Read more to vote for the operating system that should have been the Next Big Thing (TM), but that never happened for whatever reason. Please make sure you vote an OS based on its *technical capabilities*, not because you used to be its user when you were 16, or because you like its background color, or because you heard good things about it once. Vote for the one that you truly think it was technically superior at its time.Note to Web Masters:
Please do not link to this article. We are not interesting in altering the results of the voting poll by linking this page and directing a very specific crowd to vote for the OS you advocate. OSNews attracts a very large crowd from all OSes, so please let this poll be as close to the larger public opinion as possible. Thank you.
Note: The poll is now closed. Thanks to everyone who voted.
The company renamed to Apple and bought Apple (hehe, i know it is the other way round), while renaming NeXTSTEP as Mac OS X.
while renaming NeXTSTEP as Mac OS X
You’re right about one part – OSX was largely based on NeXTSTEP, but I wouldn’t say it was just repackaged and renamed to it.
So I voted for NeXTSTEP, even though it’s not technically dead. :O)
OSX is not as NeXTSTEP/OpenSTEP used to be though. OSX might use the guts of NeXT/OpenSTEP, but they have changed a whole lot of things, and neither it is binary compatible, neither it runs on 68k or x86 anymore. OSX is a brand new OS, based on principles and some of the API of NeXT/OpenSTEP. And the similarities stop there. It is like saying that the original Unix is not dead, simply because we have FreeBSD and Solaris today. The principles are still here, but the original Unix OS, is not.
Based on this facts, NeXT/OpenSTEP are dead for years now.
I wanted to vote for Be, but IMHO at the time Amiga was more ahead of its time. First is Amiga, second is BeOS, third is NeXTSTEP (and I claim ignorance on the other OSes).
Most people voted for BeOS cause they’re newbies to the computer world.
The veterans went and voted for OS/2 because quite a few things we liked so much in BeOS came from the OS/2.
If only OS/2 had won instead of WinNT…
Seems to ‘be’ a popular choice. Why? It is fast, light, has a decent simple UI, has a usefully innovative filesystem, has a unixy subsystem but isn’t hampered by full-on unix underpinnings like (IMHO) Linux (on the desktop) and Mac OSX to a lesser extent are.
I thought about OS/2, but having recently tried it out again, I wasn’t so impressed by the WPS as I thought I’d be. It is too abstract/separated from the underlying file system (hence the Shell in Workplace Shell I suppose). MacOS<X had it better conceptually I think.
If the people who voted for OS/2 were veterans, they would know about MULTICS.
Regards
I voted on Amiga, not because I still own a Amiga 500 from when I was 14 but because the platform (NOT just the OS) was revolutionairy. Amiga had very good and decent hardware with a good idea behind it, something we all see now in PC’s. Delagate control! The Amiga had a special architechture, every item has / had it’s own chip. PC’s are slowly catching on to the principle (Videocard are a good example). It’s to bad that the Gravis soundscards didn’t lift off they are also very good (architechturally).
Amiga just used 8 Mhz and run like the wind…
For me Amiga stopt being Amiga when they when bankrupt. Amiga isn’t just software and it isn’t just hardware; it’s both.
Based on my experience out of all of those, it had the most promise for me. Now I’ve never used Amiga or most of those, I’ve used OS/2, and never saw the love.. I’m using OS X now but never used NeXT so..
I voted for Be – it is the one I know the best and the most about, except for NeXT, but that only on 68k NeXT hardware. It was so painfully slow. I also wish I knew more about OS/2. I’d buy it if IBM didn’t charge so much for it. But, I really do think Be is the one. Built from scratch, so fast, so lean. I know programming is hard for it because of its complete multi-threading, but it was/is beautiful. At Apple, with Steve whispering in his ear, Amelio rejected Be and said, “It can’t even print”. Well, it was printing a long time before OS X did <g>.
If you don’t have js (for whatever reason) and you just wanta see the poll results you can sneek up on this link.
http://www.go2poll.com/cgi-bin/nph-pwnet.pl?name=eugenia&id=4&sessi…
Seems there is still issuse, cause even after i got a js browser up and running it never submited my vote. Oh well it was gona Be for Win3.11 any way – oh wait thats not a choice.
I voted BeOS because it was SO good compared to the others in it’s time but only lacked industry support and non-shareware apps.
This is the OS I would reinstall NOW if apps and drivers were available.
Close second was AmigaOS, but as good as it was, it started to stagnate well before Commodore went under since they had such a hard time breaking away from the A500 as a base system to write apps for.
Still, it was a really hard choice. AmigaOS ruled the 80’s and early 90’s easily.
Pretty good decision to exclude any OS you can go out & buy with ease that is, otherwise the feel of the resultss would be utterly swamped out.
Anyways I still use BeOS much more than Win for my HW-SW design work, it forces me to keep to stds so I can throw it over to Linux without having to deal with that right now. Still use Win for entertainment or if forced to, but only for a few apps.
If I wasn’t so busy on my own projects & the other BeOS projects hadn’t started (OBOS, Crusoe, winBe etc) I might very well have been tempted to start a BeOS desktop Explorer replacement for Windows (like PowerDesk), not just a Yellow Tab skin, but a really feels like Tracker project. Use Cygwin for the shell, maybe use wx or Qt for the Finder/Explorer engine & combine with some skinning SW to get the look right.
It wouldn’t be able to actually run any BeOS apps as I wouldn’t go so far as to include the Be APIs but the dev SW under Windows is better anyway than the BeOS dev SW & would substitute pretty well. As a bonus I would get the thoousands of things missing from BeOS. Of course if VMWare & BeOS worked together, that would be fine too.
The Atari ST had kinda two OS’s
TOS the Tramiel OS which did all the grunt work
GEM from digital research which does the GUI
This had a very similar interface to Windows 95. So similar that I picked up Win95 very easily when it was released
I didn’t vote Atari as it had no technical merit.
I didn’t vote OS/2 because there where better ones
I didn’t vote BEOS cause Iv’e never seen it
I voted NeXt Step because of the brilliance of its design.
Mainly the object orientated framework that made it easy to make apps on it. Not because it came in a sexy looking box or had a pretty GUI.
Good poll Eugenia
PINK / TAOS / VIC20 <- pick one
AFAIK OpenStep ran on HPPA/SPARC/x86/68K. Whith one nfs server and a configured netinfo automount one could have all its binaries on one server. Apple hows a patent for FAT binaries, such bins can run on multiple arch thats cool. Darwin when built is FAT and can either run x86 or PPC.
—
http://islande.hirlimann.net
hard to vote! os2 was the first real multitasking protected GUI on the 386+ that was near uncrashable, but then plan9 beats it hands down for concepts and ideas but they never felt very workable.
you got things at all ends of the spectrum!
in the end I picked the amiga because it was revolutionary. beos, next, os2, etc, were not. they were great, but not innovative.
i think, the original macos should have been in there tho, regardless of being able to purchase it now.
Although the last multics system was <A href=”http://www.multicians.org/corby-letter.html“>retired in 2000, it directly pioneered several very important OS systems concepts that gradually found their way into more mainstream OS’s. For example, it was the first to be designed with support for virtual memory, hardware pagetables and segmentation rings(one of the reasons why chips like i386 support both segmentation and paging), smp, dynamic linking, dynamic memory management, access control, hierarchical fs, programmed in PL/1—a high level language, etc. Unix, a pun on multics, of course, borrowed and simplified several of these concepts (for ex., no hardware memory protection on the original PDP’s on which Unix was first implemented—that’s right, the early days’ unix was comparable to the early days’ windows), although all the above pioneering concepts slowly wound up being added to unix. I think the chief impediments to multics’ popularity was simplicity, hackability (through being more cheaply available compared to multics), and fastness (for example, being programmed in C instead of in type-safe HLL’s). It is quite instructive and amusing to find that these are roughly the same reasons that led to windows’ dominance years later.
Multics wasn’t THAT good. I love Vax/VMS – OpenVMS, but still, OS/2 was one step behond.
IBM made OS/2 as a business OS, like they did AIX. Still, OS/2 infrastructure was great, OS/2 could have been so much more.
I use Windows, Linux and BeOS. I’ve never used another OS. And till this day I’m happy that I find link to be.com and ordered their trial demo disk for 10$, and v4.5 after that. Nither Windows nor Linux gave me the same feeling be did.
I have only read about it once, where it was called “developper OS” – is there something to it like BeOS? I thought it is largely an inhouse thing of Bell (or whoever..)?
1) During the 80s and early 90s PC manufactures and salesmen often stated that GUIs were useless and only meant for people who can’t type.
Today every mainstream OS has a GUI! The Amiga color GUI was the most advanced one around at the time, it even allows several screens with multiple resolutions and color depths to be displayed simultaniously. (Sadly modern graphic cards lost support for this feature mainly because MSDOS/Windows never supported this). And since 1985 AmigaOS offered amazing still unmatched 32-bit pre-emptive multitasking between a gigantic amounts of different programs without any noticable slowdown. To make everything complete Amigas also were the first computers to be able to display photo realistic graphics with up to 4096 colors simultaniously since 1985!
2) During that period PC people also stated that stereo sound wasn’t very useful as well and that the internal peeps gives enough feedback to users.
Nowadays everyone I know has a very advanced soundcard offering great music and soundeffect.
AmigaOS had since day *1*, support for great stereo sound as a standard, it doesn’t matter how old your Amiga is, if you have the horsepower you can still listen to MP3s with 14bit (near CD) sound quality (or get an additional sound card for even more quality).
3) Furthermore all Amigas offered shared library support, autoconfig (plug&play), multiprocessing (several custum co-processor chips), genlockable graphics (sub-titling or add other graphics to movies easily), very advanced CLI with case insensitive long file names support, etc.
But what still very much appeals to Amiga users today is its efficiency and ultra small memory footprint. Without a doubt AmigaOS and Amiga hardware were miles ahead compared to the competitors.
1) OPENSTEP
Actually, OPENSTEP 6.0 (or 6.1 for Puma, or 6.2 for Jaguar). The architecture remains the same. It has been updated and widened. Emulators (BlueBox/Classic), compatibility layers (Carbon) do not change a system. And, about not being binary compatible: that is because 68k and x86 binary code cannot run on PPC. The .app bundles structura has changed slightly, but you wouldn’t need to recompile for an old bundle to work, just change the internal directory structure. About Darwin not being the same UNIX that was under NeXTSTEP, just think that OPENSTEP 4 would run *on top* of the NT kernel. No Unix underneath, but Windows NT. Later, at the time of Rhapsody (aka Openstep 5) the same technology survived as Yellow Box for Windows. So I would say that OSX *is* OPENSTEP, but not Openstep/Mach or Openstep/NT, neither Openstep/SPARC. Think of it as Openstep/PPC v6.
2) PLAN 9 FROM BELL LABS
Plan 9 was an experimental OS that instead of considering the single machine as the logical unit for computing, it considered the whole network. That is, you would simply have a terminal (they have their own terminology, I just don’t remember it) and then there would be file servers, printing servers, processor servers, and so. All the communication (process, data, *all*) went thru the same protocol: 9P. And all the resources are represented as files. Theres also something about namespaces and so… Quite a revolutionary thing that will be ripped off by Microsoft and Apple and will then get to the main public. It source has been released by Bell Labs. Just in case you didn’t know, Plan 9 from Outer Space is the name of a sci-fi movie (I think it’s by Ed Wood).
Hi.
I wanted to vote, but I didn’t see linux in the list, which IMHO should be there in the first place…
but BeOS.
First I voted BeOS as I use(d) it since R3 but I’ll be on MacOS X soon.
What I don’t get is, why do you list Irix and AIX? Both are still available and development for them goes on. Both OS won’t target to the customer market, so they are not wide spread. Anyway my O² is running great (also my BeOS-boxes (x86)) and it’s no prob to get new releases of irix (AFAIK the latest is 6.5.15… I’m running 6.5.10).
Irix and AIX are not “made” for that list.
Oh, btw NextStep runs great at my old computer too !
Peace,
LoCal
I voted for other. I’m still waiting for the Great OS. BeOS was good, but I think the original Be hardware was better. I lost interest in Be when the stopped making hardware. Does anyone remember the original Be machines? I work with a guy that still has a blue DUAL 133 MHz PPC with the blinken lights and the Geek Port(tm) in his cube. We call him “The Kevster.” Be never made a single processor machine. When they stopped making hardware I lost interest. 🙂
— LadyJessica, geek girl.
I think BeOS will win this poll because a large part of this sites audience (first) came here from BeOS related sites. I’m sure Eugenia would verify that the first significant proportion of visitors had arrived from the BeOS sites that she and others had been involved in.
Having said that, I voted for BeOS for all the right reasons, i.e. it was a step in the right direction with a clean modern micro kernel design geared towards the increased capabilities of it’s targeted hardware platforms. It was (should I be saying is?) very responsive with plenty of room for expansion (the great FS, multiple processor aware, mime type handling, translators etc). It tried to look forward, unfortunately the masses need an OS that will deal seamlessly with their past too.
As always, it’s the lack of apps that kill and OS, WIndows has a rather large number of apps, BeOS didn’t. Maybe they should have bundled loads more apps with the OS 😉
“Pretty good decision to exclude any OS you can go out & buy with ease that is, otherwise the feel of the resultss would be
utterly swamped out. ”
In that case, AmigaOS is excluded. There is no difficulty in buying
AmigaOS 3.9, and by next Easter there will be no difficulty in buying
OS 4 with an Amiga computer.
plan9 was just sheer genius: recursively defined user-space namespaces. i can union my directories over other directories, mount my own filesystems without being root, access osnews through /net/tcp/www.osnews.com/index.php, etc…
the power and flexibility of this idea still hasnt been realised by anyone in the main-stream, although the Hurd team grok it.
To use it is to love it!
BeOS Rules!
I wouldn’t have excluded Amiga, never used it, but I appreciated it’s contribution in the 80’s. At the hight of AmigaOS time, I fell for MacOS instead, much more subtle even if only BW & no HW for anything. Anyway, they both used 68K so they were clearly far ahead of Wintel for quite some time. My boss at the time was an Amigan.
Anyway, I wouldn’t know where to buy Amiga kit unless I was a fan with some general knowledge of that scene. Its the same for BeOS, OS/2 as well. Most of the other OSs would be very hard to come by unless you still had the HW. As much as I was a Mac/PPC fan, I would be much more interested in trying out AmigaOS if it ran on x86. If Be had never done the x86 port, they wouldn’t even be on the poll & would have no significant following today.
If I had burnt even more $ than I already did, I would have owned NextStep, luckily they priced it way out of my reach.
OS/2 was a wierd set up. Like BeOS it took several versions & several PCs before I could even boot it. Once it warmed up though, I just couldn’t get it, so opposite to everything Mac.
Anyways, I am sure the OBOS, Cosmoe, othere teams will be pretty chuffed with this poll.
I think AmigaOS has proven itself to be the fartest ahead for its time. MacOS X and WindowsXP have copied alot of features originally only found in AmigaOS {but without any efficiency and far less freedom for the user}.
BeOS was good, but seriously it was always lacking behind. If it were not for DTP support then it was for 3D support. It could have been a good basis for MacOS X, but BeOS completely lack good software support. AmigaOS still has much better DTP software available like Pagestream or good 3D support with Warp3D.
I do all my serious programming on an A4000T (purchased in 1998). Even running at 50 Mhz on
a 68060, it appears to be just as fast as the Compaq presario (Celeron @ 600Mhz) for most tasks,
& much faster in others (such as text editing. Are all windows text editors purposely slowed down or
what?). The main reason I love it? I know where everything pertaining to the OS is located, & I
know what to do to fix just about any S/W problem. Jim Steichen, author of AmigaTalk
Hmm, I really wanted to vote for several, based on the time frames they appeared in…
Early: AmigaOS – A micro-kernel based real-time-lite multi-tasking OS for the masses…
Middle: NextSTEP – well, might still happen in a round about sort of way, but was clearly revelutionary.
Late: BeOS – AmigaOS on steroids, but with a more modern design and a file system that was(still is?) ahead of its time.
It had to be simply for it’s usability.
Mayuresh
AmigaOS
Spare, clean, elegant, fast, fast, fast.
Just enough, not too much.
No computer has impressed me nearly as much as unwrapping the Amiga A1000. An amazing experience. I have fond memories of all night game tournaments in college with the sound pumped through my stereo system.
BeOS
Windows NT 3.51 (the last Cutler OS, very stable/fast)
QNX Neutrino
BSD 4.4
Linux 2.5
OS X
#m
Really cool, even if in early stage:
http://uuu.sourceforge.net/
I even still got a Nextcube that I use sometimes.
I absolutely agree that linking to this article desirable, but I also think there are relatively a large percentage of ex-BeNews readers reading OSNews as that website died and Eugenia moved from BeNews to OSNews. I think that could be a factor for BeOS getting so many votes. Don’t you think so?
I personally used a lot of the OSes (MacOS,AmigaOS,
TOS/MiNT/MagiC,Nextstep,Irix,RiscOS,BeOS) – I love
diversity. So its hard to decide, its surely depends
on the timeframe:
In the beginning of the 80s MacOS was a milestone,
being the first with an appealing windowed user interface.
The Amiga ruled the homecomputer market in the mid 80s
technically wise (multitasking, multiple screens), but
gets not much credits for its usability. (IMHO)
In the end of the 80s and beginning of the 90s, Nextstep
was faaar superiour to everything – featuring display
postscript, a great object oriented user interface,
micro kernel technology and dozens of other unique
features – just ideal for productivity.
When the Acorn RiscPC600 appeared in 1994 it ran
RiscOS, featuring a lighting fast and super eye candy
user interface, antialiased fonts (!), multithreading,
drag&drop par excellence, multi processor support etc.
No Question – it was far beyond its time refering to
the home computer market! In many aspects it could easiely
compete with BeOS, when BeOS was released.
So I voted for RiscOS (maybe Nextstep deserved the vote
too, but it still lives inside MacOSX).
…as a os/2 user, but BeOS never was, and os/2 was murdered by MS.
OS/2 was amazing to me. The most powerful database software ever created (which many of the same ideas were used for creating oracle) were invented on os/2. Openbooks was an amazing system and couldve really changed the server market back in the early-mid 80s. Too bad venture capitalists kill companies with vision =(
1. NeXTstep
Clear, sleek design. Great Interface. Fast, even on my 25MHz NeXTstation Mono NonADB. Fine tools for development.
2. plan9
Strange, but really interesting design.
3. AmigaOS
…because it was *so* different at its time.
A hard choice, both have really interesting features..
Still Plan9 is not really dead, and some of those feature have percolated to The Hurd, FreeBSD (which have also unionFS).
Some won’t which is a pity because Limbo(the language not RedHat ‘s distro) is very nice too..
I voted BeOS, as it is the true AmigaOS 4 to me.
I was a die-hard Amiga user until I first saw the BeOS 4 demo CD in 1999. It had everything I ever wanted from AmigaOS 4. I ordered BeOS the same day, and a month later I sold my Amiga 4000 (which had everything – all-SCSI drives, Picasso IV, 64MB RAM, lots of software) and got a dedicated BeOS x86 computer.
I voted for Amiga, it was ahead of time………
Amiga was way ahead of its time technically, which is another proof that superior technology is really not enough to win in the marketplace.
I couldn’t read a detailed explanation in the 44 comments so far, but maybe I missed it, so forgive me if this is a repeat. Amiga had in 1984 :
* An OS consistently using a very small number of abstractions
* A single address space, message-passing OS
* A fully preemptible ‘sort of kernel’ (exec)
Of course this came with many trade-offs (including no protection for applications, like MacOS, and not possible to build it later into the system without losing the performance). But sassenrath’s consistency in design was admirable.
It still failed though
I say BeOS. Simply because in this day andage it could have hit the widest audience.
OS/2 was nice at the time. But the best thing I can say about it was that is was functional. Like a big diesle truck.
Amiga was great, but you have to account for the hardware.
BeOS was functional and beautiful, all on off the shelf hardware. The only place it was really lacking was the netwroking. But the general population is running high traffic web servers. Under normal day to day use, BeOS R5 would stay up for weeks.
OS/2 was functional, but ugly. Amiga was great, but ran on proprietary hardware.
Can’t comment on NeXT, because I wasn’t making that kind of cash when the NeXT cube came out.
Although I love BeOS, and still use it as my day to day OS. The fact that we ‘suffer’ a Wintel monopoly and not a Commodore monoply was one of the most tragic mistakes made in the computer industry so far.
Than people actually brought 2000UKP IBM’s over more capable 500UKP Amiga’s with equivalent software, before the big software lock in we have now, is something that will always disappoint me.
Since there are so many multi OS fans out there, I would offer to give away original Warp OS/2 & Tenon Mach OS? (Older Mac is fine) just for the shipping, otherwise just sitting around. The Tenon was quite a value but I will never use em.
I agree with you, most costumers are so easily brainwash-able. Good marketing sadly is far more important than having superior technology. Looking back who really still believes that MSDOS was *cutting edge* as marketed?
I voted Plan9, it failed but to me, was the must revolutionary OS I’ve ever read about. The only one who really tried to make the dreamed paradigm of “everything is a file” a truth. The Hurd is just trying to get the same thing done with a diferent aproach, but I wonder what a solid base would have Plan9 been to avoid having those nasty things as having a copy of libraries an programs in a separate directory to have ftp funcionality without compromising root, or things like the gnome virtual file system. How much extra code could have been avoided with such a basement.
recursively defined user-space namespaces. i can union my directories over other directories, mount my own filesystems without being root, access osnews through /net/tcp/www.osnews.com/index.php, etc…
recursively defined user-space namespaces? can someone explain what that means? I am not a programmer, but i do know something about OS technology …
fred
Even though I’m a hardened BeOS nut, I still believe that Amiga was destined to be the one. There were just so many innovations which came with the Amiga, things that we take for granted today, but were revolutionary in the 80’s (I wont mention them all, but I’m sure Mike Bouma will 🙂 Had the big C= had more financial sense, who knows how different the PC landscape would look like today, and how much more revolutionary it would be. The biggest WOW I had in the computing world came from the first A500 I saw. 12 years later, BeOS provided the second WOW (notice how the second WOW isn’t bold).
I’d have to say OS/2 for PPC. The betas were very nice and fast but it never got out of IBM.
I’ve voted BeOS.
I still know the day I got an illegal copy of BeOS 4.0 and installed it. Wow. Bye Bye Linux. Now I’m only using Windows (for the stuff not available on BeOS like Delphi, VirtualDub and VideoEditing (Although I have Personal Studio)) and BeOS on my PC.
I ordered (a legal) BeOS 4.5 the day it was available (as did I for the 5.0) It still feels the right OS to use.
To use it is to love it.
Maybe if I was involved in PC earlier I would’t have know the power of AmigoOS. I sense the same love for the system among it’s users as the BeOS user.
‘BE the difference that makes a difference’ – JEWEL
I voted for AmigaOS not because it was more advance when it was released, but because it is more advance at this very moment.
None of the others actually offer me the flexibility and the easy of use of AmigaOS.
So the vote goes to the best OS ever.
Actually I voted for BeOS. On reflection I think I should have voted AmigaOS… since when you think about it, what else was there? DOS – a joke; Mac – innovative but it wouldn’t have multitasking for another 15 YEARS. I’m talking about the A1000 release year, 1985, when I was 2 years old…
But I really believe that BeOS deserved to grow and be at least successful to sustain itself. It was (is?) a technically brilliant OS, a perfect desktop/user OS. I’ve used linux and linux only (read: not booting windows) for weeks at a time, trying to switch; I configure everything nicely, which takes ages, I run my commercial CAD program and freeware dev software on it, in fact I’m using it now for some reason – I don’t know why – but I always end up going back to windows to do any real work. I hate to say it. My windows install does not crash, yes it is slow, but does not ever crash (win98 first edition on a PII/300 w/256MB), and it seems like less effort to do my productive work in windows rather than linux. I always end up using the shell to do EVERYTHING in linux, which is not bad all the time, but the GUI parts of linux are very frustrating/inefficient to use.
Linux is a pain to do any real desktop work with, let alone adminster. As they always say, an average joe desktop user with linux still needs a personal sysadmin to maintain it.
So, linux is out of the question… It’s my opinion that if AmigaOS had been continuously developed over all these years, there would have been no NEED for BeOS at all; but as it stands, BeOS was the most refreshingly elegant OS of them all in recent years. There was of course OS/2, but well… it was IBM and ‘all
I can’t comment on NextSTEP etc. so I claim ignorance if anyone thinks those others were better…
– Paul
millions spent on those OSes that never saw the light of day…
I voted for NeXTstep, despite being one of the old-guard BeOS folks. At the time, NeXTstep was a HUGE step forward compared to all of the other systems, even Macs. BeOS was simply mostly-current state of the art, done without the legacy baggage.
– chrish
Most of the OSes listed on that page were quite good. NeXT had the enormous benefit of being based upon Objective C, instead of C++. But I still voted for Be.
Current operating systems are absurd. Think about it for a second. Windows CE devices often come with 64MB of ROM. That’s including the OS installed on flash, which the user never sees; all the user does see is a directory of applications, and a directory of documents. And Windows CE is the most bloated of the embedded OSes; Palm OS can fit in 8MB, and Symbian, AFAIK, is similar. I’ve seen dumb terminals with Windows CE installed. They use a web browser with the same rendering code as Internet Explorer, but it’s WinCE.
So how come the only three OSes with measurable desktop marketshare (Windows, Mac OS, and Linux) all require gigabytes of installation space, how come most applications require hundreds of megabytes, and how come both install huge, convoluted directory trees that should be invisible or non-existent?
Since 1990, Be is the only completely-new desktop OS developed from the ground up. Most of the other OSes on the list were based on UNIX, and the few that weren’t were inferior to UNIX. Be is the only one that’s better. There’s also Palm OS and Symbian, but those were influenced by Be (or at least, influenced by the same things Be was influenced by).
Unfortunately, BeIA is dead. But luckily, what it stands for is now. Due to the proliferation of mobile phones, Symbian may soon be the most widely-used OS in the world, moreso than Windows. As the desktop slowly dies, and the laptop shrinks, demand for embedded devices will grow, and Be’s ideas will live on.
AmigaOS could have been the ONE, if Commodore wasn’t strict, like Apple. Both companies could have done better if they allowed other companies to sell clones. Since the PC market was wide open everyone could build their own PC. This resulted in a nice competative market with interesting prices for users. Mac and Amiga didn’t have that. It wasn’t just the marketing it was also the fault of their ‘greedy’ creators…
veterans went and voted fore multics, you whippersnapper! 😉
Those are all nifty OS’. Except for Atari ST. That was a computer. Aside from the hybrid GEM/TOS (which was based on CP/M-68K) that came stock, there was also Minix, MiNT, and possibly MagiC (I’m not sure if that’s a different OS, and update or just a desktop replacement like NeoDesk). Oh, and maybe Oberon.
Actually AmigaOS still lives on, but at a much smaller scale and soon you will see AmigaOS4 running on ordinary off the shelf PPC clone hardware. Note that when Commodore bankrupted Amigas were still selling incredibly well in Europe. In some countries like in Sweden, Amiga had market shares up to 90%! Also there have been high-end AmigaOS clones, like Draco towers, which were mainly used for professional video editing.
What does make a huge difference however is that there were dozens of PC companies promoting their (inferior at the time) solutions, while commodore management and marketing did not focus solely on Amiga solutions.
This only changed dramaticly when they were close to being bakrupt, for a short while all resources were set behind Amigas, for instance the top basketbal team in the Holland changed their Commodore shirts into Amiga shirts.
According to insiders Amiga could have saved Commodore if they would have been able to manufacture more hardware during the last months before the actually bankrupting.
…technologically superior to Windows 95, and 98 in most repects, plus an integrated office suite with features that did not appear in another commercial office product until Office 97.
It was a shame that this one went away. I suspect if the resources had been available of Geoworks to truly develop this product to it’s full potential, it would still be giving MS a run for it’s money.
Despite being a long-time Amiga fan, I voted for BeOS. For many reasons that I have read in the postings here I believe that they are technically tied.
It really is difficult to say what might have been or what should have been.
I was a really big NeXTStep fan. It was a great OS and it pains me to see how slow it is now on Apple hardware, considering how quick it was on my Pentium Pro 200.
I think BeOS could have succeeded as a consumer OS were it not for the “focus” shift. I was very fast and it handled multimedia like there was no tomorrow. (Indeed, I would still be using it today if it were not for driver-rot.)
I do think Apple could have replaced the classic MacOS far more quickly had they bought BeOS, as it already ran on PowerPC/Mac hardware and NeXTStep needed to be ported. Still, with BeOS’ problems with large, non-multithreaded applications, Apple may have made a better choice in the long run, not to mention that regaining Steve Jobs was a boon.
This is not meant as a troll…
I’ve used BeOS and besides the file system and the fact that it was written from scratch and uses a C++ API, where’s the innovation?
No offense, but what does BeOS offer that isn’t available on, say Windows, Macintosh, or *nix? (And it just works better is not an option)
Once again, this is not a troll…i’m genuinely curious what the big deal with BeOS is…
-bytes256
Apollo Computer’s Aegis/DomainOS of course. Better Unix
than Unix, great windowing system. Beat the hell out of
Sun at the time, but everyone knows the best never wins.
HP eventually bought them out, killed Aegis and gave the
world HPUX (a.k.a. PHUX).
Okay, so you know where I’m posting from by my IP address. That qualifies me even more to say:
IRIX FOREVER!
IRIX is one of the best maintained UNIX OSs ever – every quarter a new release comes out, without fail. Then there’s the “freeware” distribution <URL:http://freeware.sgi.com/> – looks pretty similar to what lycoris and others are now starting to do (and the SGI freeware distribution has been going out for a few years now).
How many other OSs let you write and build software on a 10 year old machine (say an Indigo R4000), and then run it on your 1024 CPU Origin 3000 (Single system image, using ccNUMA)? That kind of binary compatibility is simply amazing, and that kind of scalability (whilst remaining reliable) is stunning.
IRIX is not dead, nor is it dying.
byte, when I voted, I took the poll as one where there was an OS that could have possibly taken the computing world by storm. I think that Be was in a time and place where, had things been different, it could have happened. In one sense, the poll is hard because of the historical timeline involved.
Looking at the comments really does make me wish so much I had looked into Amiga way back when. I was an Apple zealot then though <g>. But, it appears I may still get a chance at that.
Multilingual, used by all ages, doesn’t need upgrading, uses no power….
The Abacus
I am surprised that Inferno wasn’t included since you included Plan9. Basically, it is the OS that the Bell Labs folks started after Plan9. It incorporates a lot of the same concepts, plus some new ones. I have only used it under Linux, but what I was impressive. It used TK for the gui which some people may not like. I hope it starts to become more general purpose. Check out http://www.vitanuova.com/inferno/index.html for more info.
” Inferno runs in hosted mode under many different operating systems, providing an environment suitable for rapid development of distributed systems.
Native Inferno, on the other hand, is a complete operating system in its own right, running on embedded systems with as little as 1Mb of memory.”
IBM dropped the S/1 in the early ’90s, but we still run it under an emulator on AIX and SCO OpenServer
I voted for BeOS, because it is the best OS I have come across, and not because I am some newbie. I have used OS/2 v2 and v3, and was never all that impressed with it. The U of Arkansas had it for a while in a few argri computer labs (don’t ask I don’t know why) and it was interesting, but Be was the first one that had me saying “wow”.I have also tried Lunix, Mac, Apple, etc. The only I have never used is NextStep and Amiga, both of which I have heard are quite interesting.
Since I think that it was an excellent, excellent machine. It was sooooo ahead of its time. I remember coding blitter and copper chips, and still feel the exitement after creating my first copper bars! Actually, I hate Commodore because they killed Amiga for their greed. I believe that if it lived, we would not have to deal with these soulless shitty computer hw, and soulles OSs like Windows.
> If the people who voted for OS/2 were veterans, they would know about MULTICS.
Agree. I never tried any of them, but AFAIK, MULTICS was a *MUCH* more innovative OS at it’s time.
Didn’t used to see the EPOCH option (Psion handheld OS). In my opinion it was the most impressive handheld OS I ever seen (including WinCE, PocketPC, PalmOS, etc). A real shame the company fell down…
> Multics wasn’t THAT good. I love Vax/VMS – OpenVMS, but still, OS/2 was one step behond.
You have to replace it in context. Maybe I’m totally wrong, but there’s a LOT more difference between the before-and-after-MULTIX than the before-and-after-OS/2.
It would be nice if we could pick more than one to vote for.
For example, using a point system, the first place choice gets 3 votes, the second place, 2 votes and the third, gets one.
Then, record how many votes of each type was for each OS and display that in the final tally.
AIX – never used it. But it is VERY important to IBM as they have made lots of money off of it.
Amiga – the first multimedia computer that PCs are trying to be now.
AtariST – I used an Atari 400 (4k of RAM, no hard drive of floppy drive) back in 1982. I wrote my first 3D first person wireframe maze on it. Got boring fast as the maze couldn’t be bigger than 8 x 8 due to RAM limitation.
BeOS – BeOS 5 Pro can run multiple videos on the screen at one time without loss of frame rate and sound instantly kicks in on which ever video you are watching. First OS on home type computers that could do this.
GeOS – Great idea. It looked the same on laptops, desktops, etc., before any other OS could say that. It included an word processor, spreadsheet, card game and worked very well before Windows 3.1 caught up to it and looked better. It was always more stable than Windows 3.1 though. Its main flaw that it wasn’t pretty.
IRIX – never used it and don’t know much about it
Multics – never used it and don’t know much about it
NeXTSTEP – John Carmack raved about NeXTSTEP. Does anyone need to say more? No. But I will. As stated before. It still lives as a small part of OS X.
Novell – This is THE company that just can’t figure out how to market their product. They are even worse than IBM was when trying to market OS/2 with European or Itallian nuns who didn’t speak American. They have very good products that need significantly less servers to do the same thing as compared to MS. But they just know how to tell anyone this.
OpenVMS – never used it and don’t know much about it
OS/2 – IBM is the second most inept company at marketing behind Novell. When the beta for OS/2 2.0 came out I was able to replace 4 Windows 3.1 computers with one OS/2 computer running on the EXACT same model of computer. And it was far more stable even when running 16 bit Windows programs. I was even able to reduce my paper consumption by 90% at work as I could very easily receive/modify/and send faxes and other things that drastically reduced my need for printing things out. My paper consumption jumped dramatically when I had to switch back to Windows at work. OS/2 (which I still use on one of my computers at home) changed my life more than any other OS.
Other – I’ve not used this OS <wink>
Plan9 – never used it and don’t know anything about it
RiscOS – never used it and don’t know much about it
Solaris x86 – never used it. Not sure how it differs from “regular” Solaris other than what hardware it runs on.
Tru64 – never used it and don’t know much about it
All these OSs have some of their ideas live on. Software
Engineers don’t die, they just start new projects.
Preemptive multitasking, journaling filesystems, virtual memory, etc all started in single OSs. Look how they have
spread.
Now the BSD licence has made the pace of cross adoption much
faster. Anything released under the BSD liscence can be
adopted quickly to all OSs.
The only thing that doesn’t survive from good OSs is their
clean consistant code and interfaces. Besides the filesystem BeOS didn’t have anything revolutionary, but a lot of people really appreciated all the stuff it didn’t have cluttering up everything.
The same could be said for Tru64. Tru64 is in my opinion what I will miss the most. No 32-bit baggage to carry around, and everything they did was an almost perfect implementation. Nothing special feature wise, but just so darn well done.
Another question is what killed the OSs? What about the current ones, what will be their downfall.
Next – Too dang expensive and not so fast. Hard to sell that
BeOS – We are a hardware company, no a desktop OS company, no an embedded OS company, on PPC, no wait x86, with metrowerks, I mean gcc as our compiler, and netpositive, err Opera as our browser. The problem with Be is they couldn’t stick with a decision, and it’s hard for deveopers or customers to hit a moving target.
Tru64 – DEC, Compaq, HP. That alone is enough. To add insult to injury the Alpha was always sold as a low volume high price chip. Topp it off with making everybody jump to 64 bit when 16 bit was the standard.
VMS – it lived a long life and its time had come.
Multics – a great foundation who had been surpassed.
AIX – not dead yet, but quickly being commoditized by Linux
on the low end.
Solaris – not dead yet, see AIX.
OS/2 – Don’t mess with a better Monopolist head to head. Those OEM contracts won’t let you get in the market.
Linux – What is the strategy? Linux is being pulled in several competing directions. If it tries to be all things to all people it will become lousy at everything.
OpenBSD – can’t claim no remote root exploits anymore. Lacks features that are becoming more and more necessary (like SMP).
netBSD – there is only so much of a market for an OS targeting old discarded hardware.
FreeBSD – With Apple behind this now will it be overshadowed by big brother Darwin, OSX?
PalmOS – already being overtaken by Be. Nice stories in the register on Be’s takeover of Palm.
DOS, Win3.1, Win95, Win98, Win Me – NT (3.5, 4.0, 2000, XP, .net) has grown up and overcome. One united front from Redmond.
NT – When the OS starts to cost more than the hardware and has to be upgraded every other year Open Source starts to look better. Add to it a history of unreliability, poor security (even Gartner suggests running away), and infrequent fixes to well known problems and it is an uphill climb. Plus when you own the market there is nowhere to go but down.
OSX – Will PPC ever catch up in speed? Is a microkernel actually a good idea outside of a University? Certainly nicer than any other Desktop OS, and certainly more expensive than any other Desktop OS. Price may be a downfall.
MacOS – OSX has almost finished this battle.
IRIX – MIPS can’t keep up. Propriatary grahipcs workstations are too expensive. IRIX is already relegated to niche markets, and slowly being driven out of those.
There are others like OS/400, VM, NewOS, Plan 9, SCO UNIX, SunOS, etc. I’m just running out of time.
In summary all OSs will die. All of them will live on, their ideas embodied in what replaces them.
> BeOS 5 Pro can run multiple videos on the screen at one
> time without loss of frame rate and sound instantly
> kicks in on which ever video you are watching. First OS
> on home type computers that could do this.
BeOS is fantastic, but AmigaOS could do this already before BeOS existed. For playing more fancy movies simultaniously however, you would need more fancy Amiga as well or for example Amithlon http://www.osnews.com/story.php?news_id=604 which while *only emulating* AmigaOS is able to play truckloads of movies simultaniously, without any slowdown.
>Read more to vote for the operating system that should >have been the Next Big Thing (TM), but that never >happened for any reason.
I don’t really consider Atari ST and Amiga being elegable for this list, mainly because they were big for their time, but this list is meant to be for “that never happened”… progression doesn’t automatically defunct their once large status imho
I think almost every OS on that list deserves credit somewhere along the line, and when it comes down to what is the most revolutionary Operating System?, hmmm, Since the computers the ancient Inca and Greeks used analog and digital, can I assume that, since the user manipulated parts like we manipulate a mouse, it to had an “operating system”??
But on a more serious note, It’s not an easy poll to vote on thats forsure
It probably was a completely unworkable project, but it would have been nice to have a modern OS in the style of the classic Mac OS.
Zenja sez:
The biggest WOW I had in the computing world came from the first A500 I saw. 12 years later, BeOS provided the second WOW (notice how the second WOW isn’t bold).
————-
Amen. I feel exactly the same way. Nothing computer-related has excited & inspired me like those two. When I hear people talk of Linux, MacOSX or Windows as the future, I don’t know whether to laugh or cry.
when was the last time you saw an OS that wasn’t a knock-off of UNIX/Multics (cheap, high-quality, or otherwise)?
I voted for BeOS, but When I thought about it, wish I voted for OS/2, I’ve never used it, but have read detailed info on it (and thus NT), and personally, I think they outclass BeOS.
Considering the choices, I choose NeXTSTEP. To non-users, they probably don’t think there’s any difference between NeXTSTEP and OpenStep except for the name, but there’s quite a bit of difference. NeXTSTEP 3.3 is the acme of NeXT’s OSes.
If I could, I’d vote for NewtonOS 2.1. NewtonOS is, hands-down, the most user-friendly operating system I’ve ever used. Not “user friendly” like the Mac OS or Windows, but an OS that truly felt like you worked with it, rather than being forced to adapt to the way that things are done. When I first got my Newton 2100u, I was really amazed- I would love to have a desktop system that worked like that. I’m not talking just about the UI- that is superficial- but the design from the top down. It lacked in many applications that would make it suitable for using on the desktop, but Apple could’ve made the Newton OS their next generation operating system- instead of OS X- if they were truly interested in revlutionizing the computer industry.
While Microsoft talks about using databases as filesystems some day, and about all the very real benefits that will have for developers and users alike, the Newton OS had it from the beginning, and it worked wonderfully.
From the NewtonOS, I’m taking many ideas for Dynapad, my Squeak-based environment primarily for PDAs. Unlike the PalmOS, however, Dynapad’s purpose is to enable PDA-class devices to be used to do the same things as the intrusive desktop computers we use now, in a most natural way. I forsee not needing a desktop or laptop within the next couple years, as I get closer to implementing my dream software.
BeOS is probably the best it can get as far as desktop operating systems of the old-breed, the old vision of computing. The “old vision” is a bit of a misnomer however.
Rather, I shouldn’t say the old vision, rather the new one. The old vision was one that would be quite like the NewtonOS- of pioneers like Alan Kay (and the rest of the Smalltalk team), Doug Engelbart, Ivan Sutherland, Steve Wozniak, and others. The new vision, the one driven purely by profit, is stale. But the world seems to prefer computers that worked like the generation before them- which has us stuck using garbage like Windows, Linux, Unix and OS X.
Don’t misunderstand, so-called modern OSes are stable and have a wealth of applications. I use Linux and OS X on my iBook, and it’s passable I guess. But as I get more and more software ideas worked out, I find myself using my Jornada 720 running Dynapad instead of my iBook. I may have to put up with a slower CPU, but for a computer that works exactly like I think it should, it’s a miniscule price to pay.
I don’t expect many regular computer users and geek to understand what I’m trying to say. But surely there has to be at least one other person that reads OSNews that does.
I bring it down to Amiga BeOS and OS/2. First Amiga. They really had a good chance. Amiga brought all the Commodore 64/128 to the table. It had great hardware. It had lots of software. yes, games came out first on Amiga before ported to PC’s. One problem was that Commodore was a very corrupt company and that led to its demise. But there were other problems. Amiga had no concept of open hardware which guarantees it being behind the times (just like mac with no 2 ghz, ddr/rdram etc). Also I knew Amiga was doomed when I saw an Amiga 3000 running UNIX in hi res with only 16 colors when macs already had 16 million colors by default.
OS/2. Even though it looks like Windows 95 or 3.1 (in other words really ugly) it actually had some interesting UI concepts which have not yet been fully implemented in other OS’s. OS/2 had its chance during 1995. It had the biggest company, IBM behind it. But it really wasn’t. Once Windows 95 started to took over the best chance for an alternative OS was gone.
BeOS. Despite this, Be Inc kept trying. First they didn’t know what cpu to use. They picked the wrong one of course and Apple locked them out. Once the hard migration to x86 was over it was all about drivers drivers drivers. That makes it really hard to work on the real OS. They managed to get a decent lineup on apps on board. Certainly not as good as Amiga but that was ancient history. It was a start. But “killer apps” mean nothing if they couldn’t get BeOS preinstalled on PC’s. They couldn’t even have it as a dual boot option on the few companies that wanted it. They couldn’t even give it for free because then Microsoft would raise the price on Windows. That’s what really killed BeOS on the desktop. Some would say it’s a lost cause from the beginning but they had the guts for trying.
So I have to vote for BeOS. They started with absolutely nothing in their corner up against Windows which had already dominated the market. OpenBeOS might keep it alive-if they finish. But it will still be hard to get commercial apps and OEMs to support it again
> Amiga had no concept of open hardware
High-end Amigas offered fast Zorro slots and could easily be upgraded with 3rd party graphics cards and sound cards. Amigas were more open with regard to this then compared to PCs. For example try upgrading any PC from the 80s to Amiga class upgraded specs and you will see that you will never be able get a modern experience from the 80s PC while you can upgrade an A2000 to semi-modern specs with soundcards/graphics cards/network cards/etc.
Also there did exist AmigaOS clones without Amiga’s custom hardware chipsets. (DraCo)
One trouble however was that most people did not see any need to upgrade their machines as the standard machines provided them everything they needed.
> Also I knew Amiga was doomed when I saw an Amiga 3000
> running UNIX in hi res with only 16 colors when macs
> already had 16 million colors by default.
The Amiga 3000 was released in 1990 and was voted by computer experts to be the most revolutionary machine released that year. You should have bought yourself an Amiga graphics board instead.
The issue with OS X wasn’t getting it ported to PowerPC- that was easily done. Apple spent years on getting it to resemble classic Mac OS more, and getting it to work with legacy applications and legacy users.
As far as I can tell, there was nothing revolutionary about BeOS other than that it worked most of the time, more than Classic Mac OS and Windoze worked. And it is more consistent than Unix. If there was something truly revolutionary about BeOS, I’d be interested in hearing about it too, but from my (limited) personal usage and research, it’s nothing but the same stale desktop operating system paradigm done better. That’s not a bad thing- most people just want what they’re used to, not something revolutionary- and BeOS did what they were used to better.
To all those who are pretending to know what they’re talking about with the ST, but really dont, quit it.
The Atari ST isn’t the Atari 400. Or 800. Or 2600. It’s a different class of machine. The Atari ST was a 68k processor. That’s like having the option in the poll being “Macintosh,” and raving on about the Apple ][.
I’m no Atari ST junkie, but I know the difference enough to call the bluffs of some of you out there.
Now, if only someone could get me an Atari ST Notebook… mmmm, 4 pounds.
Be got it because it’s familiar and the best of those I’ve tried (Windows, Linux, Early Mac, RiscOS). In the spirist of the exercise, it could soooo easily be Riscos. When I see what it does on antiquated hardware it makes me thing what it could do if there had been more development.
I have absolutely no experience with the Amiga, so I couldn’t consider it.
BeOS was a little to much of a one trick pony with its multimedia performance, but I did like its novel file system.
But, NeXSTSTEP, wow, what a system. Today we take for granted a lot of the things that NeXSTSTEP brought to light.
Things that really grabbed my attention, besides the development environment.
Display PostScript – The fact that you used the same imaging model for both screen and hardcopy is truly a wonder to behold. The only real differences were the DPI and color model between print and screen, rather than some arbitrary printer driver making the choices for you.
The fact that many folks now use PDF as their format of choice for reports and such shows that DPS was a pretty good idea. The fact the Crystal Reports still can’t get its output to match the PDF output shows that we’re still in the ’80’s regarding imaging on some platforms.
MIME E-mail built-in – The standards were just emerging, which is why it was NeXTMail, versus the MIME standard of attachments et al. I’ll never forget the simple demo Jobs gave where he recieves a Purchase Order Request Document by e-mail, clicks on it, up pops the form from the associated application, he enters his User Password to approve it, and it goes on its merry way. Ad Hoc Document work flow using attachments in e-mail. Mundane today, but pretty darn innovative for the time.
Services – The heart and soul of the Workspace, I think. Extensible services available EVERYWHERE (when was the last time you were able to spell check your file name in a Save As.. dialog). This ubiquitous availability of functionality was very powerful. I’d like to see it get even more integrated within OS X.
Distributed Objects – Nice, lightweight, easy to use with minimal boiler plate. Hardly perfect, but it made for some interesting software at the time without all the mass of something like CORBA.
NetInfo – NetInfo was a love/hate thing. It made a lot of things easier, and even integrated fairly well, it just needed some better documentation, I think. But it was endemic to the nature of NeXTSTEP to leverage the standards it had (the Unix core) and add value to them.
OS X IS NeXSTSTEP, just NS 10.0, with a Mac compatability layer on the system. As the older Mac stuff goes quietly into the night, OS X will be able to shine more and more.
Windows for Workgroups get my vote….lol…it screams quality.
I looked up some factual information on Apple’s hardware in comparison with the Amiga 3000 in 1990.
The following hardware was released by Apple in 1990 offering the following display modes:
The Mac Classic – 2 colors at 512×342
Mac LC – 16 colors at 640×480 or 256 color at 640×400
The Mac IIsi/Mac IIfx – 256 colors at 640×480
Now compare that to the Amiga 3000 standard offerings in 1990:
In addition to lower resolutions screenmodes it offers
4096 colors at 640×512 or 64 colors at 1280×512
Conclusion
In 1990 the Amiga 3000 clearly outspecced any Apple hardware available at the time. If you need even higher resolutions you can easily add an old Picasso IV graphics card into your Amiga 3000. Then you can have a 1280×1024 24bit GUI!
Picasso IV http://www.vgr.com/picassoiv/
On technical merits, Lisp machines were much more advanced than Unix. If only they had survived…
Lisp machines were and still are the bomb! Incredibly advanced. But like anything revolutionary, no one wanted to use them.
I should mention Squeak Smalltalk, since we’re talking about LispMs now and someone else mentioned Oberon.
Squeak already has quite a few features in common with Lisp Machines.
Dynapad:
http://dynapad.swiki.net/1
Squeak:
http://www.squeak.org