AmigaOS 3.1.4 withdrawn from market after Amiga

[Romanujan]

@Hewitson - I don't consider it massively overpriced; the marked for this update is very small, this always rises the price.

And regarding the shell bugs... recent bug in M$ Windows was deleting files, recent bug in Linux kernel was damaging hard drives, and previous AmigaOS releases had serious bugs with such basic things like hard disk support :/

@kolla - Although you are fully right about this licensing issues, and I hope Hyperion will get spanked for this really hard in one way or another, you are exaggerating the problem with shell. Simply using older shell (3.9, 3.1, whichever) will fix the FailAt problem, but might as well introduce other problems - simply because such configuration was probably not tested at all.

[kolla]

Yes, using old shell fixes the problem for those of us who know how to extract the old shell from kickstart files and rom updates... it does not fix the problem for those who just install a kickstart and boot up floppies or use random scripts from aminet or wherever.

And regardless, how such a bug handled is more interesting than the bug itself - this nasty one is already more than two months old now.

[malko]

Originally Posted by kolla

Yes, using old shell fixes the problem for those of us who know how to extract the old shell from kickstart files and rom updates... it does not fix the problem for those who just install a kickstart and boot up floppies or use random scripts from aminet or wherever. [...]

On all the forums where I've read something, you are complaining about the same "failat" issue and with the same bitter criticism.

I'm not minimizing this issue but "much ado about nothing", your posts are worthy a Shakespeare sonnet... in less funny.

According to you this is such a "big issue" that :
- it annihilates all other OS 3.1.4 improvements
- the vast majority of the Amiga community is stuck and can no longer work. As only a few, of which you are a part, know...

Originally Posted by kolla

[...] how to extract the old shell from kickstart files and rom updates [...]

So, for the many users, whose interests you defend so strongly, unable to work with the 3.1.4 that they have bought spending expensive money, why don't you just post here, in a new specific thread, the step by step workaround guide on how to achieve the shell fixing. You will be a great help for ALL of them.

[kolla]

Well, first step is to remove the OS 3.1.4 kickstart and reinsert the OS 3.1 kickstart, so you have a rom to rip and extract from. Oh, you upgraded from 3.0 or 2.x? Sorry, you are out of luck. Ask Hyperion for a 3.1 shell, or buy 3.1 first. Or download from your favourite pirate site.

[Gorf]

Originally Posted by gulliver

There is a bug, the bug has been acknowledged, a solution has been found, and what I was trying to say, is that there are several ways to deliver the fix, not just one or two. And then the bugfix will be available in some form when we feel it is ready, not when someone else wants it.

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!

[indigolemon]

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!

Did you miss the 'when it is ready' part? I think you missed the 'when it is ready part' ...

[kolla]

So, around xmas times, as I said.

[Romanujan]

@Gorf - delivering a fix before developers and testers feel it is ready typically leads to even more bugs. Like in this poem:

99 little bugs in the code
99 little bugs in the code
Take one down, patch it around
117 little bugs in the code

I prefer to wait for 2 or 3 months and receive a proper fix - not a half-baked one, breaking who knows what. Especially that I know where the problem is, I can easily find all affected scripts, and I know about a temporary workaround which is probably suitable for 99.9% users.

[Gorf]

Originally Posted by indigolemon

Did you miss the 'when it is ready' part? I think you missed the 'when it is ready part' ...

No:
"There is a bug, the bug has been acknowledged, a solution has been found"

so apparently the solution has already been found...

And there is of course always the solution of delivering a version with the working old shell and re-include the new shell when it is really ready.

easy for all who purchased a digital download ... not so easy for customers with a physical ROM ... sending it back and getting a new version for free might take some time

[Gorf]

Originally Posted by Romanujan

I can easily find all affected scripts, and I know about a temporary workaround which is probably suitable for 99.9% users.

I know all that - it is just the way how all this is handled does not feel right to me.
Even Microsoft or Apple can no longer act this way and try to release fixes as soon as possible (and for free).

And to go back to the topic here:
Hyperion might not have all the time in the word .....

[roondar]

Originally Posted by Gorf

You also do know how to fix - but you not make that fix available for your customers.

I really don't want to get into discussion much, but you do realize that knowing how to fix something and having the actual fix in place are two totally different things, right?

For all we know it's much more tricky to implement this fix than we might think. And for all we know re-introducing the 3.1 shell has other potential problems that make this a poor overall solution. OS's are complicated stuff, you don't want to randomly change parts of it.

Furthermore, it's not at all clear how to me many people out of the total 3.1.4 'population' are actually adversely affected by this bug. Which is pretty crucial information for determining how big the priority for fixing it should be.

As I don't actually own 3.1.4, I can't judge this either.

[FOL]

Originally Posted by roondar

Furthermore, it's not at all clear how to me many people out of the total 3.1.4 'population' are actually adversely affected by this bug. Which is pretty crucial information for determining how big the priority for fixing it should be.

So what your saying is. You buy a TV, it wont work on one channel. So manufacturer says don't worry there will only be a few people who want to watch that channel. It should not matter.



I think the only solution is software approach, Hyperion will not pay for mass returns from customers to get replacement / reflashed roms. Sort of defeats the object of updated rom, if your back to patch resets.

[Romanujan]

Originally Posted by Gorf

I know all that - it is just the way how all this is handled does not feel right to me.
Even Microsoft or Apple can no longer act this way and try to release fixes as soon as possible (and for free).

Than rest assured, that before delivering any fix to the end user (even if it is reversing some component to the older release), they should be tested first. To be sure that it won't introduce other problems, to be sure that older component works well in the new environment, etc. Which takes time, sometimes quite a lot. Or else you will get another 'Windows 10 October 2018' quality update.

If the team would like to test the OS with 3.9 shell first (in order to provide an official quick-fix), we would have to wait for the proper fix LONGER. In some cases (bug causes risk to a valuable property or human life) it would be reasonable to go the 'older component' route first, to get the bug fixed ASAP. In other cases (retrocomputing system, not used for anything serious anymore, with very limited development manpower) to 'proper fix' route is more reasonable.

[roondar]

Originally Posted by FOL

So what your saying is. You buy a TV, it wont work on one channel. So manufacturer says don't worry there will only be a few people who want to watch that channel. It should not matter.

No, that is not what I'm saying. Try again.

[malko]

Originally Posted by kolla

Well, first step is to remove the OS 3.1.4 kickstart and reinsert the OS 3.1 kickstart, so you have a rom to rip and extract from. Oh, you upgraded from 3.0 or 2.x? Sorry, you are out of luck. Ask Hyperion for a 3.1 shell, or buy 3.1 first. Or download from your favourite pirate site.

This propensity to post non-constructive responses is quite impressive .

Originally Posted by Gorf

So you are saying, you are acknowledging that a faulty product was delivered to your customers. [...]

Originally Posted by Gorf

[...] release fixes as soon as possible (and for free). [...]

You tend to ramble.
Answer has already been provided. Strangely, like @kolla, you don't want to listen to it...

Originally Posted by Thomas Richter

First of all, let me say that I am sorry that I missed a bug in the shell, despite testing. [...] Strangely, this was not detected during testing, which still astonishes me given that the shell was under test for quite a while.
[...]
Will this be fixed? Of course it will. There will be an update, tentatively scheduled for Q1 2019, but as always, it will be ready when it will be ready. According to my information, this will be a free update for 3.1.4 owners.
[...]
I just don't want to let things stand as incorrect as they are.

@kolla & @gorf : do us a favour and go post your bad mood in another thread. Leave this one for news about the court case.

[kolla]

@ Romanujan

Especially that I know where the problem is, I can easily find all affected scripts

Really.

I know about a temporary workaround which is probably suitable for 99.9% users.

Please do share.

@malko

This propensity to post non-constructive responses is quite impressive

So what is your suggestion for a temporary fix for people who upgraded to kick 3.1.4 ROMs?

[Gorf]

@malko

this is me in perfectly good mood

And I think this matter is relevant and very on-topic:
as Hyperion might lose the rights to distribute OS3.x, the question of a bugfixed version is kind of urgent.

To be on the save side Hyperion should publish the updated source code.
(this does not necessarily mean with a copyleft permission - there are many more restrictive licenses if they feel the need)

[UberFreak]

Originally Posted by Gorf

To be on the save side Hyperion should publish the updated source code.

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

[hth313]

If there is a desire to release this fix, it may unfortunately be quite urgent to do so. A court may soon rule to stop any further distribution of this software. It is not for sure they may be allowed to distribute fixes after that... at least not for a while (one year).

[kolla]

Originally Posted by UberFreak

they can't release what they don't own.

Well, that's not an excuse they can use while also claiming full ownership.