Amiga.org
Operating System Specific Discussions => AROS Research Operating System => Topic started by: Heiroglyph on November 03, 2010, 04:17:02 AM
-
It's FrankenROM (part Aros, part OS3.0) but it's a huge step forward!
http://www.evillabs.net/wiki/index.php/AROS_m68k-amiga
If you haven't donated to the bounty yet, give this guy a boost!
http://www.power2people.org/projects/profile/5
-
I still don't understand why you guys want this. AROS was originally designed to be an open sorced re-implementation of WB3.1.
AROS still hasn't been polished enough to br released as a 1.0 package.
If ported to a 68K its likely to be slower and buggier than what you currently use.
Are copyright concerns that significant that it makes sense to expend so much energy just to realize what you already have?
-
Because once we have an open source kickstart replacement we're no longer tied to several companies that have abandoned the 68k platform.
We can fix bugs in the ROM and add new features as needed. 060 libraries, CDRom boot, etc.
Most of the 3.x OS problems are already solved on Aros, a video driver SDK for example.
Mini-Mig, NatAmi, WinUAE and others can create and sell compatible systems without legal problems.
This is nothing but a huge benefit to the community.
-
AROS still hasn't been polished enough to br released as a 1.0 package.
Says who? I've certainly seen worse - it's for sure a heck lot better shape than Amiga OS 1.0.
If ported to a 68K its likely to be slower and buggier than what you currently use.
I do not see why you think this - CBM developers were magicians? You think gcc cannot compile efficient m68k binaries?
At least with AROS bugs can be more easily be fixed, efficiency be improved, without all the hassle needed for tweaking original AmigaOS.
Are copyright concerns that significant that it makes sense to expend so much energy just to realize what you already have?
Consider this - one of the big contributers to the m68k kickstart bounty (http://www.power2people.org/projects/profile/5) is Cloanto. And people are ofcourse allowed to spend energy on whatever they feel like, the people mostly involved in the m68k port are "new" in AROS context, it's not like this is pulling lots of resources from the rest of the project.
And lastly, I can't help thinking that someone will eventually get AROS running natively on Atari ST and Falcon as well, which would be quite awesome :laughing:
-
I still don't understand why you guys want this. AROS was originally designed to be an open sorced re-implementation of WB3.1.
AROS still hasn't been polished enough to br released as a 1.0 package.
If ported to a 68K its likely to be slower and buggier than what you currently use.
Are copyright concerns that significant that it makes sense to expend so much energy just to realize what you already have?
Firstly, the people involved in 68k AROS/kickstart replacement werent working on AROS prior to this bounty, so no resources are lost. Also having AROS run on a real 68k Amiga will help greatly in improving api compatibility for other architectures as well as helping with bug fixing. There's also things like minimig, natami, and variations on UAE that benefit. Additionally it can help with UAE intergration in non 68k targets without the need for commercial ROM images. There's a lot of benefits it brings to both AROS and Amiga OS. I don't really understand how people think otherwise.
Oh, and cool stuff in regards to the original post/first screenshot :)
-
AROS still hasn't been polished enough to br released as a 1.0 package. If ported to a 68K its likely to be slower and buggier than what you currently use.
The other points people have made being well covered, I'd just like to know why you would wait until a 1.0 release (typically a major milestone in open-source projects, and long past the point of basic usability) before providing support for the nominal target platform?
And slower, maybe, but buggier? If anything, the Amiga is likely to be less buggy, as there's a more intuitive architecture (both CPU and hardware) and less extreme variations in setup than on an x86 or even PPC machine.
You think gcc cannot compile efficient m68k binaries?
GCC can't compile efficient anything binaries; what it can do is compile consistently usable binaries on all platforms ;)
Anyway, spiffariffic! I'm not an AROS partisan, but anything that gets 68k Amigas further away from the giant rights tangle-up is okay by me :) Could use a different splash screen, though; I never did care much for the KS2-3 one anyway.
-
GCC can't compile efficient anything binaries; what it can do is compile consistently usable binaries on all platforms ;)
Very true, gcc's claim to fame is just availability.
There's nothing to stop someone from hand optimizing the code where needed though.
It's still far easier than writing the whole Kickstart in assembly, especially when you have the source code available as a reference.
-
Could use a different splash screen, though; I never did care much for the KS2-3 one anyway.
I vote for a bouncing Boing Ball, or the Juggler.
-
No more silly boing balls please - I very much perfer the one and only true Amiga logo - the checkmark - which btw is free for anyone to use, since none of the Amiga copyright/IP holders cared about it.
-
I have a very bad feeling that it will be a sexy animal.
I can appreciate the artwork, but it's not what I want for a boot logo.
-
Use the checkmark for AROS/m68k I'd say (again).
In time, who knows - maybe one can put out AROS based Amiga OS 3.10 and 3.11 (for workgroups) :lol:
-
Great screenshot ! AROS is "home" ;-P
-
Nice !
Mini-Mig, NatAmi, WinUAE and others can create and sell compatible systems without legal problems.
That's the more important I guess.
Not to mention replacing Amiga modules by AROS modules will help to track and fix AROS bugs, that will benefit for the main AROS "distribution" too.
If I understood correctly FrankenROM is made of an Amiga kickstart with some libs replaced with AROS ones ? Currently only AROS exec, right ?
What's the next step ? :)
-
I don't know why you would worry about 'a sexy animal'. It's not like someone is going to put graphic genitalia on it. The Kitty pictures that have been used for AROS are no more graphic than what you would find on a Saturday morning cartoon. It is far less sexual than the stuff you would find in childrens movies like Shrek 2 or Happy Feet.
-
Originally Posted by commodorejohn
Could use a different splash screen, though; I never did care much for the KS2-3 one anyway.
I vote for a bouncing Boing Ball, or the Juggler.
Guys!!! It's AROS, it's open source... you can have whatever boot screen you like :)
I've waited 10 years to see this. So I'm pretty pleased.
Toni Willen (of UAE fame) has joined the project to help with hardware support, so soon we will have a totally free, UAE... no illegal/grey ROMs needed... Also AROS can be tailored to fit UAE better (more direct hardware control)... only good things can come from this.
-
Could use a different splash screen, though; I never did care much for the KS2-3 one anyway.
Well, don't worry that isn't AROS spash screen, but KS 3.0.
As you might remember this current FrankenROM is actually KS 3.0 with some modules replaced with AROS ones ("This is just the AROS Exec, with everything else from AmigaOS KickStart 3.0."). "strap" module (which produces the "please insert disk in the drive" animation) is obviously still KS 3.0 one. Once strap will be replaced the KS 3.0 spash animation will be gone.
This is the very first step on a long road. It gets far more interesting once you start replacing more KS ROM parts. It gets a bit messy once you run into the gazillion compatibility issues in AROS replacement modules. However, now that you can easily switch between the original modules and the AROS ones it will be trivial to use bisect method to track bugs.
-
Also having AROS run on a real 68k Amiga will help greatly in improving api compatibility for other architectures as well as helping with bug fixing. There's also things like minimig, natami, and variations on UAE that benefit.
This is so important it's worth repeating.
Being able to run real Amiga programs as test cases, even when we don't have the source, will give us pretty much all of Aminet and a bunch of old commercial apps as a resource to draw on to track down bugs.
-
The topic is a bit misleading. The kickstart boot screen is produced by Commodore graphics.library, utility.library, timer.device, ciaa and ciab resources, trackdisk.device, carddisk.device, card.resource and various other components which really are not AROS yet.
I think it's time to get excited once AROS does all of that.
-
I do not see why you think this - CBM developers were magicians? You think gcc cannot compile efficient m68k binaries?
To be fair, there *are* parts of AROS that are definitively not optimized for speed and low memory usage, so we will need to work on that once it's up and running.
As an example, the console.device in real AmigaOS has optimizations that only scrolls a single bit plane if the visible text uses only pen 1. AROS doesn't do any bitplane masking yet because it doesn't make any sense on the chunky modes used on PC hardware.
Though many optimizations like that are easy to make once AROS runs on Amiga hardware (in this case, just needs to do a SetWrMask() when we know its safe).
-
The topic is a bit misleading. The kickstart boot screen is produced by Commodore graphics.library, utility.library, timer.device, ciaa and ciab resources, trackdisk.device, carddisk.device, card.resource and various other components which really are not AROS yet.
I think it's time to get excited once AROS does all of that.
Thanks for the explanation, it's nice to learn about the boot process.
The road is still long to have a true AROS replacement kickstart indeed. But progress is progress... Any idea how long it took MorphOS to get from such level to the level where it completely overtook Commodore's OS ?
-
You think gcc cannot compile efficient m68k binaries?
I don't think it can't, I know it can't. At least it never uses scaled indexed addressing where it would be faster than using a scratch register to calculate the offset and adding it to the base pointer. This was certainly the case in 2.95 and as far as I can tell all versions of 3. Might be fixed in 4, but I wouldn't bet on it.
It also has a propensity (in C++ mode) to add expensive exception context state save operations where they'd never actually be caught in unwinding, nor does it bother to check properly whether or not it actually needs to save off the entire FPU register set when doing so. Older versions emitted multiple copies of virtual function tables for the same abstract classes which were never link optimised away. I could go on...
Luckily, nothing I haven't found workarounds for ;)
-
i've donated (only) 10$ in this moment! thank you men!
-
Any idea how long it took MorphOS to get from such level to the level where it completely overtook Commodore's OS ?
Well personally I think we overtook Commodore's AmigaOS quite early on, even though we still depended on some rarely used KS ROM parts for quite some time.
I was there doing the actual work but I've long forgotten when it was the point we could drop the KS ROM altogether. It probably was around MorphOS 0.8 or 0.9 (late 2001, early 2002) or so. So it took several years.
However, AROS case is slightly different: they already have most of the modules and they just need to bugfix and adjust them. It could be just couple of months if everthing plays out well. However, if there are major gaps to fill it can take well over a year in my estimation. Also, it depends much on how buggy version you'll consider acceptable. If you don't mind things being glitchy and some modules missing altogether then it should be possible in couple of months.
-
This is good news on a long needed replacement that everyone else has already mentioned. keep up the good work.
-
I still don't understand why you guys want this. AROS was originally designed to be an open sorced re-implementation of WB3.1.
AROS still hasn't been polished enough to br released as a 1.0 package.
If ported to a 68K its likely to be slower and buggier than what you currently use.
Are copyright concerns that significant that it makes sense to expend so much energy just to realize what you already have?
Wow, what a negative first reply, the person doing the work must feel so great after reading this.
Anyway, why do we want a 68k independent Amiga OS? Minimig. FPGAArcade. Natami. New ROMs without a dozen setpatches. Boot from other devices if you have the hardware. Proper optimised ROMs for 68060 or Natami's 68050.
-
Well, I for one am happy to hear that there'll be a KS replacement for m68k AROS.
-
I am just happy with the constant progress :-)
-
http://www.evillabs.net/wiki/index.php/AROS_m68k-amiga
If you haven't donated to the bounty yet, give this guy a boost!
http://www.power2people.org/projects/profile/5
I just donated eleven bucks, but he may need it for hosting fees, looks like evillabs.net is AmigaorgDotted right now.
-
It's FrankenROM (part Aros, part OS3.0) but it's a huge step forward!
Cool. Though I question the need to have the disk animation and Amiga checkmark exactly as they were, as it seems those images could suffer from copyright issues. Are there plans to have some new imagery to replace them once things progress and there is less and less and then none of Amiga Inc's code in there?
-
Cool. Though I question the need to have the disk animation and Amiga checkmark exactly as they were, as it seems those images could suffer from copyright issues. Are there plans to have some new imagery to replace them once things progress and there is less and less and then none of Amiga Inc's code in there?
FrankenROM is the term Jason coined to explain that it's part Aros code and part KS3.0.
I'm sure those are parts of the original kickstart.
All original parts will be replaced as individual replacements are stable enough.
It's just easier to test when you can test one section at a time for compatibility rather than trying to do it all at once, not knowing which part broke the whole.
-
Though I question the need to have the disk animation and Amiga checkmark exactly as they were, as it seems those images could suffer from copyright issues.
Again, the reason you see it because this is not AROS. It's original Commodore KS ROM.
The only part AROS is the exec. That's a tiny fraction of the KS ROM.
Are there plans to have some new imagery to replace them once things progress and there is less and less and then none of Amiga Inc's code in there?
They're not images as such, but rather instructions on how to draw the vector gfx involved. Anything else would have taken way too much ROM storage.
It will get replaced as soon as the strap module is replaced.
-
No more silly boing balls please - I very much perfer the one and only true Amiga logo - the checkmark - which btw is free for anyone to use, since none of the Amiga copyright/IP holders cared about it.
I'm with this. I like the checkmark logo, but something a little more elegant than the KS2-3 version would be good; it looks a bit too much like a mid-90s "I'm just learning Turbo Pascal!" shareware thing. Maybe something sort of like the old Windows 95 BOOT.SYS (http://www.sevenforums.com/attachments/customization/38392-how-would-i-change-startup-screen-windows-95.jpg) with some color-cycling? (Not with the clouds, obviously, but a largish checkmark against a simple abstract backdrop?)
'Course, if we do want a sexy animal, I'd be happy to assist ;D
(P.S. yeah, I know that's still the old Kickstart doing the splash-screen gruntwork; I'm just tossing out ideas for when it does get to the point that a replacement is in order. Heck, if someone would point me to some documentation on how that section works, I might take a stab at a new screen myself.)
-
(P.S. yeah, I know that's still the old Kickstart doing the splash-screen gruntwork; I'm just tossing out ideas for when it does get to the point that a replacement is in order. Heck, if someone would point me to some documentation on how that section works, I might take a stab at a new screen myself.)
Well, you have a very tight space constraint. It can be asked if AROS KS ROM even fits 512KB as it is now. Images are definitely out, so anything you do must be mathematically generated.
I don't think that AROS even implements the "please insert disk" animation of any sort as of now. So you have free hands on this one. As long as your effects fits into say 7KB it should be fine. Some free hints: You really want to look into vector graphics and perhaps procedural generation (https://secure.wikimedia.org/wikipedia/en/wiki/Procedural_generation).
-
I perosnally dont think we need any sorrt off animation.
Cant say I have seen that screen come up in a long time.
Who is going to be booting off floppy ?
-
Oh definitely vector/procedural type graphics; no question there. Would this be okay to make Amiga-specific, or would they be looking for something cross-platform? (i.e., is it safe to go stuffing registers, or should it rely on graphics.library?) What is the code structure like? Is it a function that gets called periodically, or what?
-
It's way too early to say, but if at all possible, I'd assume using built in routines.
We don't need another unmanageable asm only, hard coded to the hardware mess.
For all you know, the machine it runs on may not even have the custom chipset.
Once it does work though, I'm sure it will fork all over the place.
There will surely be 1.3-type ROMS for old games, 3.1-type, 060 specific, ROMs for super-expanded Amigas that boot directly to RTG cards with USB input, etc.
Aros will really open up a lot of options without needing OS3.1, Picasso96 or CybergraphX once it's finally complete.
-
... Heck, if someone would point me to some documentation on how that section works, ...
I suspect there is none, so maybe you could take a stab at generating some :)
I'd be interested just to see how it works. You would have to disassemble sections of KS and comment the code, for starters. Would be interesting if you could re-factor it into a callable subroutine that could be run anytime you wanted to see the 'hand'.
-
So about the original post. It did express my honest buffuddlement at your pirsuit of this endeavor.
But, as someone with no current 68K based hardware I am hardly the target for this project.
Further, if you guys want this, support this, and see benefit in it, far be it from me to discourage a project that advances the Amiga platform.
-
I suspect there is none, so maybe you could take a stab at generating some :)
I'd be interested just to see how it works. You would have to disassemble sections of KS and comment the code, for starters. Would be interesting if you could re-factor it into a callable subroutine that could be run anytime you wanted to see the 'hand'.
Well I've actually done it once. I even wrote my own gfx replacement using the routines. Unfortunately I have no idea where I might have the source code...
-
So basically once this is completed, AROS boots as the native OS on Amiga HW, correct? Or am I missing something.
-
Hmm. Well, I'm just going to play around with designs and graphics.library coding; if you ever find the information on integrating it into Kickstart, drop me a line :)
-
So basically once this is completed, AROS boots as the native OS on Amiga HW, correct? Or am I missing something.
The long term goal is to have a ROM that can replace the 3.1 ROM and still run AmigsOS and games or use AROS as the OS.
If using AROS as the OS, you could access more advanced features like the AROS CybergraphX replacement, hardware independent device layer, USB, etc.
-
the coolest project since years, wonder why this was impossible to do earlier. will have impact at much broader front than aros or 68k alone i think.
-
the coolest project since years, wonder why this was impossible to do earlier. will have impact at much broader front than aros or 68k alone i think.
Because there was no person interested and with enough knowledge. Now such person joined the team and the projected kickstarted ;)
-
The long term goal is to have a ROM that can replace the 3.1 ROM and still run AmigsOS and games or use AROS as the OS.
If using AROS as the OS, you could access more advanced features like the AROS CybergraphX replacement, hardware independent device layer, USB, etc.
Would AROS be totally contained on the ROM, or would the ROM act as a bootstrap for AROS?
-
Well, if it's going to be compatible with normal 68k Amiga software, it's got to at least contain the reimplemented Kickstart libraries. Like the old Macs with the Toolbox ROM, that's a significant portion of the OS (kernel and a number of driver, graphics, and windowing libraries.) I don't know enough about AROS or about the plans for this to say what portion of AROS that is, though.
-
I have just added a very tiny bit to the bounty, it is really worthwile supporting this kind of endeavour.
Add your grain of sand to this project :) http://www.power2people.org/projects/profile/5
-
Well, if it's going to be compatible with normal 68k Amiga software, it's got to at least contain the reimplemented Kickstart libraries. Like the old Macs with the Toolbox ROM, that's a significant portion of the OS (kernel and a number of driver, graphics, and windowing libraries.) I don't know enough about AROS or about the plans for this to say what portion of AROS that is, though.
AROS has not only rebuilt the OS, but also most of the things we have to add on for day to day use, such as RTG software and MUI plus all new things unique to AROS.
Not all of it will fit into ROM, it's just far too much.
I can't see how they will fit all the things from just a 3.1 level kickstart without using tricks that the 4000T used, like moving Workbench.library to disk. That compiled C code will be bigger in almost every case.
For my uses I don't care if it just bootstraps to a disk based kickstart personally, I only use floppies as a last resort anyway.
-
Well, you have a very tight space constraint. It can be asked if AROS KS ROM even fits 512KB as it is now. Images are definitely out, so anything you do must be mathematically generated.
I don't think that AROS even implements the "please insert disk" animation of any sort as of now. So you have free hands on this one. As long as your effects fits into say 7KB it should be fine. Some free hints: You really want to look into vector graphics and perhaps procedural generation (https://secure.wikimedia.org/wikipedia/en/wiki/Procedural_generation).
I would like something like Loonies' Luminagia or Ikadalawampu as boot screen ;P
-
For my uses I don't care if it just bootstraps to a disk based kickstart personally, I only use floppies as a last resort anyway.
I can understand that. Floppy capacity is looking really limited. I'm surprised some PC BIOS updates still require bootable floppies. Booting from a cd gives you so much extra capacity.
-
Plus if you have fast ram, most people want the ROM there anyway for faster access.
-
If this is open source software, why do they need to make up their own license?:
http://aros.sourceforge.net/license.html
( based on MPL mozilla public license? )
Why don't they just distribute with a real license (GNU / General Public License)?
http://www.gnu.org/licenses/quick-guide-gplv3.html
-
Plus if you have fast ram, most people want the ROM there anyway for faster access.
True enough, and those of us who want to boot floppy-based games can always use a ROM switcher.
Why don't they just distribute with a real license (GNU / General Public License)? (http://www.gnu.org/licenses/quick-guide-gplv3.html)
Maybe they don't want to get involved with GNU organizational politics?
-
Btw, if it's open source why do I need an account to have read-access to the sources ?
From the website:
The AROS repository is running on a password protected SVN server, which means that you need to apply for access to it to be able to collaborate in the development. At the request of Amiga Inc., anonymous read-only access to the repository has been disabled.
Seems like some parts can be checked out without an account, but I followed the instructions here http://repo.or.cz/w/AROS.git/search/b3650ddf5bd56e88b6d8277ed7a55ccd41589434?s=jmcmullan;st=author and couldn't do it because it asked for a login/pass...
-
Btw, if it's open source why do I need an account to have read-access to the sources ?
From the website:
Seems like some parts can be checked out without an account, but I followed the instructions here http://repo.or.cz/w/AROS.git/search/b3650ddf5bd56e88b6d8277ed7a55ccd41589434?s=jmcmullan;st=author and couldn't do it because it asked for a login/pass...
Why not just download the sources then ?
-
If this is open source software, why do they need to make up their own license?:
http://aros.sourceforge.net/license.html
( based on MPL mozilla public license? )
Why don't they just distribute with a real license (GNU / General Public License)?
http://www.gnu.org/licenses/quick-guide-gplv3.html
Err,... maybe because different license conditions are wanted ? You do realise there's many different licenses for open source projects dont you ? Not everything is GPL.
-
Btw, if it's open source why do I need an account to have read-access to the sources ?
From the website:
Seems like some parts can be checked out without an account, but I followed the instructions here http://repo.or.cz/w/AROS.git/search/b3650ddf5bd56e88b6d8277ed7a55ccd41589434?s=jmcmullan;st=author and couldn't do it because it asked for a login/pass...
I remember a thread on the mailing list a couple of years back, here we decided tat A. Inc were no longer a threat and access should be open again... But I guess no one bothered to do that... Probably because the people who want/need the source are developers who need write access too... The souce snapshot is available for down load from the site, so anyone can grab it if they want :)
As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets.
@piru off topic, but what was the development platform for MOS back in the old days (2000ish)?
-
I remember a thread on the mailing list a couple of years back, here we decided tat A. Inc were no longer a threat and access should be open again... But I guess no one bothered to do that... Probably because the people who want/need the source are developers who need write access too... The souce snapshot is available for down load from the site, so anyone can grab it if they want :)
As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets.
@piru off topic, but what was the development platform for MOS back in the old days (2000ish)?
Well, seems like I can't build it without login/password...
-
@piru off topic, but what was the development platform for MOS back in the old days (2000ish)?
A4000 with CyberStorm PPC and A1200 with Blizzard PPC running MorphOS. Very early on the userland was mostly AmigaOS with 68k apps. Also, not all kickstart modules were PPC native replacements yet, so some 68k modules from the kickstart ROM were used.
-
A4000 with CyberStorm PPC and A1200 with Blizzard PPC running MorphOS. Very early on the userland was mostly AmigaOS with 68k apps. Also, not all kickstart modules were PPC native replacements yet, so some 68k modules from the kickstart ROM were used.
Using gcc on AmigaOS? Or did you guys use any Mac/PC machines for large builds?
-
Using gcc on AmigaOS?
IIRC we had a cross compiler for 68k that produced binaries ppc-morphos target. But very early on we had everything running natively under MorphOS (native ixemul, native GG). Some very early things used SAS/C though.
Or did you guys use any Mac/PC machines for large builds?
We didn't. Or if we did I didn't pay much attention at that time :)
-
I can't see how they will fit all the things from just a 3.1 level kickstart without using tricks that the 4000T used, like moving Workbench.library to disk. That compiled C code will be bigger in almost every case.
Most of AmigaOS 3.1 is written in C. dos.library was written in BCPL up to 1.3 & converted to C for 2.0 (or even 1.4 beta?)
IIRC they used several compilers, one being lattice. Which turned into SAS/C. Back in the gcc 2.95 days I did some benchmarks and the software I tried was quicker with gcc than SAS/C.
There are many factors that will influence how large/fast AROS68K runs. It's far too early to even contemplate what it will be like.
However CD32/A1200's can have 1mb kickstart, A500/A2000's needs wires added. No idea about A3000/A4000...
-
Because there was no person interested and with enough knowledge. Now such person joined the team and the projected kickstarted ;)
A small correction, if I may:
Because there was no person interested and with enough knowledge and time. Now such person joined the team and the projected kickstarted.
:)
-
A small correction, if I may:
Because there was no person interested and with enough knowledge and time. Now such person joined the team and the projected kickstarted.
:)
What is nice is that since Jason has push the 68k section forward, other devs have joined in to assist! It really got the ball rolling :)
-
"As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets."
Yeah, that sounds very open source to me.
You can still make money with a GPL license. And it would really look more professional and legitimate to allow an object third party to say what the conditions are.
Selling:
http://www.gnu.org/philosophy/selling.html
-
"As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets."
Yeah, that sounds very open source to me.
You can still make money with a GPL license. And it would really look more professional and legitimate to allow an object third party to say what the conditions are.
Selling:
http://www.gnu.org/philosophy/selling.html
The world doesn't revolve around RMS or his licence you know.
In the real world the right tool for the right job is how things work. Same thing goes for licences too.
-
"As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets."
Yeah, that sounds very open source to me.
I appreciate your use of sarcasm here, but it isn't warranted. The MOS guys returned all the AROS code they used with bug fixes an a few extra features here and there... Without the MPL licence AROS wouldn't have got the bug fixes, full stop... It was a win win.
You can still make money with a GPL license. And it would really look more professional and legitimate to allow an object third party to say what the conditions are.
Selling:
http://www.gnu.org/philosophy/selling.html
AROS source code is there, you can download it, learn from it, play with it, do what you want... It will always exist in the public view, it will always be there, fork it if you like... I honestly can't see your objections in this regard. Shrug... Why waste your time arguing semantics?
-
Yeah, that sounds very open source to me.
You can still make money with a GPL license. And it would really look more professional and legitimate to allow an object third party to say what the conditions are.
Selling:
http://www.gnu.org/philosophy/selling.html
You do realize the term "open source" was coined by people that were tired of the GNU zealots and that wanted a more pragmatic approach, don't you?
I'm more concerned about what *I* can do with the code than what other people might do with the code, so to me at least it doesn't matter if someone else releases close source versions that includes some of the source.
-
"As for the licence, the modified MPL was chosen so that comercial projects like MOS could use the source code without having to publish their private code. Which is a good thing as it allowed the MOS guys to use and fix te AROS code without giving away their, trade secrets."
Yeah, that sounds very open source to me.
There are different grades of open source. GPL may be suitable for some projects while it isn't for some others. MorphOS team was happy to collaborate with AROS team on several components, donating the changes back to AROS tree, as specified by the license.
-
Well, I don't think the difference is semantic here, plainly. But I'll leave you to your version of "free and open".
-
@ The Goose
You are not making any sense. How is AROS not free and open ?
-
You can still make money with a GPL license. And it would really look more professional and legitimate to allow an object third party to say what the conditions are.
Selling:
http://www.gnu.org/philosophy/selling.html
You're allowed to make money by selling GPL software, that doesn't mean you can make money by selling it. Your first customer can give the source to whoever they want, they can even sell it.
GPL is like a parasite. It infiltrates every part of a system while keeping it alive. It seems to work for some software, but only where there is no possible business model & the programmers are prepared to do the work for free.
-
GPL is like a parasite. It infiltrates every part of a system while keeping it alive. It seems to work for some software, but only where there is no possible business model & the programmers are prepared to do the work for free.
Hi, Steve Ballmer. Didn't know you posted here. Red Hat and SUSE might want to have some words with you about this statement, though.
-
While the full AROS source tree requires SVN access, Buildable AROS sources are available anonymously through two methods:
1.) nightly source and contrib source code download
2.) Anonymous git repos AROS.git and AROS-Contrib.git at repo.or.cz
If you want a GPL'd Amiga OS replacement, start writing one. That's not impossible, Its just a lot of work. Software licenses like AROS Public License seem to encourage entrepreneurs who think they'll make some money off the work by value adding to them in various ways and being able to keep some things to themselves if they wish for a "competitive advantage in the marketplace", kind of like ...um imica or ares.