7 Jun
2014
7 Jun
'14
6:22 p.m.
AU/X for PowerPC would have given them a contender
AU/X was a pretty awful Unix cobbled together to meet a narrow market need
unrelated to the markets Copeland was intended to address. It's singular
grace was it let Apple say with a straight face "yes, you can run Unix on
our hardware" (and to a lesser extent, say "and see, it looks pretty like a
Mac"). It wasn't a contender in the Unix market, much less the "platform
to take Apple to the next level" market.
KJ
9 Jun
9 Jun
6 a.m.
On 8 June 2014 00:22, Ken Seefried <seefriek at gmail.com> wrote:
AU/X was a pretty awful Unix cobbled together to meet
a narrow market need
unrelated to the markets Copeland was intended to address. It's singular
grace was it let Apple say with a straight face "yes, you can run Unix on
our hardware" (and to a lesser extent, say "and see, it looks pretty like a
Mac"). It wasn't a contender in the Unix market, much less the "platform
to take Apple to the next level" market.
Interesting. In what way was it pretty awful?
In the mid-1990s Apple needed a proper grown-up OS to face up to the
threat of Windows 95, which for all its faults, was a 32-bit OS where
it counted, which could preemptively multitask both DOS and 16-bit
Windows apps, along with its own new 32-bit apps, with a decent GUI.
Microsoft's long-term plan was NT and it spent from 1993 to 2000
gradually improving NT while selling Win9x to keep people going until
the clean new OS was ready.
Apple tried to create a replacement for MacOS and failed; in the end,
it bought one in.
A/UX was a proper grown-up 32-bit OS with memory protection and
preemptive multitasking. It ran (some) classic Mac apps, had a Mac GUI
but was Unix underneath.
A decade later, Apple bought in... a proper grown-up 32-bit OS with
memory protection and preemptive multitasking. After a lot of work, it
ran (some) classic Mac apps, had a Mac GUI but was Unix underneath.
So how was A/UX unsuitable?
I'm genuinely curious - I'm not saying you're wrong, just that with
hindsight it seems obvious, but I don't remember anyone thinking of it
at the time.
--
Liam Proven * Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven
Cell: +44 7939-087884
3:36 p.m.
A/UX has (had?) its warts... Not the best UNIX implementation compared to
what we have nowadays but in light of its contemporaries it was probably no
worse than average. The Mac OS GUI was still single-threaded,
cooperative-multitasking, no protected-memory System 7.0 with some
application compatibility issues... so the benefits of UNIX weren't really
extended throughout.
The claim was that A/UX made for a more performant fileserver, but
honestly, a Quadra 800 running System 7.5.0 feels quicker and more
responsive than A/UX running on a WGS 95 to me? Of course, System 7.5.0 was
a few years down the line... so I suppose it had improvements in its' own
right... To be fair, I should probably compare A/UX to System 7 or 7.1
running on bare metal. Nevertheless, this is mostly controlling for hard
disk, amount of memory available and so forth.
That all said, I don't see why A/UX would have been any less suitable a
foundation for Apple's next-generation operating system than NeXTstep? At
the time, NeXTstep has criticisms of its own; it was resource-intensive
(Mach still feels like a hog no matter how modern the hardware) and the
UNIX implementation there wasn't particularly complete or astounding
either...
You did have the OpenStep APIs and toolkits; that was always believed to be
the best part of NeXT... Apple didn't really have an equivalent to that in
OpenDoc or the historical Mac Toolbox. But I've never been a fan of ObjC
anyway ;)
I'm no Apple insider but I could totally see an alternate universe where
Macs are running an operating system with a Classic-like GUI and a
monolithic UNIX kernel underneath. I bet in practice, it would have run
faster than Rhapsody, and that great, easy-to-understand UI gestalt Apple
was known for could have remained unsullied :)
Best,
Sean
On Mon, Jun 9, 2014 at 6:00 AM, Liam Proven <lproven at gmail.com> wrote:
On 8 June 2014 00:22, Ken Seefried <seefriek at
gmail.com> wrote:
AU/X was a pretty awful Unix cobbled together to
meet a narrow market
need
unrelated to the markets Copeland was intended to
address. It's singular
grace was it let Apple say with a straight face "yes, you can run Unix on
our hardware" (and to a lesser extent, say "and see, it looks pretty
like a
Mac"). It wasn't a contender in the
Unix market, much less the "platform
to take Apple to the next level" market.
Interesting. In what way was it pretty awful?
In the mid-1990s Apple needed a proper grown-up OS to face up to the
threat of Windows 95, which for all its faults, was a 32-bit OS where
it counted, which could preemptively multitask both DOS and 16-bit
Windows apps, along with its own new 32-bit apps, with a decent GUI.
Microsoft's long-term plan was NT and it spent from 1993 to 2000
gradually improving NT while selling Win9x to keep people going until
the clean new OS was ready.
Apple tried to create a replacement for MacOS and failed; in the end,
it bought one in.
A/UX was a proper grown-up 32-bit OS with memory protection and
preemptive multitasking. It ran (some) classic Mac apps, had a Mac GUI
but was Unix underneath.
A decade later, Apple bought in... a proper grown-up 32-bit OS with
memory protection and preemptive multitasking. After a lot of work, it
ran (some) classic Mac apps, had a Mac GUI but was Unix underneath.
So how was A/UX unsuitable?
I'm genuinely curious - I'm not saying you're wrong, just that with
hindsight it seems obvious, but I don't remember anyone thinking of it
at the time.
--
Liam Proven * Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven
Cell: +44 7939-087884
4:52 p.m.
A/UX on a WGS 95 with the PDS card maxed with 512KB cache is the fastest 68k
file server I have played around with in my collection. A/UX 3 as a desktop
machine is like using OS 7.01, a bit limited.
-----Original Message-----
From: Sean Caron
Sent: Monday, June 09, 2014 3:36 PM
To: General Discussion: On-Topic and Off-Topic Posts ; Sean Caron
Subject: Re: The Mac inspired loyalty - was Re: Least PC-Compatible MS-DOS
machines?
A/UX has (had?) its warts... Not the best UNIX implementation compared to
what we have nowadays but in light of its contemporaries it was probably no
worse than average. The Mac OS GUI was still single-threaded,
cooperative-multitasking, no protected-memory System 7.0 with some
application compatibility issues... so the benefits of UNIX weren't really
extended throughout.
The claim was that A/UX made for a more performant fileserver, but
honestly, a Quadra 800 running System 7.5.0 feels quicker and more
responsive than A/UX running on a WGS 95 to me? Of course, System 7.5.0 was
a few years down the line... so I suppose it had improvements in its' own
right... To be fair, I should probably compare A/UX to System 7 or 7.1
running on bare metal. Nevertheless, this is mostly controlling for hard
disk, amount of memory available and so forth.
That all said, I don't see why A/UX would have been any less suitable a
foundation for Apple's next-generation operating system than NeXTstep? At
the time, NeXTstep has criticisms of its own; it was resource-intensive
(Mach still feels like a hog no matter how modern the hardware) and the
UNIX implementation there wasn't particularly complete or astounding
either...
You did have the OpenStep APIs and toolkits; that was always believed to be
the best part of NeXT... Apple didn't really have an equivalent to that in
OpenDoc or the historical Mac Toolbox. But I've never been a fan of ObjC
anyway ;)
I'm no Apple insider but I could totally see an alternate universe where
Macs are running an operating system with a Classic-like GUI and a
monolithic UNIX kernel underneath. I bet in practice, it would have run
faster than Rhapsody, and that great, easy-to-understand UI gestalt Apple
was known for could have remained unsullied :)
Best,
Sean
On Mon, Jun 9, 2014 at 6:00 AM, Liam Proven <lproven at gmail.com> wrote:
On 8 June 2014 00:22, Ken Seefried <seefriek at
gmail.com> wrote:
---
This email is free from viruses and malware because avast! Antivirus protection is
active.
http://www.avast.com
AU/X was a pretty awful Unix cobbled together to
meet a narrow market
need
unrelated to the markets Copeland was intended to
address. It's
singular
grace was it let Apple say with a straight face "yes, you can run Unix
on
our hardware" (and to a lesser extent, say "and see, it looks pretty
like a
Mac"). It wasn't a contender in the
Unix market, much less the
"platform
to take Apple to the next level" market.
Interesting. In what way was it pretty awful?
In the mid-1990s Apple needed a proper grown-up OS to face up to the
threat of Windows 95, which for all its faults, was a 32-bit OS where
it counted, which could preemptively multitask both DOS and 16-bit
Windows apps, along with its own new 32-bit apps, with a decent GUI.
Microsoft's long-term plan was NT and it spent from 1993 to 2000
gradually improving NT while selling Win9x to keep people going until
the clean new OS was ready.
Apple tried to create a replacement for MacOS and failed; in the end,
it bought one in.
A/UX was a proper grown-up 32-bit OS with memory protection and
preemptive multitasking. It ran (some) classic Mac apps, had a Mac GUI
but was Unix underneath.
A decade later, Apple bought in... a proper grown-up 32-bit OS with
memory protection and preemptive multitasking. After a lot of work, it
ran (some) classic Mac apps, had a Mac GUI but was Unix underneath.
So how was A/UX unsuitable?
I'm genuinely curious - I'm not saying you're wrong, just that with
hindsight it seems obvious, but I don't remember anyone thinking of it
at the time.
--
Liam Proven * Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven
Cell: +44 7939-087884
5:54 p.m.
On 6/9/14 12:36 PM, Sean Caron wrote:
I don't see why A/UX would have been any less
suitable a
foundation for Apple's next-generation operating system than NeXTstep?
A/UX died with the switch to PowerPC. Apple had an AIX product which lost
all MacOS compatibility. There was a push inside for Mach around
the time that MacLinux came out, but deep down inside no exec inside Apple
wanted a Mac operating system based on Unix. It came up over and over again
from the time A/UX started until NeXT was bought, Apple
thinking that they
could do an entire OS and operating envrionment from scratch and
each time
it went down a rathole for one reason or another. The closest anyone ever
came to building a system with the Apple mindset was Be, who had a lot of
high-level ex-Apple people that designed it. NeXT was created starting from
Rich Page and BigMac, which was Apple's Sun workstation competitor. The
vision was more focused and higher end than anyone at Apple was thinking at
the time, and ultimately it too failed first as a workstation (a space that
had already switched to RISC before they tried to build something with an
88110) and as a software product (Windows had already taken over the desktop
market). OS X would have failed if CPU speed and memory density hadn't been
where it was, and if they hadn't glued enough MacOS vernier over it to make
it palatable for the user base. It didn't hurt that Apple has never had to
care much about backwards compatibility the way that Microsoft has had to
and could jettison the existing MacOS application and system code base. I
suspect that Microsoft had more testers as Apple's entire software staff
in the late 90's.
I can't imagine how big the iOS and MacOS teams are now.
10 Jun
10 Jun
6:36 a.m.
On 2014-06-09 21:36, Sean Caron wrote:
That all said, I don't see why A/UX would have
been any less suitable a
foundation for Apple's next-generation operating system than NeXTstep?
A/UX was dead and buried by the time Apple bought NeXT.
Also, you're pondering the wrong questions. Apple weren't looking for a
UNIX; they were looking for an OS that would become the next-gen Mac OS.
The underpinnings, as much as they mean to gearheads like the populace
of lists like these, were inconsequential. Appl eneeded an OS, *any* OS,
that could carry the Mac forward into the new millenium.
They got more than they bargained for, bringing SJ back into the fold as
well, although of course it has also been suggested that maybe it wasn't
Apple taking over NeXT, at all...
You did have the OpenStep APIs and toolkits; that was
always believed to be
the best part of NeXT... Apple didn't really have an equivalent to that in
OpenDoc or the historical Mac Toolbox. But I've never been a fan of ObjC
anyway;)
I think you answered your own question, there.
Look at the Mac OS -oops: OSX- as it stands, today. Everyone whining
about how great and polished and efficient System 7 was should take a
short break from OSX every now and then and try to work with System 7
for a while. You'll be pulling your hair out in no short order, because
of all the little conveniences we take for granted that wouldn't have
been possible to bolt onto classic Mac OS, at least not without turning
into an even bigger kludgy pig than it already was by the time 7.6 shipped.
Apple aren't in the business of making technical marvels that are
marvellous from a technical standpoint. Every now and again, they do do
that, but that is merely a side effect. Apple are in the business of
making things atht are marvellous from the viewpoint of users. How that
is accomplished, is entirely not the user's problem. Who cares if NeXT
is a resource hog: as long as resources are plentiful, it'll do some
damn handy things for the user.
I'm no Apple insider but I could totally see an
alternate universe where
Macs are running an operating system with a Classic-like GUI and a
monolithic UNIX kernel underneath. I bet in practice, it would have run
faster than Rhapsody, and that great, easy-to-understand UI gestalt Apple
was known for could have remained unsullied:)
In that universe, I'm afraid Apple would no longer exist today and we'd
still be jumping through hoops trying to get our Symbian feature phones
to work with our Windows machines and just grinning and bearing it when
our Outlook contact lists would be cleared again after our weekly
botched synch.
J.
11 Jun
11 Jun
12:43 p.m.
I play with System 7 and 7.5 all the time; daily, almost, and I wouldn't
say that I find it frustrating to use when compared to OS X, adjusting for
the fact that it's a 20 year old operating system running on 20 year old
hardware! I think to the contrary; to me, the classic Mac OS is still
remarkable; a truly 100% GUI drag-and-drop OS all the way down to
installing, modifying and copying the System. It was eminently easy to use,
and usability -- and consistency over time -- were considered paramount.
Someone who learned how to use a Mac in 1986, would not have been out of
place in 1996. There was just no comparison then -- or now?
It seems that nowadays Apple really couldn't care less about usability and
consistency over time; they have become just as capricious as Microsoft at
racing to implement pointless new flashy UI features, switching up the
method for completing a task pointlessly, just for the sake of doing it
(why does the method for installing a printer need to change with every
release?); etc. To Me, Mac OS X is no marvel compared to Windows 7. I miss
what Apple used to stand for in building a computer or an operating system.
Honestly it's true, the best melding of classic Mac OS usability values and
modern operating system internals underneath was not A/UX at all; it was
BeOS! I just think it is slightly inaccurate to say that Apple had nothing
internally they could have drawn upon in creating a next-generation
successor to the classic Mac OS. Would it have been any greater effort to
port A/UX to PPC and bolt on the Finder from System 9, than it would have
been to port NeXTstep to PPC and graft a few Classic-reminiscent visual
queues on top of it? Would taking the alternate route of basing their
next-gen OS on A/UX (or BeOS??) instead of NeXTstep have dramatically
altered Apple's fate in the long run? It's hard to say, as the introduction
of OS X kind of co-incided with their move more into the consumer
electronics space (iPod, iPhone, etc). How much of Apple's revenue is
derived from selling computers today? Probably not much.
Best,
Sean
On Tue, Jun 10, 2014 at 6:36 AM, Joost <gyorpb at gmail.com> wrote:
On 2014-06-09 21:36, Sean Caron wrote:
That all said, I don't see why A/UX would have been any less suitable a
foundation for Apple's next-generation
operating system than NeXTstep?
A/UX was dead and buried by the time Apple bought NeXT.
Also, you're pondering the wrong questions. Apple weren't looking for a
UNIX; they were looking for an OS that would become the next-gen Mac OS.
The underpinnings, as much as they mean to gearheads like the populace of
lists like these, were inconsequential. Appl eneeded an OS, *any* OS, that
could carry the Mac forward into the new millenium.
They got more than they bargained for, bringing SJ back into the fold as
well, although of course it has also been suggested that maybe it wasn't
Apple taking over NeXT, at all...
You did have the OpenStep APIs and toolkits; that was always believed to
be
the best part of NeXT... Apple didn't really have an equivalent to that in
OpenDoc or the historical Mac Toolbox. But I've never been a fan of ObjC
anyway;)
I think you answered your own question, there.
Look at the Mac OS -oops: OSX- as it stands, today. Everyone whining about
how great and polished and efficient System 7 was should take a short break
from OSX every now and then and try to work with System 7 for a while.
You'll be pulling your hair out in no short order, because of all the
little conveniences we take for granted that wouldn't have been possible to
bolt onto classic Mac OS, at least not without turning into an even bigger
kludgy pig than it already was by the time 7.6 shipped.
Apple aren't in the business of making technical marvels that are
marvellous from a technical standpoint. Every now and again, they do do
that, but that is merely a side effect. Apple are in the business of making
things atht are marvellous from the viewpoint of users. How that is
accomplished, is entirely not the user's problem. Who cares if NeXT is a
resource hog: as long as resources are plentiful, it'll do some damn handy
things for the user.
I'm no Apple insider but I could totally see an alternate universe where
Macs are running an operating system with a
Classic-like GUI and a
monolithic UNIX kernel underneath. I bet in practice, it would have run
faster than Rhapsody, and that great, easy-to-understand UI gestalt Apple
was known for could have remained unsullied:)
In that universe, I'm afraid Apple would no longer exist today and we'd
still be jumping through hoops trying to get our Symbian feature phones to
work with our Windows machines and just grinning and bearing it when our
Outlook contact lists would be cleared again after our weekly botched synch.
J.
1:37 p.m.
On 6/11/14 9:43 AM, Sean Caron wrote:
I miss
what Apple used to stand for in building a computer or an operating system.
That was one of the reasons that I quit. I came into Apple after a solid
Unix background and did everything I could to get them to abandon the existing
OS because of how primitive it was, but the people that made it work and cared
very deeply about building the best system they could given the constraints
placed on delivering something that wasn't TOO different gave me a different
perspective on the problem. Pretty much all of them were forced out of the
company during the purges by NeXT management. The Apple of today is a money
machine and really, really, really does not care about you. THAT was the direct
result of Steve coming back.
They are all that Gates dreamed of in the 90's.
This is not the Apple that I was a part of and after being away since 2005
I'm not even afraid of saying that in a public forum any more.
Honestly it's true, the best melding of classic
Mac OS usability values and
modern operating system internals underneath was not A/UX at all; it was
BeOS!
Duh... They were APPLE people. Lots of refugees from the insanity of Pink/Taligent
who wanted to build something small and elegant. Many of them ended up at Danger/Android
which sadly is now heading where Apple was 10 years ago wrt lock-in an user disdain
(eyeballs ARE just the product).
I've heard of many people bailing for other groups and startups because of the circus
that is Google.
19 Jun
19 Jun
7:32 a.m.
On 11 June 2014 18:43, Sean Caron <scaron at umich.edu> wrote:
I play with System 7 and 7.5 all the time; daily,
almost, and I wouldn't
say that I find it frustrating to use when compared to OS X, adjusting for
the fact that it's a 20 year old operating system running on 20 year old
hardware! I think to the contrary; to me, the classic Mac OS is still
remarkable; a truly 100% GUI drag-and-drop OS all the way down to
installing, modifying and copying the System. It was eminently easy to use,
and usability -- and consistency over time -- were considered paramount.
Someone who learned how to use a Mac in 1986, would not have been out of
place in 1996. There was just no comparison then -- or now?
I agree. I don't use Classic nearly that much, but I still love it.
*The* most elegant GUI OS ever built.
(For me, the rivals are Acorn's RISC OS, sadly almost unknown in the
USA, but a lovely system - but with weird hairy text-file ugliness
underneath, although that did mean it was far more customisable than
classic MacOS; and BeOS.)
It seems that nowadays Apple really couldn't care
less about usability and
consistency over time; they have become just as capricious as Microsoft at
racing to implement pointless new flashy UI features, switching up the
method for completing a task pointlessly, just for the sake of doing it
(why does the method for installing a printer need to change with every
release?); etc. To Me, Mac OS X is no marvel compared to Windows 7. I miss
what Apple used to stand for in building a computer or an operating system.
A little harsh - I think Apple still has a significant edge - but I
know where you're coming from.
Honestly it's true, the best melding of classic
Mac OS usability values and
modern operating system internals underneath was not A/UX at all; it was
BeOS!
Agreed.
I just think it is slightly inaccurate to say that
Apple had nothing
internally they could have drawn upon in creating a next-generation
successor to the classic Mac OS. Would it have been any greater effort to
port A/UX to PPC and bolt on the Finder from System 9, than it would have
been to port NeXTstep to PPC and graft a few Classic-reminiscent visual
queues on top of it? Would taking the alternate route of basing their
next-gen OS on A/UX (or BeOS??) instead of NeXTstep have dramatically
altered Apple's fate in the long run?
Well, actually, yes. I mean, given unlimited resources, they could
have done a Microsoft-style split product line: BeOS as the desktop
product, a notional PowerPC A/UX as the server and high-end
workstation product, with limited software compatibility between them
- as MS did with Win9x and NT for nearly a decade. (NT 3.1, 1993; XP
on GA, 2002.)
But honestly, it would have killed the company, long-term.
Why? Because...
It's hard to say, as the introduction
of OS X kind of co-incided with their move more into the consumer
electronics space (iPod, iPhone, etc).
Well, yes, but more importantly, that is what saved the company. As you say:
How much of Apple's revenue is
derived from selling computers today? Probably not much.
And why? Because the NeXT deal bought back Jobs, and Jobs got them out
of the hole they were in.
Pre-Jobs, not only did Apple have a clapped-out OS (but with hindsight
a potential successor and replacement already there, but languishing
in the labs), but it also had a product range all over the shop:
* dozens of models of Macs - Jobs cut that down brutally and even
simplified the nomenclature
* peripherals - cameras, printers, etc. - that were no better than the
competition's offerings
One of Jobs' brilliant moves was to kill off the cameras, printers,
most of the monitors, indeed, most of the peripherals, and force the
Mac to accept and support commodity peripherals by removing its
proprietary or semi-proprietary expansion ports (ADB, Localtalk) and
replacing them with industry-standard, modern, robust, plug&play ones:
USB and Firewire. You got nothing else so you /had/ to use them.
Voil?, Apple gets out of a low-margin business and users get more
freedom of choice.
But then, once the migration to the new OS was in place, he
diversified aggressively into the market of smaller standalone smart
devices - MP3 players, then phones, then tablets - which were already
out there but where the existing offerings were, frankly, crap.
He got Apple out of its niche, by leveraging its new OS into places
where it hadn't occurred to any other company to try to take a
general-purpose non-specialised OS - Unix. This transformed whole
product categories by replacing horrible UIs with great ones.
Too many geeks focus on the small techie details that are utterly
irrelevant - CPU architectures, for example, or OS design, or
touchscreens - and miss the bigger picture.
E.g. my office has 2 networked smart multifunction printers, capable
of printing, scanning, faxing, emailing, etc. all on their own or
talking over Ethernet to devices running any OS - we use them from
Linux.
These devices have touchscreen UIs.
They are horrid and almost unusable. I and 2 colleagues -- all very
smart professional techies -- spent more than 30min trying to persuade
one of these MFDs to scan a document for me. We failed, individually
and collectively. In the end, switching to the other, we did it after
"only" 5-10min of fiddling.
Touchscreens are no panacea. Lots of pocket devices had touchscreens
before the iPhone, and frankly, they were all utter crap compared to
the iPhone.
And I am no fanboy. There's an iPhone 4 sitting next to this laptop,
but it's turned off and unused. I use an Android phone, because it's
better-specced, more open and far more customisable, even in its
limited way. But it's a ripped-off, but improved (for my purposes),
iOS.
Same as the Unity desktop I'm using on Ubuntu is a ripped-off, but
improved, Mac OS X desktop.
The real genius was in inventing the originals.
--
Liam Proven * Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven
Cell: +44 7939-087884
3686
days inactive
3698
days old
8 comments
6 participants
participants (6)
-
aek＠bitsavers.org -
gyorpb＠gmail.com -
lproven＠gmail.com -
scaron＠umich.edu -
seefriek＠gmail.com -
teoz＠neo.rr.com