What happened to Amiga Anywhere?

[nar001]

So I remember seeing a very old interview of two employees of Amiga Inc I think right before or after the 2000s talking about AmigaDE/Anywhere and how it could be portable, I think something about the Tao Group too? I was just curious what happened to that? I couldn't find any information about it other than the announcements I think. Thanks for the help!

[Radertified]

The AmigaDE Player was released. I used to play Boxikon on it all the time: http://www.amigahistory.plus.com/dep...ugust2001.html

[Hewitson]

Ironically Amiga Anywhere didn't actually go anywhere at all. Maybe they should have called it Amiga Nowhere.

[Fastdruid]

It was a great idea...that floundered and went nowhere.

http://www.amigahistory.plus.com/amigade.html
https://arstechnica.com/gadgets/2018...2-red-vs-blue/

[Thomas Richter]

Originally Posted by Fastdruid

It was a great idea...that floundered and went nowhere.

It was a stupid idea that went nowhere for a reason. I still have the manual in a box somewhere in the basement... This was nothing more than another "Java, me, too!" project that copied the idea of a virtual machine, though register based rather than stack based.

You are not successful as a small company by copying the idea of a big company while being incompatible to it - you will only run behind the market without the chance of overtaking them, less resources, less success.

It takes novel ideas to be successful. This one wasn't, just a "oh, let's create another virtual machine because Java is so nice but stack based".

I remember the time back then I was also asked to join the company, luckily I never did that.

[Fastdruid]

https://news.ycombinator.com/item?id=9808159

[lesta_smsc]

Originally Posted by Hewitson

Ironically Amiga Anywhere didn't actually go anywhere at all. Maybe they should have called it Amiga Nowhere.

that's hilarious.

[Hewitson]

Originally Posted by lesta_smsc

that's hilarious.

Thanks, I'm pretty sure that's the first time anyone has laughed at one of my jokes on EAB

[Megol]

Originally Posted by Thomas Richter

It was a stupid idea that went nowhere for a reason. I still have the manual in a box somewhere in the basement... This was nothing more than another "Java, me, too!" project that copied the idea of a virtual machine, though register based rather than stack based.

Pretty impressive to copy something long before that something existed. TAOS originated in the age of Amiga/Atari ST as a message passing portable operating system using JIT/AOT translation, JAVA originated as a stack based interpreter for embedded systems.

The big difference was that JAVA was a co-designed language and interpreter, TAOS was created by someone loving assembly language and wanted it portable. While there were high level languages they weren't the basis of the system.

TAOS was designed for multiprocessing in heterogeneous systems with translation from the portable format into machine code usually done while loading (as mentioned above ahead-of-time translation was also supported).

You are not successful as a small company by copying the idea of a big company while being incompatible to it - you will only run behind the market without the chance of overtaking them, less resources, less success.

See above.

It takes novel ideas to be successful. This one wasn't, just a "oh, let's create another virtual machine because Java is so nice but stack based".

See above.
The system was never based on JAVA or intended to only run JAVA, that they were successful in making JAVA run well on their system opened up more markets.

I remember the time back then I was also asked to join the company, luckily I never did that.

Luckily because of what?

TAOS and its descendents still have innovative features I've not seen anywhere else, the big problem being the lack of any type of protection. I guess that made it a more suitable fit to the Amiga philosophy...

[Megol]

Originally Posted by nar001

So I remember seeing a very old interview of two employees of Amiga Inc I think right before or after the 2000s talking about AmigaDE/Anywhere and how it could be portable, I think something about the Tao Group too? I was just curious what happened to that? I couldn't find any information about it other than the announcements I think. Thanks for the help!

Amiga DE was mostly a skinned Elate/intent. Only thing I'd like to try from the Amiga DE project is the Sheep language.

[nar001]

Was it ever released or open sourced? Seems it could've been useful if anything for small embedded systems?

[Thomas Richter]

Originally Posted by Megol

Pretty impressive to copy something long before that something existed. TAOS originated in the age of Amiga/Atari ST as a message passing portable operating system using JIT/AOT translation, JAVA originated as a stack based interpreter for embedded systems.

Look, I do not know which came first. Message passing is not a language feature, it is an operating system design decision which would have certainly be possible with other virtual machines as well - this is orthogonal to the VM.

Originally Posted by Megol

The big difference was that JAVA was a co-designed language and interpreter, TAOS was created by someone loving assembly language and wanted it portable. While there were high level languages they weren't the basis of the system.

That sounds like another bad idea: How many low-level assembler programmers are out there? How many high-level software engineers are out there? Looks like this product was driven by engineering, not by the customer, which is typically a bad idea.

Originally Posted by Megol

TAOS was designed for multiprocessing in heterogeneous systems with translation from the portable format into machine code usually done while loading (as mentioned above ahead-of-time translation was also supported).

What is the target application for that? Why is that a good idea, or why is it any different from Java?

Originally Posted by Megol

The system was never based on JAVA or intended to only run JAVA, that they were successful in making JAVA run well on their system opened up more markets.

It might have been a bit more successful if it would have used the java VM instead - instead of a custom VM. Who builds products on a technology of a small company with a big competitor with a technology that is sufficiently similar to be used instead.

Originally Posted by Megol

Luckily because of what?

Because it went bankrupt shortly after. Which was a very obvious development that was easy to predict.

Originally Posted by Megol

TAOS and its descendents still have innovative features I've not seen anywhere else, the big problem being the lack of any type of protection. I guess that made it a more suitable fit to the Amiga philosophy...

I'm not quite sure I would call that a "philosophy". It is a substantial shortcoming which makes the Os unsuitable for any type of modern application.

[Gorf]

Originally Posted by Thomas Richter

Look, I do not know which came first. Message passing is not a language feature, it is an operating system design decision which would have certainly be possible with other virtual machines as well - this is orthogonal to the VM.

tell that BEAM (Erlang virtual machine).

That sounds like another bad idea: How many low-level assembler programmers are out there? How many high-level software engineers are out there? Looks like this product was driven by engineering, not by the customer, which is typically a bad idea.

So WebAssembly is a terrible idea? Yet every major browser supports it...

No: the advantage is, that it allows you to compile other languages into that assembler - and the same was true for TAOS/ELATE

What is the target application for that? Why is that a good idea, or why is it any different from Java?

again look at Erlang and BEAM. There is reason why it is preferred over Java in some cases.
TAOS/Elate would have been somewhere in between: fast oder faster as Java but with more concurrency in mind, but not as extreme as BEAM.

It might have been a bit more successful if it would have used the java VM instead - instead of a custom VM. Who builds products on a technology of a small company with a big competitor with a technology that is sufficiently similar to be used instead.

Well that custom VM allows them to build one of the first Java-JITs and it stayed the fastest JIT for a couple of years.
They were asked to deliver the official JVM for BlueRay, but stupidly declined that offer...

I am not saying that TAOS/ELATE where a good fit for Amiga, as it had nothing in common - but the ideas of TOAS/Elate/Intent and the VM itself were not so wrong and there is no need to bash that technology.

[Megol]

Originally Posted by Thomas Richter

Look, I do not know which came first. Message passing is not a language feature, it is an operating system design decision which would have certainly be possible with other virtual machines as well - this is orthogonal to the VM.

(I use TAOS to describe everything up to the system under the "Amiga" thing)
The fact that TAOS came much earlier than Java undermines the argument you have repeated several times.
You are assuming a freestanding VM on an unrelated OS not a co-designed VM and microkernel. The whole idea was to have something that performed well on low end systems while providing a portable object oriented system using message passing, really look it up - it's still a radical idea.

That sounds like another bad idea: How many low-level assembler programmers are out there? How many high-level software engineers are out there? Looks like this product was driven by engineering, not by the customer, which is typically a bad idea.

I agree. That's the problem for anything innovative, how to keep compatibility to have customers. There was support for HLLs BTW.

What is the target application for that? Why is that a good idea, or why is it any different from Java?

The Java that came much later? Shouldn't we ask why Java was a good idea when there was something at least partially better available?
Anyway Java support was added and performed well, can't see the problem.

It might have been a bit more successful if it would have used the java VM instead - instead of a custom VM. Who builds products on a technology of a small company with a big competitor with a technology that is sufficiently similar to be used instead.

The Java VM was crap even when launched. But sure innovation is boring, making yet another clone is the way to go, hooray...

Because it went bankrupt shortly after. Which was a very obvious development that was easy to predict.

And yet they lived longer than the Amiga line (I don't count NG systems that aren't real Amigas). It should have been obvious that the Amiga was a failure*.

I'm not quite sure I would call that a "philosophy". It is a substantial shortcoming which makes the Os unsuitable for any type of modern application.

I agree hence my apparently failed attempt of irony.
(* next attempt)