOS app that would take control of the device and load the new OS. In such an instance, the OS would stay resident in memory until you did a hard reset. Eventually, I want to support a flash install, but I won't even start on that until it's reasonably stable and a viable replacement for PalmOS.

Quote

Sorry for the annoying(?) questions but you're WAY beyond my level of coding experience here.

They're not annoying at all. I enjoy answering technical questions that I haven't been asked 500 times before (you wouldn't believe the number of e-mails I get asking if its possible to boot standard Genesis ROMs off of a Sega CD disc, I really need to write an FAQ or something)

Quote

So the most that I'll be able to offer is "good luck in your task".

Once the foundation (memory management, process management, etc.) is put together there will be plenty of work that won't require super-human coding skills, just a working knowledge of C. I also enjoy teaching so if you want to help at some point, but aren't sure you're up to the task feel free to jump in and I'll give whatever help I can. Of course, if you don't want help that's fine too, the "good luck" is appreciated either way.

Quote

Since I just lurked on Tapland, I just wanted to also say that I've enjoyed your work for years. (Ever since I found ConvSCD and your Sega CD dev pages.)

Glad to hear you like my Sega CD stuff. Suggestions for new features are always welcome (though be warned my attention for my Sega CD projects comes and goes).

[Omniyukai]

If there were a viable replacement for the Palm OS that could multi task and did not require flashing to test, I would try it out in a second. Heck, if it worked just as well as palm OS, I would pay for it. I would also be willing to help any way I could, although I do not yet know any coding.

[theorem]

Neat!

I'm certainly interested in such a project and have, at the least, a working knowledge of c programming (I'm hardly an expert, though). I can't guarantee that I'll have time to work on this kind of project, but I'll keep an eye out.

If you were to put together a nice project-plan that laid out the necessary steps in relatively easy chunks (where possible, of course), it would probably help people like myself considerably in contributing. Could be great for anyone who wants to learn to code or, like myself, would like to improve his or her coding skills.

Anyway, neat idea, and hopefully I'll be able to lend a hand in making it happen.

Good luck!

[andyakadum]

What we need is a binary compatability layer, like wine.
So linux pdas can run native palm apps

This post has been edited by andyakadum: 24 June 2006 - 11:40 AM

[Mask of Destiny]

theorem, on Jun 24 2006, 03:46 AM, said:

I'm certainly interested in such a project and have, at the least, a working knowledge of c programming (I'm hardly an expert, though). I can't guarantee that I'll have time to work on this kind of project, but I'll keep an eye out.

Yay, a possible volunteer! Though even if you can't help out you're welcome aboard as a user of course.

Quote

If you were to put together a nice project-plan that laid out the necessary steps in relatively easy chunks (where possible, of course), it would probably help people like myself considerably in contributing. Could be great for anyone who wants to learn to code or, like myself, would like to improve his or her coding skills.

Not a bad idea. I'll probably start working up a proper project plan once I get some rudimentary support for form drawing and their associated events. Once I've got a preliminary plan put together I'll start fishing volunteers and if I get a handful of them I'll set it up as a proper open source project on sourceforge.

'andyakadum said:

What we need is a binary compatability layer, like wine.
So linux pdas can run native palm apps

I'm not personally fond of the idea of Linux on a PDA, but if someone wants to make a Linux hosted version of the OS I won't try to stop them. I don't have any problems with it being part of the main tree as long as it doesn't make a complete mess of the source for the native ports.

[theorem]

My personal view of Palm OS has always been that it has a great interface (for the most part), but that it feels a bit like an OS from the '80s or early '90s, otherwise. Much of what I would like to see done with this project is to modernize the OS, adding features like true multitasking and, on the GUI side, across-the-board font anti-aliasing (I think this is already in some of the newer Palm devices?). Obviously, we don't want to get ahead of ourselves, though.

Quote

Not a bad idea. I'll probably start working up a proper project plan once I get some rudimentary support for form drawing and their associated events. Once I've got a preliminary plan put together I'll start fishing volunteers and if I get a handful of them I'll set it up as a proper open source project on sourceforge.

Making this a Sourceforge project seems like a good idea.

Quote

What we need is a binary compatability layer, like wine.
So linux pdas can run native palm apps

Of course, thats only useful if you have a PDA that runs Linux; that's a no-go for a Zodiac user like myself, as far as I'm aware. I use Linux on the desktop and wouldn't mind it on my PDA, but I like the idea of creating an extended functionality Palm OS clone even more.

This post has been edited by theorem: 24 June 2006 - 08:45 PM

[theorem]

So I recently started a blog (very recently) and decided I might try to draw a little attention to this project. The traffic on my blog at the moment is about 0, so it probably won't make much of a difference, but we'll see.

See the blog post here.

[Mask of Destiny]

It occurs to me that there are some ways non-programmers can help out even in these early stages.

First off, this project needs a proper name. I've been referring to it internally as "Project Flatfoot" but that doesn't seem to be a very marketable name and I suspect someting similar is already being used for some type of computer project (though I could be wrong on both accounts). It also will need a logo. If someone wants to offer some suggestions they would be welcome.

Also, I don't really have the faintest idea how to setup a project on sourceforge and at the moment I'd rather focus on coding than figuring it out. It would be a big help if someone could set one up. The sooner the sourceforge site is up, the sooner source code will be released.

I'd also like some input on what license to release this under. I'm not personally a big fan of the GPL, but it would make it easier to grab components from other projects since the license is so popular.

[Newspaz]

Hey that's really interesting. I'm not a programmer myself but I'd like to help out too. I have done some webdesign in the past so let me know if you'd like my help.

For a name I'd go for another plant/tree name. Something like FicusOS could be pretty funny

This post has been edited by Newspaz: 26 June 2006 - 03:52 PM

[tbuskey]

POSE or Xcopilot?

Are you looking at these at all?

Some of the posts mention wanting a palm emulator running on Linux. Well, POSE and Xcopilot do it. And on Mac and Windows.

I use them all the time to access a read only copy of all my palm apps and data.

I'd think it'd be possible to create your OS in the rom file POSE loads for testing purposes.

I'm not sure if POSE does the ARM and PalmOS 5 but it certainly does 68k and PalmOS 4, 3, 2. It's base on the UAE 68k emulator.