@Carl-S

Quote:

One other positive distinction about the Amiga is that of being a lightweight system

You've definitly struck my chord here with the synergy of mating lightweight and creativity. I began in the late 80's as a flegling musician with a midi interface and a C64. Those simple tools let me an my companions create music and tracks, that once tooks days or weeks, in a matter of hours. Later came Amiga/BarsNPipes/NewTek and the skies opened. I began to do multimedia for fun an profit. Nothing on any other platform compared.

Cutting the story as short as possible, later I had to move to PC equipment. Much of the fun left as energy went into learning bloated apps, buggy OS's and the best way to back up work as not to loose it in a crash. Things have improved some since XP, but I keep holding out hope Amiga can let me get back to creating and spend less time tinkering and recovering. Other OS's were always much "heavier" in maintenace needs than Amiga. Besides, I never could part with my Sunrize cards

Plaz

Last edited by Plaz on 24-Oct-2006 at 02:02 AM.

@itix

Indeed, it may not be Amiga, considering the complexities of the situation (legal, technical, and financial).

But, you cannot rule out a new influence. If we know anything about the world, we know it is always changing.

_________________
When replying, please keep it short and sweet, and I will read it. If it is long, I read only the first few lines to decide if I want to read the rest.

@tomazkid

Quote:

tomazkid wrote: @wegster Quote: he idea of handing each student a USB flash drive, with OS, apps, and personal data on it, is appealing, and could re-use the above hardware, if/were OS4 ever to be run on x86. You mean similar to these ? Seems they have managed to build Mac-On-Stick since I last checked I myself am using Firefox, VLC, and Gimp from USB flash drive, really nifty

Note I didn't say it's not already possible. People can, and have been using, Linux on USB, as well as Solaris for single sign-on, etc, for some time now. The point being made is there are specialty distros out there to allow something similar, today. Whether or not it's enough 'added value' to be of use in Carl's scenario...is debatable.

'Mac on a Stick' is insane and inappropriate to even bring into the discussion- it's MacOS Classic, under emu...yay? Not something I'd ever want to do work on, today..ok, or then, either

_________________
Are we not done with the same silly arguments and flames yet??!

Not sure if I could contribute anything useful to the thread, but I'm compiling
a linux kernel so there is time to think .

Back in the days when the Amiga was released it attracted many people that
never were into computers before. It was just amazing what you could do with
with this system, things that were not possible before or with other systems
existing at that time or price. Though, a lot of people remember the amiga for
games, also a good thing.

If this could be done again, I mean create a system that "enables" people
to do new things not possible yet, things people never would have thought a
computer could do and in an easy elegant way at a reasonable price, not only
for the sake of being different, but better in a way. If it can surprise
people, make them smile and make them feel this is a good product to buy and
to use, something that many people want to have. Then I believe in a return.

I have not really a clue if and how this could be done in our days with a computer
again, but this is what the Amiga did to me and maybe many others back then.

Remember this (slightly modified) one?

Carl,
as much as anyone
has made the Amiga
a success, you have!

Thanks, Markus

@Carl-S

Welcome back to the Amiga community! I use OS4 every day as my OS of choice
because of the very things you mention. The ONLY things that need to be
improved are low cost and millions of users.
I sincerely hope that Samantha helps to take care of these two points!

Carl, do you have an eyetech machine running OS4? Have you experienced the
Amiga like it was in the old days, only much faster and more up to date?

I hope you do!


Scott

_________________
AmigaOne uA1-c 512M RAM - Only Amiga Makes It Possible!
Check my blog AmigaOne Computing

@Carl-S

Quote:

Carl-S wrote: @wegster Yes, perhaps a specially "mixed" small Linux could do it too. It would need to be really smooth. But, the issue is, the solution has to be a "platform" -- something concrete, defined, and supported, because the school is not of a mindset (nor ability) to do that themselves, nor do the want to deal with the chaos of the open source domain when it comes to things like support.

Sure. Enter the plethora of companies that are out there now offering just that- support for open source alternatives. Lindows (ok, whatever they renamed to), RedHat, Novell/SuSE, MySQL, and others.

I don't know if DSL, as an example, has a company out there willing to support it, but it _could_ be done. Updates to users' individual 'systems' (on flash/USB) would/could be a challenge, or they could simply add an init script into the system to go and check for, and install, updates, at each system boot.

Other 'challenges' would present themselves, such as ensuring the bootable system hasn't been hacked to allow 'additional priveleges,' (if small enough, could use something like tripwire, or md5sum config files, etc), but those would also exist for other solutions..


Yes, it would be possible for a company to build some kind of Linux that satisfied those requirements, but do you really think that Linux is good enough for most students? I use a dozen Linux distros here, and I've not found that really special one yet.[/quote]
It depends. Without any formal requirements, we're really just guessing back and forth
If the requirements were, say:
- able to be updated easily - with some work.
- ability to store user data on flash or USB device - check, mountable filesystem
- ability to boot on all school systems of configurations X, Y, Z for immediate use - ok, custom kernel and modules, or monolithic kernel
- ability to provide modern web browsing experience - firefox
- ability to provide word processing, spreadsheet applications - OpenOffice.org
- ability to limit the addition of additional applications - some problems here, but initially, just disallow/don't give out root access (can still be hacked..), similar problems would exist for other systems as well, though.
- ability to ensure some/all sites/URLs are blocked outside of a subset, or local network only. - Ok, use school nameservers and http proxy. Same for other OSes here..
- Ease of use - big black hole of unknowns, need more info
Others ?

So again, it comes down to, what would OS4 be better at, than a system similar to the above, or what problem is OS4 a solution for? Most of us on AW love 'Amiga,' and/or AOS, but that doesn't mean I'd willingly deploy it at work, and expect people to be able to do their jobs on it, currently. Given more limited requirements, it becomes more of a possibility, but there are currently few areas IMHO, that OS4 'beats the competition' on, sadly, including the 'small footprint' if you look at DSL.

That isn't to say it _can't_ happen, only that, at a minimum, additional development work would be needed, and ideally, a specific set of requirements be targetted, to make it a more feasible option, whether for the scenario you outlined, or some other, possibly not yet known/discovered, option or 'problem to solve.'

Edit-
Note, I'm not disagreeing with you here, as sure, there's 'potential' for OS4, but instead wondering what OS4 does, or possibly, CAN do in the future, better than other alternatives, to make a scenario like yours compelling, where OS4 is 'the solution.'

Last edited by wegster on 24-Oct-2006 at 02:22 AM.
Last edited by wegster on 24-Oct-2006 at 02:20 AM.

_________________
Are we not done with the same silly arguments and flames yet??!

@Carl-S

Wow, first Dave Haynie and now Carl Sassenrath... this place is getting better very day

I voted that I agree, however I must disagree when it comes to price. Ideally, yes, but reality is very different and has always been (in fact, in South Africa where I grew up, us poor kids had C64s and Speccys; the rich kids had STs, Amigas and Macs). When I bought my first Amiga (an A1200 in 1993), I had to do a careful cost-benefit analysis of it vs a 386 PC, because the price didn't make it a slam-dunk. I'd also been lusting after an A500 since practically the day they came out, and the reason I didn't buy one was also price - I was a teenager with a choice: keep using my C64 and have a social life, or spend about a year working weekends, saving everything I earned, and watching the other kids have one. I chose life

Amigas and their 3rd party addons have always been more expensive than equivalent peripherals for other systems, and this seems to be an acceptable fact of life (case in point: someone was selling, just a few months ago, a wireless ethernet card for something like 5x what you'd pay for it from a PC store - the exact same hardware!)

So, I agree with your basic premise. I disagree that it's going to happen any time soon. The reasons for that are social rather than technological though.

@Carl-S

It was nice seeing you at AmiWest again this year, your recolections of the early days at Amiga were interesting and entertaining. Funny, the last time I went to AmiWest was when you also were there, I think in 99. I did not get an opportunity to talk to you at the show this time. Sorry for being off topic (I did vote yes), but I would like know how well Rebol is being received and how many users and developers you think are using it? I believe your philosophy behind Rebol is viable if you can get enough momentum behind it to reach critical mass. I have not researched it much yet, but I was intrigued by your description of it being able to use words in different ways depending on the context they are used in.

Signature under construction

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@Carl-S

I voted no, because...

I first will say that the current (or soon comming again) situation
is that we have OS4 on PowerPC and that is meant to be a user
and developer-platform for the existing Amiga-community.
There are plans to port it also to the comming gameconsole PS3
but those are only plans for now and it is not sure if it will happen.
If the PowerPC hardware (systems/embedded + OS4) sells good
then it can be sold to a larger new- and ex-amiga-user-base.

Now, why i said "no" is because the defenition of an Amiga has
changed. You can not say that it has to be a completed system in
the sense that you have to buy new hardware + OS. The future
lies in the fact that you can take your OS every where you want
and that you can plug it into whatever system is out there.
For example take an older computer out your basement (one with
usb) plug in your OS-usb-stick, start your own OS and of you go.

Now, i am very interested in your REBOL IOS. Lets say you could
write OS4 in REBOL. Then you would have an OS that is very portable
and very light. You could use it as a stand alone system or you could
use it as an Internet-OS. You could also use it with remotedesktop or
some other internet way of running your applications and data.

So, how much would an USB-stick + an OS cost me as a user?

Dirk.

.

Last edited by Dirk-B on 24-Oct-2006 at 04:21 AM.

_________________
A1G3-SE + OS4.1 u1 iso (x2)

@Thread:

Best discussion since ages there.Haven't we got more arguments like this one to discuss freshly?

Thumbs up to the Topic Starter, and respect from me for the way he made the Amiga back in the day.

_________________
Cuida tus piedras gordas.

A1200/030 32Mb
A4000D
A1-X1000.

@Carl-S

Quote:

Carl-S wrote: @falemagn I understand your points. Well stated. Yes, those systems have come a long way! I use Linux, BSD, OS X, and Windows daily, but none truly satisfies me. Only OS X gets me close to the Amiga feeling and concept. Linux and BSD both have their merits (e.g. I use them for servers, firewalls, etc.), but I don't find them that friendly when used as a desktop system.

Others have already partly responded to that, and you partly agreed with them, but let me give my contribution as well.

As I myself stated, those systems more or less cover your points, OSX even comes close to fulfill all of them, but none of them really could be called an "Amiga", in my opinion, for the following reasons.

1. OSX still uses most of the paradigms used by classic MacOS, and some (most) of them never struck me as "intuitive". OSX is an OS you still need to get used to and learn how to do things with. Yes, once you do it, you may become productive with it, but it would still feel "Alien" to me, in some ways. This is not to say that AmigaOS wouldn't pose the same problems to someone who'd never used it, but inuitivity is part of my personal definition of "Amiga" (which seems to be a synonym of "ideal computer", as someone else says). However, for someone who's never used a real Amiga before, I'd say that OSX plus an apple machine do all that the old Amiga's did and much, much more.
   
   
2. Linux, to all the above issues, adds the one that most of the time the user has to find her way around the OS. You need to know how does Linux work "behind the curtains" in order to accomplish certain tasks, and this is because all the work that has been done on Linux so far, in the ease-of-use department, has been to try and hide the complexity under the carpet, but that means that you not only need a quite thick carpet, so to not make it show bumps here and there because of the hidden complexity, but just try to move it a bit from its original position and there you go, the complexity comes out again, because it was never really gone.
   

1. distributed - build your own super computer, uniting the computing resources of as many amiga's as you want


2. totally componentized - mix and match various components and build your own applications.


3. totally scriptable - components would be automatically scriptable because the system would know their type and thus their interfaces, making it possible for any script engine to control them.


4. tightly integrating the shell with the components - it would be possible to "cd" into a program's window and invoke its methods as if they were executable programs. Controlling the execution of a program, in every one of its aspects, should be doable with just the help of a simple shell script (Monads of the Microsoft fame comes to mind, but it's something I had already thought of a couple of years before MS came public with that idea )

1. reimplementing the linux kernel's system calls in terms of AmigaOS library functions
   
   
2. base AmigaOS upon, say, L4 and have linux run alongside AmigaOS on top of L4.
   
   
3. port usermode linux to AmigaOS.
   

Last edited by falemagn on 24-Oct-2006 at 08:28 AM.
Last edited by falemagn on 24-Oct-2006 at 07:49 AM.

_________________
“It is well enough that people of the nation do not understand our banking and monetary
system, for if they did, I believe there would be a revolution before tomorrow morning.”
~~ Henry Ford

@Carl-S

_________________
retired

@Tesla

Quote:

It might just be because I am interested in processor architecture and think that assembler is something which all programmers should know (even if they might not program in it),

In any processor architecture discussions, one should factor the external Instruction set and the micro-architecture implementation e.g. out of order execution vs in-order execution, dynamic branch vs static branch, branch table size, deep pipeline vs short pipeline, the amount of instruction issues/ instruction retirement and 'etc'.

There's no overriding need for application programmers to know the assembler language. One should focus on doing the job quickly and receiving income.

Quote:

but I feel that PPC, MIPS and possibly SPARC are a better fit than x86 (I am not mentioning ARM since I have not yet had the opportunity to examine it closer).

Have you considered infrastructure that surrounds the CPU (e.g. distribution channel, core logics, motherboards, ODMs/OEM/ISV/IHV support and etc)?

Quote:

For many the processor is irrelevant, but for others PPC (or MIPS etc.) is a better choice than x86 and perhaps that is enough to warrant a moments pause. For me at least, x86 is the less enabling choice.

Why x86/x64 is the less enabling choice? What market are you targeting?

At least in X86 market, there’s a strong mobile/desktop/server market competition between AMD and Intel relative to PPC market.

MIPS family of CPUs are been repositioned to embedded market that targets specific workloads.

Last edited by Hammer on 24-Oct-2006 at 09:45 AM.
Last edited by Hammer on 24-Oct-2006 at 09:39 AM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

@Carl-S: while I think this is a good start, I think this definition needs to be precised, with the following points...

- Name: You are referring to the "Amiga" all along your definition... Is the name important ? Should it be called an "Amiga" ?

- Compatibility: Should it be backward compatible with the Amiga (Hardware and/or OS) ?

I think these two parts are important... it is important because these two points add two important constraints. And I think that's what kept the Amiga "alive" (well, sort of), and at the same time, that's what prevented the Amiga from evolving...

Leo.

Last edited by Leo on 24-Oct-2006 at 09:35 AM.

_________________
http://www.warpdesign.fr/

@Carl-S

Quote:

For me, I reached a point where I just imagine the x86 instruction set to be a form of microcode... which is always ugly and difficult to read, but no one really cares.

There’s no need to imagine i.e. all modern X86/X64 processors translates (hardware and* software emulation**) X86 ISA to internal RISCy/VILW*** like instructions prior to execution.

*Hardwired X86 to RISCy instruction translations.
**Micro-code decoding engine that fetch translation parameters from firmware (on-chip).
***pack/bundle of instructions within an instruction issue.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

Go easy on me, I'm trying to rationalise past history here!

I always wondered where the "we must move to PPC" momentum came from.

Going back in time, my guess is that it started for one of two reasons:

1. Parallelism with Apple.

Apple used the 68k. Apple moved to the PowerPC. It's a very thin-on-the-ground guess, but since Apple developed a 68k emulation for the PPC System series to run 68k software, everyone saw it as the upgrade path of the Amiga.

2. It was the next generation of Motorola CPUs.

Brand loyalty. Everyone loved their 68k and thought that the PPC was the one true path.

One of these two reasons started a motion of people producing PPC accelerator boards for Amiga systems. My impression is that this motion was carried forward until the decision was that this should become the next hardware platform. Once you've got a community revved up for such a shift, it's hard to shake them from the target.

I hear people quote complaints about the PC architecture and say how inefficient it is, how much of an ugly hack it is - it really isn't any more. Intel CPUs hold the most amount of legacy cruft (does the Pentium 4 still have the A20 hack? And start execution at $RAMTOP - 16 bytes?), but AMDs are mostly cruft free once we have seen the back of BIOS once and for all. The impression I get now is that more people want the OS than they want the full platform to go with it (see the number of people willing to run out and buy one of the two platforms recently announced with fairly low speed CPUs for an example of OS over platform).

I'm not asking anyone to port anything, or give up any plans. I'm just platform agnostic, and I'm interested.

@Carl-S

Quote:

So, there is a perfect opportunity for an alternative OS, but instead... all those computers will be replaced. What a huge waste.

I admit its a waste, and XP isnt a solution.

Wouldnt it be easier to install good AV software and ban Internet Explorer and Outlook Express? Arent they the biggest vectors for spyware/adware/virii?

Roll forward a few months/years, there will be someone who will want to take down a room/school/world full of computers if they can. Amiga (on x86 or anything else) doesnt solve that (yet!?) - or only solves it as much as using a Mac.

Security through obscurity isnt a solution.

@nine

Quote:

nine wrote: Go easy on me, I'm trying to rationalise past history here! I always wondered where the "we must move to PPC" momentum came from. Going back in time, my guess is that it started for one of two reasons: 1. Parallelism with Apple. [..] 2. It was the next generation of Motorola CPUs. [..]

Pardon me, but there's more to this.

At the time this decision was made Motorola was very supportive of Amiga's intention to go with the PowerPC. They also tried to push the ColdFire, but at the time the binary compatibility issue was still unsolved (some ten years later it's no longer an issue). I was present at the meetings with the Motorola folks, so I'm not making this up Motorola wanted the Amiga to use their CPUs.

Then there's the issue of what kind of CPU you could buy, and what it would deliver. Back then you could still take your pick: MIPS, Sparc, Hobbit (don't laugh), PowerPC, 68k (including the 68060 and the ColdFire family), ARM, Alpha, SH-3/SH-4 and several i386 flavours. Now go through the list: MIPS and Sparc were rather expensive for the kind of market the Amiga was intended for, same goes for the Alpha. The 68k family was going the way of obsolence. ARM and Hobbit didn't necessarily deliver the kind of computing power that was needed (even Be, Inc., which had originally used the Hobbit CPU, went for the PowerPC). And the binary compatibility issues for everything but the PowerPC would have been deadly. At least Motorola knew that they had to offer binary compatibility to successfully migrate their 68k clients to the new platform. Sure, we struggled with the 68k emulation idea, but at least we knew that it was possible to move it all to the PowerPC.

Last, but not least, the overall PowerPC architecture wasn't so ugly either. Compare this to what the i386 legacy brings with it today and weep.

In total, I'd say that going to the PowerPC made sense then and still makes sense today. The reasons for choosing this platform and no other are not as simple as you may have concluded.

@Leo

Quote:

- Compatibility: Should it be backward compatible with the Amiga (Hardware and/or OS) ?

As I see Carl's definition is more like a vision than feature set specification. What ever is seen as the best effort to reach this vision (i.e. low cost) should be used... big endian CPUs probably arent the first in queue...

_________________
Amiga Developer
Amiga 500, Efika, Mac Mini and PowerBook

@Carl-S

I agree with that definition. That's what an amiga was and should be. It's not what it's today as it's expensive and doesn't have many apps that allow you to be creative.

I will add that just like AmigaOS used datatypes to interchange information easily in a object oriented way and the custom filesystems like crossdos or crossmac allowed you to access easily data, any future amigaos should have as a priority that it should be easy to be able to interchange and access information between our "amiga" and other systems.

That includes being able to access and edit .doc/.xls/.ppt files, being able to access the same webpages, being able to watch the same media files...

datatypes were a step in the right direction but unfortunatily no amigaos flavour has really evolved in that way.

MorphOS is the only amiga-like system with a competent datatype/codec system (Krashan's Reggae) but unfortunately the number of classes is too small to attract developers. Another downside (and probably Fabio will agree on this, although Grzegorz is the author and he's free to do whatever he wants with his work) is that Reggae isn't open source so it won't be adopted by AROS/AmigaOS3/4 in the future.

MorphOS had clear plans in the right direction, as they designed it in a way that all AmigaOS compatible software would run in a sandbox and future applications and games would run in the unveiled "Q-Box". Unfortunately the Q-Box API didn't evolve enough and being unfinished, the public docs didn't were published

OTOH AmigaOS4 coders prefered to concentrate on compatibility and decided to improve the old API. They are adding improved memory handling but it's obviously limited by the nature of AmigaOS. But it allows future programs to be more stable than in the old amigaos (although not as secure as in a real memory protected environment)

Each path has his advantages and disadvantages. Unfortunately no path is using "cheap" or "affordable" hardware so most of people won't even try it out. And most of developers won't create software for these platforms as the user base will be really small due to the high price and poor features of the hardware.

IMHO a release for mainstream machines like PSX3 may make sense (and thanks to its big endian cpu it would have the advantage of being compatible with old software) to provide funding and a nice platform. I'm not against a release for little endian machines like x86 or AMD64 but we should keep in mind that little endian cpus won't be able to run old legacy software so we won't have much software to offer. If PSX3 version sold well the user and developer base may grow enough to acceptable levels. Maybe like 2001. Of course most of future code for the OS and apps should be created endian-agnostic.

_________________
The only spanish amiga news web page/club: CUAZ