AmigaOS 3.1.4 withdrawn from market after Amiga

[Romanujan]

Kolla, if despite being present on this forum for over 10 years (and more on Amiga.org), and despite writing well over 1500 posts here (plus over 3500 on Amiga.org), you are not able to checks which scripts on your disks use FailAt, than I don't think I'm able to help you

[Gorf]

Originally Posted by UberFreak

Repeating this every other post won't make it happen, they can't release what they don't own.

if they sell if, they must think they own it - there is no rule, to only distribute a compiled version.
(and even "compiled" could mean all sorts of things...)

and repeatedly mentioning something worked before:
https://en.wikipedia.org/wiki/Carthago_delenda_est

[SpeedGeek]

Originally Posted by UberFreak

Repeating this every other post won't make it happen, they can't release what they don't own.

WTF, if Hyperion only released what they owned then there would be no court case.

[AMIGASYSTEM]

Originally Posted by Romanujan

Kolla, if despite being present on this forum for over 10 years (and more on Amiga.org), and despite writing well over 1500 posts here (plus over 3500 on Amiga.org), you are not able to checks which scripts on your disks use FailAt, than I don't think I'm able to help you

I'm not a developer but the Shell bug does not only affect scripts that use FailAt.
As I mentioned earlier, OS3.1.4 does not stop in the presence of any wrong command.This in my opinion very important, means that if there are errors in starting a game or a program you will never know why.

[kolla]

Originally Posted by Romanujan

Kolla, if despite being present on this forum for over 10 years (and more on Amiga.org), and despite writing well over 1500 posts here (plus over 3500 on Amiga.org), you are not able to checks which scripts on your disks use FailAt, than I don't think I'm able to help you

All scripts use it, default failat is 10, and that is currently ignored. You gave the impression that you have something that will work for everybody, so what is it?

[Romanujan]

I suppose you have some older OS 3 release then? You already know what to do

[daxb]

In OS3.1 S-S there is "FailAt 21". What means that return codes up to 20 don't break the script. What means S-S is designed to not stop if any error happens. So you need to remove the FailAt command from your S-S if you want to break S-S while executing.

I used FAILAT 21 in three ARexx scripts (only for DOS commands) to not break the script while executing. Error handling is done by me of course.

I don't know. Exists a situation where a script should break instead of doing a proper error handling?

[Gorf]

Originally Posted by daxb

I don't know. Exists a situation where a script should break instead of doing a proper error handling?

It is useful for sub-scripts that get started by another script and should exit if certain conditions are not met...

Well - an in general you can usually not foresee every possible error, so you would make some error handling for a couple of cases but break if something else is happening.

[AMIGASYSTEM]

Originally Posted by daxb

In OS3.1 S-S there is "FailAt 21". What means that return codes up to 20 don't break the script. What means S-S is designed to not stop if any error happens. So you need to remove the FailAt command from your S-S if you want to break S-S while executing.

I used FAILAT 21 in three ARexx scripts (only for DOS commands) to not break the script while executing. Error handling is done by me of course.

I don't know. Exists a situation where a script should break instead of doing a proper error handling?

Hi daxb, in my test with OS 3.1.4, i entered the wrong command at the top of the startup-sequence, before setpatch and before the command Failat 21 and the startup-sequence did not stop, with other OS this does not happen

[gulliver]

Originally Posted by Gorf

So you are saying, you are acknowledging that a faulty product was delivered to your customers.
You also do know how to fix - but you not make that fix available for your customers.

Only at some point in future, when you "feel" it might be the right time there might be a better release for your paying customers ... but not if your customers want it, because ... how dare they - stupid users!

I dont have customers. I dont sell anything.

I am as you say just a "stupid user" who got bored of waiting for things to happen and decided to put some time and effort to see something actually happening in the AmigaOS front.

I am truly happy you endorse open sourcing AmigaOS. But it is not my call, nor my property, and there is absolutely anything I can do about it in any remote way.

The good thing is that there are still quite a lot of other operating systems that can be used on an Amiga that are open source and do desperately need people using/supporting them:

OpenBSD which has been unmaintained since 2002
Minix that has been unmaintained since 1.5.x days (aprox. 1992 ?)
Linux that has been struggling to revive its port for over 2 years.
NetBSD that could use more developers and testers to have broader hardware support.
And then you also have AROS, that if you dont like the unix kind of approach, then fits the bill. And they also need both users and developers to progress further.

So if you are not happy with AmigaOS as it is, dont worry this is not a monopoly, there are plenty of options to pick up from.

But everything means hard work and involvement.

Again, I am not selling anything. I am just another "stupid user" that wants to put its effort where its will is. Nothing more. I bet you can do the same.

[Gorf]

Nice list of Open Source OSs - I tried them all on the Amiga back then.
They were even still supported on the Amiga long after other hardware was much faster.
But now it just makes absolutely no sense to use an Amiga (real or emulated) for Linux or BSD.

Well - the OS3.x branch was now unmaintained for almost 20 years - the Kickstart itself even longer... so the closed source approach did not help at all.

[gulliver]

Originally Posted by Gorf

Nice list of Open Source OSs - I tried them all on the Amiga back then.
They were even still supported on the Amiga long after other hardware was much faster.
But now it just makes absolutely no sense to use an Amiga (real or emulated) for Linux or BSD.

Well - the OS3.x branch was now unmaintained for almost 20 years - the Kickstart itself even longer... so the closed source approach did not help at all.

To be fair, no approach has helped. We are all still in kind of survival mode no matter the flavor.

In the Amiga domain, open source failed and closed source too. Just see how we are doing right now.

Not only that, but the sources to AmigaOS 3.1 have been leaked to the wild in 1994, more than two decades ago, and no one even recompiled them or made a tiny change at all.

IMHO, the issue is the lack of active users and developers and not the source model we use in each variant.

We have both source models coexisting in the Amiga world, and yet we still see the light getting dimmer, so the issue is certainly not there.

Where do you think it is?

[kolla]

By "Linux" you mean Debian, I presume?

What also doesn't help is the bickering between the companies that claim to own amiga.
If there were one company I would help them with funding if they made new hardware. A-EON tries but I would have to sell my house to own one of their systems...

[daxb]

Originally Posted by Gorf

It is useful for sub-scripts that get started by another script and should exit if certain conditions are not met...

Sorry, I wasn't clear. I meant break by accident (failat) or break by code (proper error handling). You can exit a sub-script by code (even with own return codes) and IMO that is the better way. With FailAt you only set break at level <n> or higher, but you don't know what error and where the error happens. Of course against something strange happens you can't do anything. Hence my question. FailAt is no error handling, right?

[Gorf]

sure - you can do better error handling.
The point is: FailAt was and is used.
The shell was now behaving the same way since 30 years.
3.1.4 was supposed to be a bugfix release with some small improvements.

Is was not supposed to be 3.2 or some 4.x on 68k.
So we were told by the developers.

the missing FailAt functionality is acknowledged as bug by all parties.

It should not be required to change a single line in your shell-scripts.

[gulliver]

Originally Posted by Syntrax

What also doesn't help is the bickering between the companies that claim to own amiga.
If there were one company I would help them with funding if they made new hardware. A-EON tries but I would have to sell my house to own one of their systems...

Yes, I totally agree, the fight between companies does not help and further tires/divides users.

But trying to see the good side of it: at least at the end we will have a single entity that without any doubt will hold all legal rights (whatever they are worth).

And I also agree that we need affordable and better performing hardware to push any Amiga OS ahead.

[wawa]

Originally Posted by gulliver

We have both source models coexisting in the Amiga world, and yet we still see the light getting dimmer, so the issue is certainly not there.

Where do you think it is?

the issue is that the people offer themselves to support the companies and sign the ndas or look at the stolen sources and are therfore unable to contribute to open approach even if they wanted.

we are debugging aros boot on vampire as you talk. but then i wish you good luck with the closed source initiative, anyway.

[Gorf]

Originally Posted by gulliver

To be fair, no approach has helped. We are all still in kind of survival mode no matter the flavor.

That is not true.

Linux is doing great - as base of Android, as THE server OS, and and and
It has become the most used OS in the world!

BSD is still doing good on servers and firewalls.

Minix3 is now deployed in every new motherboard with Intel-chipset (even if most customers will ever know...)

In the Amiga domain, open source failed and closed source too. Just see how we are doing right now.

the software, that helped the most to keep things alive in the Amiga Domain ist of course open source: (Win)UAE

Mainly because we can not really use modern hardware - except for AROS.
And AROS was and is suffering big time from excluding everyone, who has had contact with the original source-code..

Just imagine how far AROS could be now with all the people and experts, that instead worked for OS4 or MOS (or now classic again...)

Not only that, but the sources to AmigaOS 3.1 have been leaked to the wild in 1994, more than two decades ago, and no one even recompiled them or made a tiny change at all.

probably very few managed to do so, because of the missing dev environment.
How many secretly succeeded? Who knows - no-one was allowed du distribute anything... even worse: AROS had to the extra care and doing things differently on purpose, just to avoid any accusations

IMHO, the issue is the lack of active users and developers and not the source model we use in each variant.

Please look at 3.6 and 3.9 Some parts are now not longer accessible for developers - that is why we have now a discussion about 3.1.4 and not some 3.9.9

And what will happen now, if Hyperion loses in court?
2 years of development lost?
All for nothing? And if maybe Cloanto wants to do an update, they have to start again at zero?

We have both source models coexisting in the Amiga world, and yet we still see the light getting dimmer, so the issue is certainly not there.

AROS was clearly hindered by many factors as mentioned above.
And they also did choose a non-standard license but also relying on many GPL-parts ... sadly it is quite a mess in this regard.
The deal once made with MOS did not turn bring back much benefits either...

That is why a releasing OS 3.x under a OSS-licence would still be very useful and could even bring back some developers.

Where do you think it is?

In still not working together!
We need all hands on deck - in one single ship.
(And the only chance to get there is actually open source and copy-left, as this new legal war now proves)

[gulliver]

Originally Posted by wawa

the issue is that the people offer themselves to support the companies and sign the ndas or look at the stolen sources and are therfore unable to contribute to open approach even if they wanted.

we are debugging aros boot on vampire as you talk. but then i wish you good luck with the closed source initiative, anyway.

I see your point, but I kind of disagree in that everyone is not a developer and each of us have different tastes that we should persue in our own way. That is the good part of having different choices to pick from.

Good luck with the open source initiative too.
And I mean it! We need diversity to survive.