@XDelusion
From what I read, I do believe this basically is the game plan.
...
All in all I don't like this idea. I have been looking for modern(ish) hardware that I could score at a cheap price, and be able to run a large chunk of my more resoure intensive classic Amiga apps on, and until my eMac died, despite it's limited resources, has been the answer to my praryers! I would only hope that in the future that the MorphOS team would strive to maintain classic support on what ever new hardware that they may be targetting, for as long as possible. I like not needing Amiga rom images, I like not having to load up an emulator, I like being able to run my old software as if it were native, and it does not bother me in the least that I am not able to play the old games without emulation because of the chip set because when you are gaming, you are not using the OS, so I could really care less. It's when I'm running classic apps that I don't want to be running an emulator.
...
Anyhow, there is my UAE rant, and why I prefer MorphOS like it is. Backwards compatible. Though again, by the time all this takes place, we'll probably be at version 4.x which I'm sure is going to be VERY impressive, and I'm sure that when 3.x comes out, developers will get excited again and will begin to start coding fresh new apps and games for it.
Thank you for your long and well written post!
I agree that this route would be a big and non-trivial mental step of acceptance a MorphOS user would have to take, *should* it happen (which again is highly hypothetical, and something in the future anyway). There is a beauty in MorphOS's way of simply not caring if the binaries are 68k or PPC but treats them just the same, allowing them to truly work together in the same environment, they all using the very same resources, arexx communicating with them and binding them together (arexx being 68k itself even), etc, etc.
1) But if you want this to remain unchanged in the future, I think you must sacrifice the possibility of moving to a different architecture (due to big/little endian problems (at least on x86, not sure about ARM) *together* with the enormous effort it is to write a good 68k JIT native for that new ISA, which again would be pointless (at least on x86) due to endianness issues), and by that you will sacrifice the chance of having MorphOS running on future proof HW (the PPC is dead, and everything tied to it is bound to die as well). You would also sacrifice the possibility of having true SMP, true MP, 64-bit, etc, i.e. many of the things people have been crying for for several years now. You would also sacrifice the only chance for MorphOS of reaching a wider audience, outside this shrinking little community, or to even survive more than a couple of years ahead from now.
2) The flip side of the coin is that if you want to move the MorphOS platform to an architecture that still has a pulse and is future proof (meaning a chance of long term survival of MorphOS, the PPC is dead), and if you want to harvest the benefits of multiple CPU cores, memory protection, 64-bit, etc, then you are in for a break from the past. It means a new endian model and an Amiga API that is different in maybe few but way too fundamental ways for even trying to uphold any kind of backwards compatibility to the old Amiga API environment, a new set of API's, rules and guides would apply. The old (and current) 3.1 API centric applications have prerequisites, a way of function, and makes assumptions of their surrounding environment that simply wouldn't be true anymore. It would be a "3rd generation" system, and it won't be free of sacrifices to go there. To a user, it would look and feel the same, have much of the same features, the same Ambient, the same applications in a recompiled version (mostly, at least). It wouldn't at all be like the Mac's migration from OS9 to OSX, not even close, it would still be MorphOS with most of its advantages intact (and a couple of major *new* advantages on top of that). But we would probably have to settle with something like AROS is doing today for the 68k emulation part. And personally, I honestly think that would be an *acceptable* price to pay. And as has been construed by many people in this thread (including you), MorphOS is about so much more than just its superior 68k emulation...!
So the way I see it, there are basically two choices, leading in two different directions, and there are sacrifices to them both. Question is - which one do we prefer?
