Why was A/UX developed?

[beachycove]

It was marketed in adverts along with the IIx and the SE/30 as "Apple's new advanced multi-tasking system." But UNIX types did not really need it, so what was Apple doing in this market?

In the end, it never made money. Was it an attempt to lay the groundwork for a new OS or was the decision to develop the product just a bad one commercially?

[Unknown_K]

Probably to compete with SUN and SGI in the UNIX workstation market (they needed a multi user, multitasking system which the classic macos did not have). Even commodore came out with an Amiga 3000UX UNIX workstation, there was a ton of money to be made by taking a standard 68K machine and selling UNIX (compared to the competition). I think around that time Apple was spending alot of time and effort trying to get the Mac II line going as a business machine (which is what it was designed for). There was a lot of effort going on to get Tokenring and Ethernet connectivity to IBM and UNIX workstations and servers. I could be wrong but Apple never made an ethernet card for the compacts for interfacing with the rest of the world (outside of serial port modems).

The WGS95 is the fastest 68K Apple made server (using A/UX).

Do you have any information about the money spent on development compared to sales of all versions of A/UX? They sunk a bunch into OS 8 copeland that was never released let alone sold.

[Anonymous Freak]

There were a lot of 68k-based UNIX workstations back in that timeframe (HP, Sun, SGI, DEC, NeXT, for example.) Apple just tried to join in.

At the time, the Motorola 68k architecture looked like it had a very bright future, at both the "PC" end of the spectrum to the "workstation" end of the spectrum. Apple would have been dumb *NOT* to try to hit into the workstation market, too. Commodore, in the form of Amiga UNIX, tried, too.

[johnklos]

People who were reading computer magazines at the time should remember all of the craze and buzz about the idea of the convergence of Unix (which was always too expensive for us mere mortals) and personal computers. Also, everyone talked about Unix System V Release 4 as though it was going to solve the Unix market fragmentation problem. Both things didn't happen in the way that the journalists imagined - no inexpensive Unix or Unix-like OS came to the personal or microcomputer market, and the Unix market itself remained as fragmented as ever.

Lots of people - myself included - thought that the advent of real CPUs and MMUs would herald a new era of proper OSes which didn't crash, didn't need to be rebooted, were secure, and internetworked easily. The Dark Ages of computing lasted MUCH longer. In 1985, I thought it'd just be a handful of years, in 1990 I thought it was imminent, but when nothing materialized I realized that it wasn't going to happen for a much longer time. I'd say that it took until 2005 before we began our Age of Enlightenment and people started realizing that computers didn't need to crash, didn't need proprietary networking protocols, and didn't need to be as insecure as a screen door.

Unix is what has saved the day, just as everyone had predicted, but it took 15 years longer than we all thought to get to OS X. OS X is NeXTSTEP merged with Mac OS. A/UX could have been the beginning, but AT&T Unix was not the way to go. NeXT is much more BSD-ish and NeXT was not afraid to make completely new software APIs along with its new interface. If Steve had done NeXTSTEP as a project at Apple, we might've had OS X in the mid 1990s.

The perfect time to do the transition would've been during the transition from m68k to PowerPC. m68k code needed to be emulated, so that would've been an excellent time to encapsulate the "classic" code in a "classic" mode while running a new OS with protected memory, preemptive multitasking, a new, updated filesystem, and a permissions model.

Regardless, from the perspective of anyone in the late 1980s, A/UX was certainly reasonable. Or at least the idea of A/UX was reasonable - Unix was the future, as Steve proved with NeXTSTEP.

Oh, well!

[ChristTrekker]

but AT&T Unix was not the way to go. NeXT is much more BSD-ish

Do you really think the flavor it was based on makes that much difference?

[jruschme]

People who were reading computer magazines at the time should remember all of the craze and buzz about the idea of the convergence of Unix (which was always too expensive for us mere mortals) and personal computers. Also, everyone talked about Unix System V Release 4 as though it was going to solve the Unix market fragmentation problem. Both things didn't happen in the way that the journalists imagined - no inexpensive Unix or Unix-like OS came to the personal or microcomputer market, and the Unix market itself remained as fragmented as ever.

In the late 80s and even into the early 90s, a number of the major players thought that desktop Unix combined with some degree of PC compatibility was going to be the way of the future. This produced a number of interesting systems including:

• AT&T UnixPC (aka 7300, 3B1)- a 68010-based Unix micro which could also run DOS programs via a compatibility card
• AT&T 6320- an 80286-based PC which ran Unix and something called DOSMerge that allowed DOS programs to run side-by-side with Unix programs
• Sun 386i- a 386-based Sun workstation

and some interesting software software solutions including SCO OpenDesktop (x86 unix combined with a Win 3.1-compatible evolution of DOSMerge) and Sun's WABI (Windows Application Binary Interface- an environment for running Windows 3.1 inside Solaris or Linux (including processor emulator under Sparc)).

I suspect that A/UX was either aimed at the same market or the related engineering market (target for the Sun 386i) which was more at home with Unix.

I would argue, as others have, that Mac OS X represents the first successful mainstream Unix desktop.

An ironic twist is that the *real* market for personal Unix systems was mostly the "hacker"/developer community. By that I mean, software types exposed to Unix on minicomputers who wanted to run it at home. This is the group which became the big 7300/3B1 users after the AT&T fire sale and which was the driving force of much early Linux and post-4.4 BSD development.

John

[Gorgonops]

Regarding the "Flavor" question:

Almost certainly the biggest fatal flaw preventing UNIX from getting any serious foothold in the market during the 1980s was licensing. UNIX had such a convoluted history it was really unclear at the time who deserved to be paid for it, and just how much. Unix was "invented" at Bell Laboratories (AT&T) at a time when AT&T was a government-blessed monopoly... they were "the phone company" in about the same sense that the "post office" still is today. Because of their unique position there were a number of consent degrees in place that dictated exactly what sort of products AT&T could sell "commercially". UNIX fell outside of that, so despite being owned by Bell laboratories and technically costing something like $40,000 for a full license it was essentially a non-profit exercise, and its wide distribution at universities and the loose (nearly nonexistent) enforcement of the licensing agreements made it essentially "open source". However, in 1983 when the government finally broke up AT&T's phone monopoly the company was no longer constrained by the consent degree and was technically free to sell UNIX as a product. And thanks to that UNIX development went to heck in a handbasket.

Here's an article talking about what happened. Look at the 1980-1990 section.

"BSD" is sort of a red herring here, because despite being "different" from AT&T's System V commercial product there were still legal issues with it. At the time BSD was technically packaged as a set of changes to an original Bell Labs source port... to be squeaky-clean legally you still needed to have an AT&T source license to work with it. In the late 1980's active efforts were organized to clean out the licensed code and replace it with "free" versions, but it wasn't until 1993 that BSD was indisputably available as a full stand-alone operating system without any AT&T encumbrances. Sure, by the 90's if you were developing a proprietary UNIX you'd almost certainly be better off going with BSD's source than AT&T's if for no other reason then you could simply download it all for free, slam the door, and just note that you used it in the footnotes of the user's manual. But in the 1980's there were legal landmines scattered all over it.

As to why so many computer vendors had this strange obsession with producing one-off UNIX versions in the 1980's... well, in retrospect it does look pretty silly. There certainly was a desire in academia for UNIX boxes then, and the pricing of the systems being used to provide it were stratospheric despite the hardware not being that much more powerful then the state-of-the-art in home computers by the middle 1980s. If adding UNIX could turn your $3500 home/business computer into a "workstation" which then could be sold as a replacement for a $10,000 SUN then the math must totally make sense, right? Even with the overhead of a UNIX license and associated development costs?

Of course, it never did actually work out, but the theory was probably sound on some level.

[ChristTrekker]

Almost certainly the biggest fatal flaw preventing UNIX from getting any serious foothold in the market during the 1980s was licensing.

Ahhhh! I was looking at it from a technical standpoint. I didn't even think of the legal morass that surrounded AT&T Unix back then.

[Unknown_K]

Even Microsoft used to sell UNIX back in the 80's (XENIX).

[Dog Cow]

To answer beachycove's question: Apple developed A/UX because the US Government only wanted to buy computer systems which were POSIX-compliant.

[johnklos]

That's doubtful; A/UX came out in 1988, and the first set of POSIX standards came out in 1988, too. While POSIX is a requirement nowadays, it certainly wasn't pre-1988, and I doubt that Apple started the A/UX project because of something the government might or might not do in the future.

The Mac II was seen as a workstation class machine when it first came out, and A/UX was an attempt to sell Macs to a market where people were used to spending lots of money.

[Dog Cow]

Look it up. I didn't just make that off the top of my head.

[beachycove]

The convergence of UNIX and desktop OSes is interesting. I did not know that there were a series of parallel developments. I am mostly interested in the question of human interface and in the application layer, rather than with these substructures.

GREAT discussion, by the way. What I take from it is that there was a cluster of factors that pointed in this direction, one of which might have been that Apple wanted a toe if not a foot in these waters, with a view to a possible, later plunge into mainstream OS development. They were also not alone in thinking like this, it seems. And in the meantime, there were niche markets like the US military who would buy the software.

Does this entail that the later decision to abandon A/UX at the time of the PPC move and to go with the Copland project instead meant that the company eventually decided that UNIX was not, in fact, the way of the future? Oh dear....

[johnklos]

Look it up. I didn't just make that off the top of my head.

I'm sure you didn't, but I can't see Apple in the mid 1980s worrying about POSIX before it was even invented, that's all. I'm sure they wanted to be standards compliant, which is why they went with System V release 4, but I don't know if that was their primary motivation.

On the other hand, some links discussing it would be educational. I've been wrong before, and I'm certain I'll be wrong in the future, too.

[Unknown_K]

Apple like any other company chases money and markets when they can (or think they can). Sure A/UX was dumped when they switched to PPC , but by the time the 604 was around they jumped back in for a little while with the ANS servers running a special AIX and then dumped it again. Apple still dables in commodity servers for some reason.

[ChristTrekker]

I wonder if it had something to do with the "Big Mac" project. Seems I recall reading that running Mac OS on a more robust foundation was an idea that predated NeXT, Copeland, or OS X, and I thought I recalled reading this in relation to Big Mac.

If only the initial Mac OS development had used a few different conventions (using \n line terminators, using / as a path separator in the filesystem, etc), marrying Mac and Unix could have been easier. Since Apple's target market at the time was primary education, and Unix was already widespread in higher education, one would have thought that someone along the way would have thought, "Hey, let's make transitioning between the two easier."

[z180]

Congratulations,you catched it.
The project "Big Mac" existed in protoype stage,you can find screenshots on the web.
It was a 68020 UNIX system but with b/w display and some prelimiary ADB bus.