The situation

[Amiwell]

meanwhile we start with the sale by aros here, who wants a copy or two you book, we open a special thread

[Amiwell]

for me the thing is positive and maybe not only for me we paid 2000 euros for audio evolution, maybe it would take an effort of us all for terminills

[Amiwell]

@terminills

the program is finished?

[deadwood]

from what i remember when m68k target was introduced and jason and toni picked up the task, which was when i got into aros,

You might be correct in that shortly prior to Jason's work the 68k target was been taken more serious and the abi split went shortly before that. It can be checked from the commit history (which i was too lazy to do).

The history is a bit different.

ABIv1 was planned for x86 (32-bit) and was being worked on by Staf on separate branches for several years prior to moving it to trunk. Then two things happened:

a) AROS was growing fast and amount of changes made to it were causing Staf's branches to have conflicts all the time and Staf spending time on resolving the conflicts instead of working in ABIv1
b) m68k target of Jason needed the change in several APIs and LVO locations which Jason was still trying to do without impacting compatibility, which meant workarounds

In order to help both topics, the decision was made to "break" trunk. This was Staf's codes could move from branch to trunk and Jason could make all the binary compatibility changes also without need for workarounds. I think at that time the developer community was fairly uniform in supporting that move (including me to some degree) or was not voicing concern.

[terminills]

Final note on this.   I can tell you I've spent over $20,000 on Final Writer and I've literally almost cancelled it multiple times due to lack of interest and feedback.

I wonder how you may recoup all these money. You'd need 200 users paying $100 each or, more realistically, 400 users paying $50. You must release it for all platforms to get this result.

I never expect to recoup my money tbh.  Sure it would be great but in all actuality, it is more of a gift to the community.

[deadwood]

Why don't you wanna work on abi1? I am innocently asking, excuse my ignorance but I don't don't know the entire AROS history

I'll answer this as well as similar question from @wawa.

The reason is that by now, seing what I have seen, I have no belief that ABIv1 will ever be finished.

With that in mind I had a few options:
a) wait some more for a mirracle
b) forgot about the whole thing
c) take what is already there and decide that it is good enough for me and others who have similar view and move forward

I was "waiting some more" for a few years, tried twice to persuade to have a change in direction and then I finally just decided to take another route.

Believe me, I would be extremly happy to be proven wrong with my initial statement. The last thing I need is maitaining a parallel repository and backporting changes from the main one and several satelite ones. Seing an annoucement AROS in main repository is now stable and backwards compatible and devs will obey by this, would be a great thing to experience.

Also please be aware that you only keep listening to my view of the topic. To make a ballanced opinion for yourself you need to listen also to people who hold a different view.

[terminills]

for me the thing is positive and maybe not only for me we paid 2000 euros for audio evolution, maybe it would take an effort of us all for terminills

I'm fine financially.  I was merely pointing out that the lack of proper input is an issue in the AROS community.

[Amiwell]

@deadwood maybe a miracle will happen then, @terminills ok thank you

[paolone]

I was "waiting some more" for a few years, tried twice to persuade to have a change in direction and then I finally just decided to take another route.

Believe me, I would be extremly happy to be proven wrong with my initial statement. The last thing I need is maitaining a parallel repository and backporting changes from the main one and several satelite ones. Seing an annoucement AROS in main repository is now stable and backwards compatible and devs will obey by this, would be a great thing to experience.

I see there is life in AROS repository, but at the moment I pragmatically consider the whole x86-64 version "too raw" and also too much 'regressed' for final users: no 3rd party software, no sound, no 3D drivers, even programs in contrib fail to work. That's why I stick with ABIv0 i386 version for the mainstream version of Icaros Desktop. But I believe this:
1. some day or another, ABIv0 AROS will be utter obsolete. There won't be anyone willing to use less than 2 cores or 4 GB of RAM. Cheap hard drives have terabyte of space and we can use only 128 GB partitions (even less: if you get too near to this barrier, your filesystem will fail soon). Moreover, the last person working on it was you. After you 'left', the only improvement to the system has been the VMware SVGA driver, which is still incomplete and lacks all the 3D acceleration stuff. How many chances there are I can continue with this system codebase in the future? Unless anyone provides a new backport, it will starve as it is now.
2. There is a time when you have to look forward. I am not killing ABIv0, but AROS users are. Our userbase has ever been composed by a few people actually using the system everyday, and a vast majority just updating Icaros to see what's new. In particular on the social networks, I've received many vocal requests about moving to ABIv1 x64, I've been even accused of keeping the OS development "stopped" for many years just because I didn't choose to move to ABIv1 i386, as if Icaros had all this power. There was also some interest by Nikos to create a 64bit AspireOS (interest that - as far as I can see - has fallen a lot afterwards), but AROS x64 can't even boot from the hard drive once installed, and I guess this put him off, really. I chose to insist, because problems are here to be resolved, bugs to be fixed, and so on. I sincerely want a 64bit version of Icaros, in the hope it will help AROS grow like the 32bit version did 10 years ago. I know it won't happen with the same pace, but I still hope it will happen.
3. "The last thing I need is maitaining a parallel repository and backporting changes from the main one and several satelite ones": that's exactly what I am afraid of, when I see forks or potential forks of the main repo. Today we have your x64 flavour and Kalamatee's one: I am quite sure programs compiled for one will fail on the other. And if this does not happen today, it will happen tomorrow. Just imagine me, trying to make a distribution, with half the 3rd party programs working for a flavour and the other half for the other (BTW: this is the exact situation I got with current few programs released on the Archives for x86-64 AROS. NONE OF THEM WORKS on main repo's system). I would get mad in no time. We already have a stable branch and it is i386 ABIv0. Let's keep the others work in progress, but as unified/coherent as possible, and try to enhance/fix this one. The last thing we need is more fragmentation.

4. I am trying to push development on AROS and for AROS as much as I can, trying to convince newcomers into using the AROS build system as a 'standard' environment. The more our build chain is standardised and automated, the faster and the easier recompiling will be, if something in the ABI changes. That's the scope of my VM, but there are interesting efforts on Docker as well from other people. I am also trying to evangelize the open source model as much as I can. You were working on a great instrument for development too: has it progressed?

[miker1264]

The good news is that AROS 64bit which also means Icaros x86-64 is headed in the right direction. We are getting more people involved that are interested in programming and trying to prioritize what needs to be done at a minimum for AROS x64. So far we are focused on Scalos, Wanderer, Dopus. Others are focused on Docker. AROS 68k is also a priority and hopefully sometime soon someone will address the 64bit boot issues. There are just so many things to get done.

[deadwood]

3. "The last thing I need is maitaining a parallel repository and backporting changes from the main one and several satelite ones": that's exactly what I am afraid of, when I see forks or potential forks of the main repo. Today we have your x64 flavour and Kalamatee's one: I am quite sure programs compiled for one will fail on the other. And if this does not happen today, it will happen tomorrow. Just imagine me, trying to make a distribution, with half the 3rd party programs working for a flavour and the other half for the other (BTW: this is the exact situation I got with current few programs released on the Archives for x86-64 AROS. NONE OF THEM WORKS on main repo's system). I would get mad in no time. We already have a stable branch and it is i386 ABIv0. Let's keep the others work in progress, but as unified/coherent as possible, and try to enhance/fix this one. The last thing we need is more fragmentation.

Once upon a time there was a symbiosis between two groups of developers which allowed AROS to grow and flourish. First group was interested in major advancements that changed the landscape. As they did those advancements, they inherently broke a lot of other stuff. Second group was interested in more gradual changes and were often fixing what get broken when making major advancements. From that symbiosis, while they often argued with each other, both groups got what they needed: first got enjoyment from experimentation and advance, second got enjoyment from seeing the advancements being brought to user base and actually being used by people.

What held that symbiosis together however was a simple covenenant: "Though shall not break backward compatibility". Regardless of how many bugs were introduced, as long as this one thing was kept in place, there was continuity.

This changed in 2011 (yes, it is 8 years already), when the mantra became: "This is development version, we can break what we want when we want, it is not intendent for the users (or more radical versions of the same)". With this, both groups drifted apart.

I believe AROS will not grow again without this symbiosis being in place and this symbiosis will not happen without the covenant being brought back. What I did in my fork, I brough back that covenant. Time will tell whether it was the right approach.

You were working on a great instrument for development too: has it progressed?

It did not during summer. I spent energy on some outdoor activities I came to enjoy in recent years. The winter is coming however, so we shall see

[OlafS3]

@deadwood

yes that is how I see it too

Aros became pure "research" but not really "replacement"

Users do not care about advanced USB support as long the whole system is not stable and there are bugs that make normal use impossible. The system must run stable and (as far as possible) bug free, it must be easy to install and hardware must be supported. As long this is not the case most users will drop it immediately

My idea (being no OS developer of course) concentrate on hosted versions regarding X86/X64 (Windows, Mac, Linux) with full support of the underlying OS (sound, USB, 3D, Network) and on V4 (68k) and RPi (ARM) native (because hardware is relative static)

[phoenixkonsole]

Well .. aros hosted / Wanderer is the ui of around 6000 sold Odroids XU4 in 2018/19

Problem: people don’t care if it is aros or any other os as long the devices do what they shall ..

Just saying that aros has by far the largest distribution base .. but completely wrong audience : )

The audience we are looking for doesn’t seem to exist anymore in today’s world.

[phoenixkonsole]

Next thing with completely wrong audience is the Bluephone .. just a China  oem phone with Aeros + android .. ~ not Amiga focused but it is aros hosted „powered“



So users are not equal to userbase : )
But to recap:
All the transcendence projects lead to a shipload of work / payment possibilities.
My idea last year was to have aros being a piece of the construct to get funding..

So market ? Check and marked as solved
Funding? Checked and marked as solved

Now we need a foundation to move money form a to b ..
we experimented with https://governance.rocks to create a voting but also funding platform..

[deadwood]

@phoenixkonsole

I'm not sure I understand. You are telling me there are commercial ARM based devices out there shipped with linux-hosted + Wanderer as their main UI?