- Home
- Discussion Forum
- AROS FORUMS
- Rants and Other Discussions
- ABIv1 vs ABIv11 and ABIv0, Branches, Upstream, Downstream, and other related items
ABIv1 vs ABIv11 and ABIv0, Branches, Upstream, Downstream, and other related items
Separating the two different conversations that are happening in the same thread.
Quote
@terminills It's bad enough this site hijacked the AROS name and the steers people away from the actual main project
How is that so?
This site exists as a place for developers, AROS users (new and old), former Amiga users, and others to discuss the project. How is it any different from forums that came before it — aros-exec.org, ae.aros-exec.org — or the Discord channel?
How else would you expect people to give feedback, suggest ideas, report bugs, share new software, or discuss new developments without a forum? It's a site about AROS, so it uses the name AROS — that's so people aren't left confused about what it's for, not because it's trying to hijack anything.
Nobody here is hijacking the name. People are free to talk about AROS as long as it's done respectfully, without attacking anyone's personal preferences or opinions.
If I've misread your comment, I apologize — but I'm not sure what else you could have meant by it.
Edited by amigamia on 04-09-2026 17:41, 3 days ago
1 user reacted to this post
sonountaleban
terminillsMember
Posted 19 hours agoQuote
This site exists as a place for developers, AROS users (new and old), former Amiga users, and others to discuss the project. How is it any different from forums that came before it—aros-exec.org, ae.aros-exec.org—or the Discord channel?
How else would you expect people to give feedback, suggest ideas, report bugs, share new software, or discuss new developments without a forum? It's a site about AROS, so it uses the name AROS—that's so people aren't left confused about what it's for, not because it's trying to hijack anything.
Nobody here is hijacking the name. People are free to talk about AROS as long as it's done respectfully, without attacking anyone's personal preferences or opinions.
If I've misread your comment, I apologize—but I'm not sure what else you could have meant by it.
This site actively directs people away from AROS.org and the main AROS development branch.
I have nothing against Deadwood or his work. I know he puts considerable effort into it. However, his repository is his own fork; it is not the upstream AROS project or its main development branch. Hopefully, one day the separate fork will no longer be necessary and the work can be fully aligned upstream.
The problem is not that this forum uses “AROS” in its name. The problem is how it presents the project. ABIv1—the actual ADT main development branch—is conspicuously omitted from the site’s introduction and its general “Development Resources” article. That article presents ABIv0 and Deadwood’s ABIv11 as the available development choices, then directs developers to contact Deadwood while omitting the ADT repository, its GitHub issue tracker, pull requests, developer mailing list, and official contribution process.
The site’s introduction claims AROS support for x86, x86-64, m68k, ARM, and RISC-V, but fails to explain that ABIv1 is the upstream branch connecting those architectures. Most mentions of ABIv1 elsewhere on the site consist of people discouraging others from using it, customizing its nightlies, or basing distributions upon it.
That is what I mean by hijacking the name: presenting this as the general home of AROS while steering users and prospective developers toward one preferred fork and away from the actual upstream project. A forum for discussing AROS is perfectly reasonable. Replacing the public perception of upstream AROS with a fork is not.
Edited by terminills on 04-09-2026 17:55, 3 days ago
Terminills, no one here is against AIBv1, and I hope that one day there will be just one AROS—divisions have never led to anything.
If all the distributions use the fork, there must be a reason. Even though AIBv1 is more advanced, Deadwood has done remarkable work to improve AROS; there have been many fixes and improvements over the years. You probably haven’t been following along and may not be aware of these improvements.
Let me give you one example above all others: if you try to compile with the main AROS 68k, you’ll get a “GCC not executable” error. The same goes for the Contrib 68k software, which includes non-executable binaries—and no one has ever bothered to fix it, something that deadwood did in his 68k version.
If all the distributions use the fork, there must be a reason. Even though AIBv1 is more advanced, Deadwood has done remarkable work to improve AROS; there have been many fixes and improvements over the years. You probably haven’t been following along and may not be aware of these improvements.
Let me give you one example above all others: if you try to compile with the main AROS 68k, you’ll get a “GCC not executable” error. The same goes for the Contrib 68k software, which includes non-executable binaries—and no one has ever bothered to fix it, something that deadwood did in his 68k version.
terminillsMember
Posted 19 hours ago@AMIGASYSTEM - Terminills, no one here is against AIBv1, and I hope that one day there will be just one AROS—divisions have never led to anything.
If all the distributions use the fork, there must be a reason. Even though AIBv1 is more advanced, Deadwood has done remarkable work to improve AROS; there have been many fixes and improvements over the years. You probably haven’t been following along and may not be aware of these improvements.
Let me give you one example above all others: if you try to compile with the main AROS 68k, you’ll get a “GCC not executable” error. The same goes for the Contrib 68k software, which includes non-executable binaries—and no one has ever bothered to fix it, something that deadwood did in his 68k version.
oh please show me the bug reports submitted in the AROS issue tracker? ;D And you specifically are the exact reason I will not allow software I own to be used on ABIv11. and as for following ... I literally spend $1000's a year supporting developers while you beg for coffee money(my name is literally the first non commercial sponor on the AROS sponsor page). You literally come on here begging people to port software that has already be in ABIv1 for months or years so maybe it's you who hasn't been paying attention.
Edited by terminills on 04-09-2026 18:30, 3 days ago
Terminills, we're just discussing this, i don't beg for anything; I've never asked anyone for anythingdon't get upset. I'm not a developer; I'm just a 73-year-old guy who enjoys playing around with AROS. Tell me what you want me to remove from Aros One, and I'll remove it!
terminillsMember
Posted 19 hours ago@AMIGASYSTEM - Terminills, we're just discussing this, don't get upset. I'm not a developer; I'm just a 73-year-old guy who enjoys playing around with AROS. Tell me what you want me to remove from Aros One, and I'll remove it!
There was a quite simple solution… “sorry for the omission” and adding a couple paragraphs and links to the site about the actual main project.
But instead you doubled down.
Do not ever include Finalwriter Demo or otherwise in any of your distros.
No problem, Finalwriter will no longer be included in any version of AROS One. If you have any other software of yours, let me know and I’ll remove it.
I’m really sorry that a normal discussion made you angry. If I did something wrong or if the translation didn’t reflect what I meant, I certainly didn’t mean to offend you. Have a good evening.
I’m really sorry that a normal discussion made you angry. If I did something wrong or if the translation didn’t reflect what I meant, I certainly didn’t mean to offend you. Have a good evening.
terminillsMember
Posted 19 hours ago@amigamia -@terminills -Quote
This site exists as a place for developers, AROS users (new and old), former Amiga users, and others to discuss the project. How is it any different from forums that came before it—aros-exec.org, ae.aros-exec.org—or the Discord channel?
How else would you expect people to give feedback, suggest ideas, report bugs, share new software, or discuss new developments without a forum? It's a site about AROS, so it uses the name AROS—that's so people aren't left confused about what it's for, not because it's trying to hijack anything.
Nobody here is hijacking the name. People are free to talk about AROS as long as it's done respectfully, without attacking anyone's personal preferences or opinions.
If I've misread your comment, I apologize—but I'm not sure what else you could have meant by it.
This site actively directs people away from AROS.org and the main AROS development branch.
I have nothing against Deadwood or his work. I know he puts considerable effort into it. However, his repository is his own fork; it is not the upstream AROS project or its main development branch. Hopefully, one day the separate fork will no longer be necessary and the work can be fully aligned upstream.
The problem is not that this forum uses “AROS” in its name. The problem is how it presents the project. ABIv1—the actual ADT main development branch—is conspicuously omitted from the site’s introduction and its general “Development Resources” article. That article presents ABIv0 and Deadwood’s ABIv11 as the available development choices, then directs developers to contact Deadwood while omitting the ADT repository, its GitHub issue tracker, pull requests, developer mailing list, and official contribution process.
The site’s introduction claims AROS support for x86, x86-64, m68k, ARM, and RISC-V, but fails to explain that ABIv1 is the upstream branch connecting those architectures. Most mentions of ABIv1 elsewhere on the site consist of people discouraging others from using it, customizing its nightlies, or basing distributions upon it.
That is what I mean by hijacking the name: presenting this as the general home of AROS while steering users and prospective developers toward one preferred fork and away from the actual upstream project. A forum for discussing AROS is perfectly reasonable. Replacing the public perception of upstream AROS with a fork is not.
Actively directing people away from aros.org would mean this site telling users to avoid aros.org or to avoid ABIv1 specifically — but that's nowhere to be found. You're confusing chip architecture with ABI. As you acknowledged in your post, "The site's introduction claims AROS support for x86, x86-64, m68k, ARM, and RISC-V" — that has nothing to do with ABIv0, ABIv1, ABIv11, or any other ABI version. To the general public, it's simply presenting which chip architectures AROS can run on.
Which ABI does what is a developer-level concern. ABIv11 is the work of Deadwood, which, in my opinion, has pushed AROS on x64 further than ABIv1 was doing at the time. Deadwood is a user on this site, just like you, and like him, you're free to post and advocate for ABIv1 if you feel it isn't getting the attention it deserves. No one is stopping you.
So instead of projecting negative assumptions and accusing this site or its members of misconduct, why not participate more constructively? You're free to express yourself, and if you feel something is heading in the wrong direction, you're welcome to voice your opinion — but direct attacks won't be tolerated, as they may violate this site's terms of service.
Since we're apparently claiming nobody here directs users away from ABI v1, here are some examples from this site.
October 18, 2024 — Odyssey Web Browser 64-bit
AMIGASYSTEM responds to discussion of the official experimental ABI v1 SMP build by saying that it does not work for him under VMware and characterizes ABI v1 as effectively having almost no software.
Odyssey Web Browser 64-bit
In the same thread, Amiwell79 states:
Quote
but owb will work on abiv11 not abiv1
Deadwood immediately corrects this in the following post, explaining that Odyssey can work on any AROS variant, including ABI v1, provided somebody compiles and debugs it.
Odyssey Web Browser 64-bit — posts #30/#31
December 18, 2025 — Grub is broken on 32 bit x86 nightly
This one is particularly interesting because the user is actually trying to use the official ABI v1 nightly.
AMIGASYSTEM tells the user that ABI v1 is incompatible with the distributions and almost everything on AROS Archives and states:
Quote
There is no third-party software for AROS ABI-v1
He then points the user toward ABI v0/v11 builds instead:
Quote
You could try these Nightly Build here
Grub is broken on 32 bit x86 nightly — post #4
This isn't merely someone saying that they personally prefer ABI v11. A user arrived using the official ABI v1 nightly and was actively directed toward another ABI.
December 19, 2025 — same thread
After the claim that ABI v1 has no third-party software is challenged, AMIGASYSTEM changes the argument to there being relatively too little software for ABI v1 compared with ABI v0/v11 and uses that as the reason for not producing an ABI v1 AROS One.
Grub is broken on 32 bit x86 nightly — post #7 onward
December 19, 2025 — same thread
AMIGASYSTEM again argues against an ABI v1 distribution and says that he previously created an ABI v1 AROS One but abandoned it because there was no software.
Grub is broken on 32 bit x86 nightly — post #10
And here is the important part: the redirection actually worked.
After being told the above, Build-0-Matic says:
Quote
OK, so I'm now on the hunt for ABIv0
He subsequently reports downloading ABI v0 and getting it working.
Grub is broken on 32 bit x86 nightly
So this isn't a hypothetical argument about whether comments might discourage ABI v1 use.
A user came here using the official ABI v1 nightly, was told that there was essentially no software for it, was pointed toward ABI v0/v11 instead, and subsequently went looking for ABI v0.
To be clear:
I'm not counting every post where somebody simply says that they use or prefer ABI v11. People are free to use whatever they want.
I'm specifically talking about posts where users are discouraged from using ABI v1, told software isn't available for it, told software only works on ABI v11, or are directly pointed toward ABI v0/v11 instead.
So the claim that nobody here is against ABI v1 or that nobody is steering users away from it simply doesn't match what has actually been posted on this site.
There is also a second problem with steering users away from ABI v1: it creates a distorted picture of who actually did the work.
We can see an example of that in this discussion, where the advancement of AROS on x86-64 is being broadly attributed to Deadwood and ABI v11.
To be clear, Deadwood has done substantial work and deserves credit for the work he actually did. This isn't an attempt to take anything away from him.
The problem is that a large amount — arguably the lion's share — of the x86-64 development being attributed to Deadwood was actually done upstream by Kalamatee/Nick Andrews.
The upstream commit history is there for anyone to examine.
Nick's x86-64 work goes back years and covers a huge portion of the platform development: x86-64 build support, toolchain work, SMP, kernel development, interrupt handling, memory management, modern PCIe/hardware support and the continuing work required to make AROS operate properly on modern multicore x86-64 systems.
That distinction matters because ABI v11 did not spring into existence independently and then somehow advance x86-64 AROS in isolation. It is a fork of AROS and necessarily contains and benefits from upstream AROS development.
Now combine that with users being actively steered away from ABI v1.
If users are repeatedly told that ABI v1 has little or no software, that applications only work on ABI v11, or are directed toward ABI v0/v11 instead of the official builds, those users are much less likely to see the upstream development as it happens.
The result is a rather predictable feedback loop:
Quote
Steer users away from upstream.
Users don't see upstream development.
Upstream work is incorporated downstream.
Users encounter it downstream and credit the fork.
The fork is then described as being further ahead.
That claim becomes another reason to steer users away from upstream.
Eventually you end up with exactly the misconception demonstrated here: the lion's share of years of upstream x86-64 development gets casually attributed to the downstream fork and its maintainer.
Again, this isn't about denying Deadwood credit.
Deadwood deserves credit for Deadwood's work.
Nick deserves credit for Nick's work.
And if we're going to discuss which branch "pushed AROS on x64 further," then we should look at who actually wrote the code rather than simply attributing everything found in ABI v11 to ABI v11.
If the stated goal is eventually having "just one AROS," actively steering users away from upstream while simultaneously attributing upstream development to the fork seems like a very strange way of achieving it.
Edited by terminills on 04-09-2026 20:13, 2 days ago
1 user reacted to this post
olivier2222
@terminills -Quote
This site exists as a place for developers, AROS users (new and old), former Amiga users, and others to discuss the project. How is it any different from forums that came before it—aros-exec.org, ae.aros-exec.org—or the Discord channel?
How else would you expect people to give feedback, suggest ideas, report bugs, share new software, or discuss new developments without a forum? It's a site about AROS, so it uses the name AROS—that's so people aren't left confused about what it's for, not because it's trying to hijack anything.
Nobody here is hijacking the name. People are free to talk about AROS as long as it's done respectfully, without attacking anyone's personal preferences or opinions.
If I've misread your comment, I apologize—but I'm not sure what else you could have meant by it.
This site actively directs people away from AROS.org and the main AROS development branch.
I have nothing against Deadwood or his work. I know he puts considerable effort into it. However, his repository is his own fork; it is not the upstream AROS project or its main development branch. Hopefully, one day the separate fork will no longer be necessary and the work can be fully aligned upstream.
The problem is not that this forum uses “AROS” in its name. The problem is how it presents the project. ABIv1—the actual ADT main development branch—is conspicuously omitted from the site’s introduction and its general “Development Resources” article. That article presents ABIv0 and Deadwood’s ABIv11 as the available development choices, then directs developers to contact Deadwood while omitting the ADT repository, its GitHub issue tracker, pull requests, developer mailing list, and official contribution process.
The site’s introduction claims AROS support for x86, x86-64, m68k, ARM, and RISC-V, but fails to explain that ABIv1 is the upstream branch connecting those architectures. Most mentions of ABIv1 elsewhere on the site consist of people discouraging others from using it, customizing its nightlies, or basing distributions upon it.
That is what I mean by hijacking the name: presenting this as the general home of AROS while steering users and prospective developers toward one preferred fork and away from the actual upstream project. A forum for discussing AROS is perfectly reasonable. Replacing the public perception of upstream AROS with a fork is not.
Actively directing people away from aros.org would mean this site telling users to avoid aros.org or to avoid ABIv1 specifically — but that's nowhere to be found. You're confusing chip architecture with ABI. As you acknowledged in your post, "The site's introduction claims AROS support for x86, x86-64, m68k, ARM, and RISC-V" — that has nothing to do with ABIv0, ABIv1, ABIv11, or any other ABI version. To the general public, it's simply presenting which chip architectures AROS can run on.
Which ABI does what is a developer-level concern. ABIv11 is the work of Deadwood, which, in my opinion, has pushed AROS on x64 further than ABIv1 was doing at the time. Deadwood is a user on this site, just like you, and like him, you're free to post and advocate for ABIv1 if you feel it isn't getting the attention it deserves. No one is stopping you.
So instead of projecting negative assumptions and accusing this site or its members of misconduct, why not participate more constructively? You're free to express yourself, and if you feel something is heading in the wrong direction, you're welcome to voice your opinion — but direct attacks won't be tolerated, as they may violate this site's terms of service.
2 users reacted to this post
sonountaleban
olivier2222
Terminills, first of all, I want to point out that I’m using a translator because I don’t know English, and a lot of what’s written here isn’t my exact wording but rather the translator’s interpretation…
If a user tells me that AROS ABiv1 won’t start in VMware, and after testing it I confirm this, what should I say to that user? And if the user then tells me that the new ports and the new software released for v11 don’t work on v1, what should I tell them?
If a user tells me that AROS ABiv1 won’t start in VMware, and after testing it I confirm this, what should I say to that user? And if the user then tells me that the new ports and the new software released for v11 don’t work on v1, what should I tell them?
OK, looks like I managed to get you to open up and explain in detail what you could have done in the first place. Perhaps I would have used a better and more constructive tone and less attacking one.
You accused this site being responsible for steering away users from aros.org and/or ABIv1 when you really meant was for specific posts by certain members.
Please keep in mind, this is a forum site where members are free to join to discover aros, share ideas, experiences, software, etc. etc. and in the mix sometimes users came up with an incorrect view or a misconception about AROS.
As I said earlier, you have the power to correct them and explain (or better "prove") that what they are posting in here is incorrect or not based on facts. You are a member of this site just as they are. And if you feel that this site doesn't work for you in conveying your message to the masses, you are more than welcome to post a statement on aros.org that explains what is ABIv1, ABIv11, ABIv0 and who does what, who is responsible for whatnot.
You accused this site being responsible for steering away users from aros.org and/or ABIv1 when you really meant was for specific posts by certain members.
Please keep in mind, this is a forum site where members are free to join to discover aros, share ideas, experiences, software, etc. etc. and in the mix sometimes users came up with an incorrect view or a misconception about AROS.
As I said earlier, you have the power to correct them and explain (or better "prove") that what they are posting in here is incorrect or not based on facts. You are a member of this site just as they are. And if you feel that this site doesn't work for you in conveying your message to the masses, you are more than welcome to post a statement on aros.org that explains what is ABIv1, ABIv11, ABIv0 and who does what, who is responsible for whatnot.
Edited by amigamia on 04-09-2026 21:12, 2 days ago
3 users reacted to this post
Argo, sonountaleban
olivier2222
terminillsMember
Posted 19 hours ago@AMIGASYSTEM - Terminills, first of all, I want to point out that I’m using a translator because I don’t know English, and a lot of what’s written here isn’t my exact wording but rather the translator’s interpretation…
If a user tells me that AROS ABiv1 won’t start in VMware, and after testing it I confirm this, what should I say to that user? And if the user then tells me that the new ports and the new software released for v11 don’t work on v1, what should I tell them?
Well the correct response would have been to explain where the AROS bugs are reported… that was a perfect example you told him not to use ABIv1 … I reported the bug to kalamatee and the bug was squashed in less than 24 hours. Nothing is constructive by ignoring bugs and steering people towards your preferred platform.
As for software … quite simple not there’s no software for but instead explain the difference.. you’re creating a self fulfilling prophecy.
Edited by terminills on 04-09-2026 23:21, 2 days ago
sonountalebanMember
Posted 19 hours agoNot sure where to start from... anyway, I used to lurk in AROS.org and its associated forum in the previous decade until somebody decided to close it down, I remember very few guys (the Icaros guy for example) kept posting occasionally there, which meant for me (and probably many other people) that AROS was dead. I've rejumped again into AROS last year thanks to this forum, by the way the only place that actually looks alive with a sort of "community". I don't know the real reasons why Deadwood has created a new fork few years ago, I really want to know them because I suspect they were mostly related to the stagnating development of 64 bit version rather than technical issues - correct me if I'm wrong please. I found out here just recently there are these v1 and v11 versions and to be fair nobody talked badly (at least explicitly) about v1, last week Deadwood himself advised me to install and try v1 to check hardware compatibility. At practical level I see v11 is pretty supported with a growing archive website and community, while I can't say the same for v1 at the moment, apart AROS.org there are no other web resources... From a point of view I think it's fair to push to use v11 to be honest. Also I have to spend some frankly and unpleasant words about AROS.org: it's still stuck at Geocities era as I left long time ago, with most pages ridiculously outdated. I still see the Developers page with the original '90s developers pictures, who already abandoned AROS 20 years ago! So, you Terminills come here, occasionally post a bit snidy replies and you expect people to know the story, while your website that you v1 guys should take care is in questionable state and offers basically little information to newcomers. No mention about differences between v1 and v11 there, too... The v11 guys may have their faults but you v1 guys have a part of guilt about the current state of the things, of course. It's sad to read "this is my precious" type of stories above, it reminds me too much the AmigaOS4 drama, the Belgian mafia and all their sick ****. 
I think the best thing to do is to revamp completely AROS.org with more transparent information and the same here, maybe adding sections for x86 v1 and v11 and other flavours. The rest is just a BS.
I think the best thing to do is to revamp completely AROS.org with more transparent information and the same here, maybe adding sections for x86 v1 and v11 and other flavours. The rest is just a BS.
Edited by amigamia on 05-09-2026 10:34, 2 days ago
2 users reacted to this post
Argo
olivier2222
terminillsMember
Posted 19 hours ago@sonountaleban - Not sure where to start from... anyway, I used to lurk in AROS.org and its associated forum in the previous decade until somebody decided to close it down, I remember very few guys (the Icaros guy for example) kept posting occasionally there, which meant for me (and probably many other people) that AROS was dead. I've rejumped again into AROS last year thanks to this forum, by the way the only place that actually looks alive with a sort of "community". I don't know the real reasons why Deadwood has created a new fork few years ago, I really want to know them because I suspect they were mostly related to the stagnating development of 64 bit version rather than technical issues - correct me if I'm wrong please. I found out here just recently there are these v1 and v11 versions and to be fair nobody talked badly (at least explicitly) about v1, last week Deadwood himself advised me to install and try v1 to check hardware compatibility. At practical level I see v11 is pretty supported with a growing archive website and community, while I can't say the same for v1 at the moment, apart AROS.org there are no other web resources... From a point of view I think it's fair to push to use v11 to be honest. Also I have to spend some frankly and unpleasant words about AROS.org: it's still stuck at Geocities era as I left long time ago, with most pages ridiculously outdated. I still see the Developers page with the original '90s developers pictures, who already abandoned AROS 20 years ago! So, you Terminills come here, occasionally post a bit snidy replies and you expect people to know the story, while your website that you v1 guys should take care is in questionable state and offers basically little information to newcomers. No mention about differences between v1 and v11 there, too... The v11 guys may have their faults but you v1 guys have a part of guilt about the current state of the things, of course. It's sad to read "this is my precious" type of stories above, it reminds me too much the AmigaOS4 drama, the Belgian mafia and all their sick ****.
I think the best thing to do is to revamp completely AROS.org with more transparent information and the same here, maybe adding sections for x86 v1 and v11 and other flavours. The rest is just a BS.
You said "correct me if I'm wrong please." That's the one part of the post worth taking at face value, so here is the correction, and it comes from the repositories rather than from anyone's website.
The forum was not AROS.
Start here, because everything else in the post rests on it. The forum that closed was not aros.org and it was not the development project. Its disappearance tells us that a community venue disappeared. It tells us nothing whatsoever about whether anyone was writing code.
Nor is the history "the forum closed, everyone assumed AROS was dead, and then this community brought it back." That retelling skips the part worth knowing: why that forum stopped being a place the people actually developing AROS wanted to be. When developers and ordinary users leave a venue over the conduct of the same small group, what remains is not a representative community. It's the people who didn't leave, talking to each other. A handful of accounts posting for years is not evidence of where a project lives.
So "this is the only place that looks alive" doesn't mean development happens here. It can just as easily mean this is where the ones who stayed are still talking.
The same substitution runs through the whole post. "I didn't see much activity" became "AROS was dead." "This forum is active" became "this is the AROS community." "The website looks dated" became "the people behind upstream AROS are responsible for the state of the project." Not one of those follows from the one before it.
"AROS was dead."
Since the Git repository was created in May 2019, upstream AROS has taken 6,860 commits. In the last twelve months alone it took 2,492, from 32 different people. The most recent landed this week. Across that whole period Kalamatee authored 4,656 of them, about 68%. Deadwood authored 852.
Nobody discovered that AROS was dead. What was discovered is that a forum had closed, and a conclusion about a software project was drawn from the availability of a website. Those are not the same thing, and the distinction takes under a minute to demonstrate.
"Deadwood forked because 64-bit development was stagnating."
The post says "I suspect," so let's not pretend it's established. What can be established is that the fork diverged in May 2019, at the SVN-to-Git migration, and what it has consisted of since.
His repository holds 6,584 commits. He is the committer on 6,454 of them - 98%. He is the author of 2,443, about 37%. The largest body of authored work in that repository belongs to Kalamatee: 3,314 commits, just over half. Kalamatee has never committed to it. Not once, in its entire existence.
That isn't an accusation against Deadwood. His tree records authorship correctly, which is precisely how those figures were produced - from his clone. Anyone in this thread can run git shortlog -sne --no-merges against it and get the same numbers. He also does substantial original work, and his 2,443 commits are real.
But committing another person's work is not writing it. So when newcomers are told that v11 is well supported while upstream is finished, what is actually being described is upstream development reaching them through a downstream distribution, weeks to months after it was written. The fork gets the credit for it. The people who wrote it get described as having abandoned the project.
"v11 is pretty supported."
Some specifics, since this is where it stops being abstract.
RISC-V: 48 commits upstream, all by Kalamatee. Zero of them exist in the fork.
ARM and AArch64, including the Raspberry Pi work: 249 commits upstream, the majority by Bo Ståle Kopperud, who has contributed 281 commits in total and whose name rarely comes up in these threads. The fork contains 44 of those 249, and none of his.
So the architectures cited as evidence that AROS is progressing are, in large part, simply not in the tree being recommended.
Provenance.
A busy forum, a polished archive and enthusiastic users do not retroactively make upstream development the work of a fork. And upstream developers do not become answerable for a downstream community merely because both use the AROS name. Those are separate things, and the only record that distinguishes them is the commit history - who wrote what, when, and where it landed.
Why this is being raised.
Because this thread is not an isolated case. It is the same account of AROS that newcomers here are given repeatedly: that upstream is finished, that development happens downstream, and that the fork is where AROS now lives. None of it is true, and it doesn't survive five minutes with either repository, but it doesn't need to survive scrutiny when everyone in the room already agrees and no one checks.
That is how a project's work gets reassigned. Not by anyone lying, but by a small number of people repeating a convenient summary to people who have no reason to doubt it, until it becomes the version that gets passed on. The post above is that version being passed on again, by someone who says outright that he didn't know the history and had to be told it here.
If newcomers are going to be given "transparent information," give them all of it - including how this forum came to be the only one left, and why the people writing the code aren't on it. Don't take the consequences of a community that drove people out and present them as proof the project had died, and don't then use the space that left to claim its development history.
aros.org.
It isn't the upstream project's job to host an account of a downstream fork, which would necessarily mean hosting that fork's explanation of why it exists. If the difference between the lines needs documenting, ask the people who created the second one.
Use Deadwood's fork if that's what you want. Nobody is arguing otherwise. But stop calling it AROS, and stop presenting it as an official or sanctioned version of it. It is a downstream line, maintained by one person, containing a majority of code he did not write.
Edited by amigamia on 05-09-2026 10:36, 2 days ago
@amigamia: Your are the one threatening others for their opinions. I urge you to tell this to my ISP.
@amigamia - ... you have the power to correct them and explain (or better "prove"that what they are posting in here is incorrect or not based on facts. You are a member of this site just as they are...
In amending my own response, I want to take this opportunity to remind everyone that whatever you chose to post in here is governed by the TOA (Terms of Agreement) you agreed to when you registered your account. In particular this:
Quote
.. You agree not to post any abusive, obscene, vulgar, slanderous, hateful, threatening, sexually-orientated or any other material that may violate any applicable laws. Doing so may lead to you being immediately and permanently banned (and your service provider being informed). The IP address of all posts is recorded to aid in enforcing these conditions. Creating multiple accounts for a single user is not allowed. You agree that the webmaster, administrator and moderators of this site have the right to remove, edit, move or close any topic at any time should they see fit...
2 users reacted to this post
miker1264
olivier2222
DeremonMember
Posted 19 hours agoI don't know if it can help but I can bring my experience of what I saw when I was approaching AROS in 2022/2023.
A bit of background:
I started with an amiga 500 in my youth then I had to switch to a PC (DOS/Win3) cause of school and, well, all my friends had a PC so...
When I started working I started using linux and I had the chance to buy a "last stock piece G3 mac before G4" with OS9 and OSX these gave me an alternative to Win2k/XP BUT all I miss was the Amiga experience.
I had looked on Amithlon and AROS but I had no time to keep an eye on them at the time.
So comes 2022/2023, when I first looked back to AROS, I started from aros.org, I found the site (from a new user point of view):
- quite outdated
- snapshot wasn' t updated (last is dated 2017 even now)
- progress page was not updated from 2016 (as it is now)
- documentation was scattered around, on the current version is better organized
- all I was left with were nightlies and the distro page, now vanished
- at the time project feel abandoned, is better now with the updated news feed
So at the time in the distro page I found IcarOS I really liked it and then I found this forum.
IcarOS was working fine (but it was already been abandoned) while I had difficulties with nightlies.
Also I could not find software I needed at time (an updated scummvm in particular but not only that), I was tempted to abandon AROS and just stick with amithlon and the upcoming pistorm+emu68 (I still have my amiga, now pistormed) but my "system engineer part" wanted drill down and solve the problem (it's my work and I like it) so I started tinkering with abiv0, abiv1 and abiv11 where I discovered all the incompatibilities.
Before the "system engineer part" probably is were an "average user" end without an help.
But I kept on it and I started porting software, not what an "average user" will do, seriously
If I look at projects like Haiku (one can not like Haiku, it's fine, I'm not discussing of the OS) webpage is very well done even is not "bleeding edge html5", I have attached a screenshot with the "things that help new users":
- Easy to find links to download a stable version (more on that later)
- Easy to find links to a community forum (very important)
- Easy to find user guide
- Presence of a donation counter that help the user undestand that the project is *alive* aside from news feed
What I can do to improve the situation:
While I cannot contribute much economically (I'm trying saving a bit for that), I'm spending my free time to port things.
I started a project to "adapt" or port much software as possible from archives to 64 bit architectures and, with Kalamatee blessing, push it to mainline ports repo.
So, now the important part:
I think we are all circling the missing point, that from me is, AROS mainline severely lacks of a stable release.
From my point of view you cannot expect an "Average user" to download a nightly or distro maintainers to base their work on nightlies and also this is not expected from the description from the aros.org nightly page itself:
"
Note
- These builds are experimental and not for general consumption. Users should instead download one of the available AROS distributions.
- Binaries/Developer Tools are incompatible with previous ABI versions.
- The use of any information or materials on this page is entirely at your own risk. The AROS Dev Team accepts no liability whatsoever for any loss or damage incurred from using the files listed on this page.
"
It's a situation that in english could be translated by "to be between a rock and a hard place." a situation without a favorable exit or a circling situation.
What can be done from my point of view, and of course in this Kalamatee MUST be involved to in the long run eventually "re-unite" ABIs:
Fundamental:
- Understand what AROS should be, just an experiment for niche devs without any real world application (aside from people who take the code with their projects or sell it with a closed source project) or a *real* open source "amiga like" OS for the masses? If the reasponse is the former than we can just close the discussion here.
Very important (given the answer is the latter):
- Have an initial stable release ABIv1, graphics and icons may not matter now, stability should, it can pass different betas (like haiku do) but every subsequent stable version must backward compatibile or is not a stable version by my definition of stable. Something may break in nighlies but it should work in betas and stable.
- Have a defined roadmap like Deadwood do for his fork with milestones to be reached and periodically progess announcements.
- Bring the community to mainline site.
Important:
- Distributions based on the stable release and betas (yay nice icons!).
- Distributions brought back to main site.
- FAQ and easy docs for the "Average user" and below.
Nice to have:
- Donation to the project to fund additionals developers (payed devs) to help Kalamatee and Deadwood.
- Merchandise to fund the project? ( just an idea one may not like it but as latins said: "Pecunia non olet" ) .
These are my "two cents" I hope a point of view from someone who skipped all the dramas (even amiga ones) may help.
A bit of background:
I started with an amiga 500 in my youth then I had to switch to a PC (DOS/Win3) cause of school and, well, all my friends had a PC so...
When I started working I started using linux and I had the chance to buy a "last stock piece G3 mac before G4" with OS9 and OSX these gave me an alternative to Win2k/XP BUT all I miss was the Amiga experience.
I had looked on Amithlon and AROS but I had no time to keep an eye on them at the time.
So comes 2022/2023, when I first looked back to AROS, I started from aros.org, I found the site (from a new user point of view):
- quite outdated
- snapshot wasn' t updated (last is dated 2017 even now)
- progress page was not updated from 2016 (as it is now)
- documentation was scattered around, on the current version is better organized
- all I was left with were nightlies and the distro page, now vanished
- at the time project feel abandoned, is better now with the updated news feed
So at the time in the distro page I found IcarOS I really liked it and then I found this forum.
IcarOS was working fine (but it was already been abandoned) while I had difficulties with nightlies.
Also I could not find software I needed at time (an updated scummvm in particular but not only that), I was tempted to abandon AROS and just stick with amithlon and the upcoming pistorm+emu68 (I still have my amiga, now pistormed) but my "system engineer part" wanted drill down and solve the problem (it's my work and I like it) so I started tinkering with abiv0, abiv1 and abiv11 where I discovered all the incompatibilities.
Before the "system engineer part" probably is were an "average user" end without an help.
But I kept on it and I started porting software, not what an "average user" will do, seriously
If I look at projects like Haiku (one can not like Haiku, it's fine, I'm not discussing of the OS) webpage is very well done even is not "bleeding edge html5", I have attached a screenshot with the "things that help new users":
- Easy to find links to download a stable version (more on that later)
- Easy to find links to a community forum (very important)
- Easy to find user guide
- Presence of a donation counter that help the user undestand that the project is *alive* aside from news feed
What I can do to improve the situation:
While I cannot contribute much economically (I'm trying saving a bit for that), I'm spending my free time to port things.
I started a project to "adapt" or port much software as possible from archives to 64 bit architectures and, with Kalamatee blessing, push it to mainline ports repo.
So, now the important part:
I think we are all circling the missing point, that from me is, AROS mainline severely lacks of a stable release.
From my point of view you cannot expect an "Average user" to download a nightly or distro maintainers to base their work on nightlies and also this is not expected from the description from the aros.org nightly page itself:
"
Note
- These builds are experimental and not for general consumption. Users should instead download one of the available AROS distributions.
- Binaries/Developer Tools are incompatible with previous ABI versions.
- The use of any information or materials on this page is entirely at your own risk. The AROS Dev Team accepts no liability whatsoever for any loss or damage incurred from using the files listed on this page.
"
It's a situation that in english could be translated by "to be between a rock and a hard place." a situation without a favorable exit or a circling situation.
What can be done from my point of view, and of course in this Kalamatee MUST be involved to in the long run eventually "re-unite" ABIs:
Fundamental:
- Understand what AROS should be, just an experiment for niche devs without any real world application (aside from people who take the code with their projects or sell it with a closed source project) or a *real* open source "amiga like" OS for the masses? If the reasponse is the former than we can just close the discussion here.
Very important (given the answer is the latter):
- Have an initial stable release ABIv1, graphics and icons may not matter now, stability should, it can pass different betas (like haiku do) but every subsequent stable version must backward compatibile or is not a stable version by my definition of stable. Something may break in nighlies but it should work in betas and stable.
- Have a defined roadmap like Deadwood do for his fork with milestones to be reached and periodically progess announcements.
- Bring the community to mainline site.
Important:
- Distributions based on the stable release and betas (yay nice icons!).
- Distributions brought back to main site.
- FAQ and easy docs for the "Average user" and below.
Nice to have:
- Donation to the project to fund additionals developers (payed devs) to help Kalamatee and Deadwood.
- Merchandise to fund the project? ( just an idea one may not like it but as latins said: "Pecunia non olet" ) .
These are my "two cents" I hope a point of view from someone who skipped all the dramas (even amiga ones) may help.
Edited by Deremon on 05-09-2026 13:54, 2 days ago
2 users reacted to this post
sonountaleban
olivier2222
You do not have access to view attachments
Not treating anyone of their opinion. That is in extreme cases such as pornography, other abuse of such as race, religion or other illegal activities.
Edited by amigamia on 05-09-2026 11:18, 2 days ago
miker1264Software Dev
Posted 19 hours agoAlthough this is a forum to express opinions, in my case, sometimes it's better to keep quiet & keep working!
Very interesting discussion though.
Very interesting discussion though.
2 users reacted to this post
deadwood, sonountaleban
You can view all discussion threads in this forum.
You cannot start a new discussion thread in this forum.
You cannot reply in this discussion thread.
You cannot start on a poll in this forum.
You cannot upload attachments in this forum.
You cannot download attachments in this forum.
You cannot start a new discussion thread in this forum.
You cannot reply in this discussion thread.
You cannot start on a poll in this forum.
You cannot upload attachments in this forum.
You cannot download attachments in this forum.
Users who participated in discussion: amigamia, terminills, AMIGASYSTEM, cdimauro, miker1264, OlafSch, arosbeginner, Deremon, olivier2222, sonountaleban
