Awareness by classic Amiga users about modern Amiga's?

[stefcep2]

Karlos wrote:

It's mindnumbingly easy. It's the easiest software install ever conceived. You don't have to do anything. It goes away, downloads the package, sorts out any dependencies, installs it, configures it and unless it is something very low level, your package is immediately ready to use.

In theory.  How well the "dependencies are sorted" depends on the person writing the package.  That is not 100% reliable.  A 100k dependency ( within a total of 10 MB of dependencies) that wasn't updated when installing a 700k SNES emulator meant that I could no longer boot.  So much for "robustness".  I then dual-booted into XP, logged on the forums and found the solution eventually..how would an average user with one PC and only Linux at home do this?  In contrast my XPPro system has  not blue-screened once in 3 years, and never failed to boot and I've installed and removed all sorts of rubbish.  

The concept of installing from a repo is flawed because it depends on the writer, of the particular package you want to install, knowing exactly what the writers of the dependencies have done.  Thats impossible to be 100% certain when 1000's of little dependencies exist.  But thats what happens beacsue Linux is a collaboration of thousands of independent programmers in all parts of the world, each with their own area of expertise. Its ridiculous that a 700k emulator needs an additional 10 MB of additional Operating system dependencies-written by various authors, at various times, just to work!!!  10x more code in the dependencies than in the actual program: OFCOURSE there is a high probability of soemthing going wrong. Most Amiga software installs from commercial software were self contained: you might need a 3rd party library but that was rare.  Hell in many cases you just drag the folder onto the hard drive..

If you have the time and inclination to maintain your Linux OS by forever logging into forums, often to cut and paste text commands that you have no idea what they do and will never remember, then Linux is for you.  For the rest of us we just want things to work, and i'll gladly pay $120 for a commercial operating system upgrade for that.

[Astral]

DiskDoctor wrote:

I think marketing is the only way to hit any numbers easily.

Personally I think that when MorphOS (Or AmigaOS) is released on Macs Minis, that's a big chance to attack the huge potential user base - for the very first time.

I don't mean paid ads or something - that's pretty much silly.  I mean forums mostly, that's all.

I agree in principal with the marketing. Without exposure, people don't recieve knowledge of anything.

The exact method of the marketing and the timing is hopefully part of Hyperion/ACube's plan. Premature marketing, with consideration to the companies' economics, hardware base, software base, user base, and so on, may lead to failure to go forward.

I agree with your suggestion that MorphOS or AmigaOS being released for the MacMini will provide a "big chance to attack the huge potential user base". Whether this path is taken, I am sure we will see.

I also agree that forums are a good place to market new Amiga items to Amiga enthusiasts - it is cost effective, and you are marketing directly to an already established user base. Currently, this seems to be the main marketing avenue by Hyperion and ACube, and also supporting companies such as AmigaKit and Vesalia. But if the Amiga is to grow, there will be a point where different methods of marketing will probably need to be used so that growth doesn't stagnate.

It will be interesting to see the next move by Hyperion and ACube is, and at what point this is done.

[freqmax]

EyeAm/DiskDoctor, Linux work may be to ruin data. But I run FreeBSD and it has never ruined data for me , and is quite trimmed for being unix.
AmigaOS still wins the compact code contest There's one advantage too with AmigaOS exokernel model. You can access low-level hardware directly for a real speed boost. MMU can eat 3x speed.

Karlos, I wouldn't trust Mac OS X with any central filesystem. Mac is a desktop machine and probably excel at that, not as a server.

Also there's an security integrity advantage to separate central filesystem service and desktop and multimedia use onto different machines. In essence bad code have less chance blow things up.

Karlos, Nasty thought.. maybe one could design a "kernel sandbox" for 3vil NVidia/ATI drivers? , ie fake that they are in kernel land when they are infact not :-D

On the feature of AmigaOS. One modern application for AmigaOS can be to run heavilty trimmed portables. To kick Asus eee but 4 MB RAM demand rather than 512 MB to start with.

stefcep2, Rarely had that problem on FreeBSD. Other than having to watch install text for 30 different dependencies etc.

[persia]

The main reason Amiga's don't sell because of their outrageously high prices, their lack of real world software and the lack of an niche apps.  An OS is a program loader, people want to exchange songs with their iPod, update the schedule and contacts on their phone, open the .docx they were sent from work or friends.  The can't edit their personal videos or photos.  

No amount of advertising will overcome these problems...

[redfox]

@Astral

As you can see clearly now, there are many opinions available and a topic like this will always go way off on many tangents.

---
redfox

[redfox]

{ snip ... }

[DiskDoctor]

@ Karlos

Well I highly exaggerated on the first point.  That was a part of the sentence yielding Mac userbase is pretty good for AmigaOS introduction.  Being disputing this ambiguous and pretty casual sentence, it ended up with quarrel.  I also said if someone uses Linux and loses data, in 99% it's his/her fault.  To conclude - Linux is good for expert users, programmers mostly, been unlike Win users or even Mac users.  It's not much reachable Desktop system so I do not consider it a Desktop system.  Same argument like Win Server 2003 might serve as desktop but it's against its purpose.  Sorry for my language but I'm anxious on twisting my statements, especially those pretty casual and off the merit.  The only reason I quoted Linux is because many people did it here already.

I generally agree with all you said (now clearly enough and  conscienciously) and try to stay more cool hereafter.

@ Astral

Just to go beyond Mac forums posting.  I am an advocate to sacrifice several licenses in some big fish forums - e.g. for the most comprehensive review (MOS or AOS for Mini, I keep talking about Minis).

@ persia

This is not anymore since this expensive Sam dongle is no longer needed for Minis.  Also the system's price might drop which is really a key turning point here as they say.

There's more.  System's distribution throughout Apple downloads only (if this is legally possible to put it out there, unfortunately I don't think so).  No?  Well Hyperion or MOS devs should make some free Amiga demo / presentation with some usable features for macs (on top of E-UAE I suppose).  And then go for Apple/downloads with it.  More.  Some alliance with some popular Apple software vendor is needed to sell a packet.  Also some printed magazines maybe, i.e. limited-time demo CD (a demo OS is a must, again).

And so on.  Forums is just a cliche, not so difficult to expand, also economically.

[Piru]

@DiskDoctor

Well Hyperion or MOS devs should make some free Amiga demo / presentation with some usable features for macs (on top of E-UAE I suppose). ... Also some printed magazines maybe, i.e. limited-time demo CD (a demo OS is a must, again).

MorphOS is a bootable livecd since forever, and it will be the same on Mac Mini (why would you need UAE?) You can install it and test freely, for limited time per session. To continue testing you just need to reboot.

If you feel that MorphOS is worth your money then you can buy license online. The license keyfile will unlock the MorphOS for unlimited use on that particular hardware.

[Karlos]

freqmax wrote:

Karlos, Nasty thought.. maybe one could design a "kernel sandbox" for 3vil NVidia/ATI drivers? , ie fake that they are in kernel land when they are infact not :-D

It's an interesting idea. I strongly suspect that the drivers don't need to be in kernel space - after all most of the open-source one's don't. It would seem more likely that the drivers are ported from windows, where root access may be required to implement a particular feature that might not be supported under linux and tus they inherit this dependency. Of course being closed source there's just no way of knowing.

On the whole, I've not had any major problems with the nvidia drivers (currently using 180.44), but there's no way of knowing how robust they are. Running in kernel space, if they crash, they could bring down the whole OS. How's that for Amiga like? :-D

[Karlos]

In theory. How well the "dependencies are sorted" depends on the person writing the package. That is not 100% reliable. A 100k dependency ( within a total of 10 MB of dependencies) that wasn't updated when installing a 700k SNES emulator meant that I could no longer boot. So much for "robustness"

Again, this isn't really a criticism of the package manager, though is it?

I could write a bad install script for AmigaOS that could format your hard disk without even asking you. Is that the fault of the installer mechanism?

Most Amiga software installs from commercial software were self contained: you might need a 3rd party library but that was rare. Hell in many cases you just drag the folder onto the hard drive..

This is shifting the goalpost. Now you are comparing commercial and free software, not the mechanism of installation.

[HenryCase]

quarkx wrote:

Open source improves code stability because there are more people involved who are able to track down and fix bugs.

It also invites a lot of half baked code that causes more problems than what its worth. Any "Nut" can add code whether it works or not. There are no strict guidelines to go by, no quality control.
-But I am off topic.
Just my 2 cents

You clearly have no idea about how open source projects are commonly organised. Access to the source repository is limited to those who have earned a certain level of respect/trust. That trust is often earned by submitting patches, which would be reviewed before being implemented. Often case there are coding style guidelines too to keep the code readable.

stefcep2 wrote:
Its ridiculous that a 700k emulator needs an additional 10 MB of additional Operating system dependencies-written by various authors, at various times, just to work!!!  10x more code in the dependencies than in the actual program: OFCOURSE there is a high probability of soemthing going wrong. Most Amiga software installs from commercial software were self contained: you might need a 3rd party library but that was rare.  Hell in many cases you just drag the folder onto the hard drive..

Actually what you're talking about is one of the strengths of Linux, the libraries/dependancies are shared so HDD space is saved and work doesn't need to be duplicated as much. Libraries are a good thing, just because your emulator author chose to use them rather than bloat their own code doesn't make them bad.

[DiskDoctor]

Piru wrote:
@DiskDoctor

Well Hyperion or MOS devs should make some free Amiga demo / presentation with some usable features for macs (on top of E-UAE I suppose). ... Also some printed magazines maybe, i.e. limited-time demo CD (a demo OS is a must, again).

MorphOS is a bootable livecd since forever, and it will be the same on Mac Mini (why would you need UAE?) You can install it and test freely, for limited time per session. To continue testing you just need to reboot.

If you feel that MorphOS is worth your money then you can buy license online. The license keyfile will unlock the MorphOS for unlimited use on that particular hardware.

I know.  And what I was referring to in my broader expression was:

System's distribution throughout Apple downloads only (if this is legally possible to put it out there, unfortunately I don't think so). No? Well Hyperion or MOS devs should make some free Amiga demo / presentation with some usable features for macs (on top of E-UAE I suppose). And then go for Apple/downloads with it. More. Some alliance with some popular Apple software vendor is needed to sell a packet. Also some printed magazines maybe, i.e. limited-time demo CD (a demo OS is a must, again).

So when I said E-UAE distribution, sort of, I meant to unlock the Apple/downloads channel.  When I said demo CD, I wasn't considering MorphOS only.

[downix]

EyeAm wrote:
I can't say I'm surprised people don't know about those--I barely know of them, myself (like, where to get one, etc.). I understand enough that not one of those things mentioned is a full-fledged computer, but instead is either a motherboard or a card that goes into some slot; and that all are 'classic' (if that) in nature, and not a piece of it with a CPU that is AMD or Intel.

As far as steps to be taken, including advertising, etc., I won't offer any until I see on the Amiga Inc. website a press release that "Amiga will now be put on x86-64 CPUs" and run on CPUs by AMD and Intel. Til then, what's the point? (beyond this, I have extremely hostile words)  :-D

Why do you want to go for the SECOND most popular 32-bit processor group?  I say, go ARM, the single most popular CPU family on the planet (over 10 billion so far).  I'll always prefer MIPS, but got to face the world as it is now.  x86 is a monopoly, and unless your initials are MSFT or you have millions of man-hours to dedicate to OS development, you don't have access to the party.  ARM has dozens of chip vendors, dozens of OS's, lots of options, lot more flexibility.  Most importantly, opens up the future on our own terms, not those of Mighty Redmond.