Originally Posted by gulliver
I do agree with what you say, but up to some extent. Aros 68k fails terribly, IMHO, in the way it is implemented. I share much the view of Heiroglyph on that particular subject. Aros68k fails massively due to the way it is structured. It is not a 1:1 AmigOS replacement project, which would have been very useful for regression tests. We cannot mix and replace Amiga binaries and expect the system to work. There is a lot of non Amiga API code (HID for example, or whatever it is called) that have nothing to do with AmigaOS. Then you have compiler generated code, which seems to be huge and bloated compared with ancient AmigaOS C compilers. There is now way to mix and reuse RomTags without a great rework. And the list could go on forever.
So, for me, in order to put Aros68k back on track a fork or a complete roadmap change is required, because if it continues to follow the way it has, it will certainly still be what it is now: an outcast from real Amiga hardware destined solely for some corner case use.
I really hope this changes.
hoping doesnt account for much.
such things mus be discussed. not aggressively, and one by one, compromises may be found. but it needs that people get involved, they of course cant simply overthrow how aros is structured for now, but working their way in, earning trust of the team as jason did, they can adjust things in a way more favourable for us (68k). it is at least a possibility.
i dont see any other infrastructure thet would provide such possibility, since simply there is no other open and working amiga os reimplementation ot there.
heiroglyph approach is all fine, and im supporting him, but i doubt that it is maintainable. he had several projects worked in in the last years but he didnt publish them up till now, so this is an example of what im saying, an individual cant lift such a burden alone, no matter how skilled, and how much good will he has at a given point. all alone he will eventually give up. we need to team up, as apollo people did.