Designing the First Apple Macintosh: The Engineers’ Story (1984)
spectrum.ieee.org
spectrum.ieee.org
Seeing a Mac for the first time, from the smiling icon at bootup to the way the pointer moved over things, and the way you could click to make things appear on the screen, was a transformative moment similar to the first time people a generation later encountered an image-capable web browser. It was one of those moments (like seeing VisiCalc in action) that made large numbers of people realize that a world was coming when computers might be usable by, and useful to, regular mortals.
I was nine years old when I saw my first Mac in 1984, and my only previous experience with computers was teaching an IBM PC to print 'BUTT' in an infinite loop using BASIC. I'd love to hear the impressions people older than me got the first time they used a GUI.
Here's one I shipped in 1983:
https://archive.org/details/byte-magazine-1983-11-rescan/pag...
A better-known system from the time was Framework, developed by Robert Carr and team and released in 1984:
Here's a link with examples of a later version: https://winworldpc.com/product/framework/xe
I had only ever seen "embedded" TUIs like on word processors, but even then, I only saw them after the Mac came out, and they were being dumped at thrift stores and flea markets. ...one time in '96, I came across someone selling a complete Xerox Daybreak at a flea market.
(the iPhone transformed him into the biggest Apple fan on the planet, somehow)
The very first GUI I ever used was a thing called Fleet Street Editor on a BBC Micro. It was slow, but since the task at hand was about graphics, it made sense to use it. All other things you could do on a micro were a lot faster on the CLI.
And I'm not judging, I was myself like this since a young geek in the 90s to a greybeard in the early 2010s, before something finally clicked in my head.
While us geeks were criticizing the productivity/effectiveness of the Mac, the iMac, the iPhone (and presumably soon the VisionPro), Apple was casually changing computing for everyone (literally for all of humanity) and forever.
What cemented it for me was re-watching 10-20 year old videos of Jobs at All Things D: https://podcasts.apple.com/us/podcast/steve-jobs-at-the-d-al...
I had no idea how patient and prescient he was about how the various pieces of technology Apple was working on would change the world. How revolutionary something like iTunes truly was.
The intersection of technology, usability, and BUSINESS (Only Nixon could've gone to China; Only Jobs could've convinced the Music Industry to digitize all their music to sell on iTunes) is truly astounding.
In our industry of frivolously inflated startup unicorns, it somehow seems just that Apple is the most valuable company in the world. And I say that as someone who spent most of my career criticizing them.
I got my first computer in 1983, and eventually got a Mac, when the "affordable" ones came out in the early 1990s. I don't think I'm dumber, or less successful for having waited that long.
The next closest watershed moment to that would probably be getting to uni and being able to telnet around the world.
Then after that, probably getting to use an SGI Onyx to do "military grade" flight sims.
It cannot be understated what the Mac did to the computing industry. Yeah, the Star did it first, and the Mother of All Demos showed it off in 1968, but those things were never in people's living rooms. Thr Mac was.
Just shows the power of presentation and a company behind a product that could launch it to bigger things. Commodore never really understood what it had with the Amiga, but Apple did...which is why the Mac has stuck around and the Amiga is just a footnote now.
The Amiga was very flashy but, ultimately, it felt like a toy to me. It was more fun to hack on than the Mac, but most people don’t care about that.
^ I’m so happy it has one today.
I suppose it was supposed to be an impressive demo, but even though I was in my early 20s I was like, wtf? Why would anyone want this?! Especially when I had a laser printer in my office!
The Amiga didn't look as slick but I could digitize images and record sound, and edit them in full color and in stereo. On the Mac, you were limited to desktop publishing because you could do that with only black and white. The Amiga OS looked a lot better with v2.x.
Amiga had a big impact on audio and art of I remember correctly. I took an art course in 1993 and the lab was full of Amigas. Not only did they have color screens but there was a rudimentary video capture system as well. We couldn't actually edit videos but we could incorporate video into projects.
We did have a Mac in the electronic audio lab though. Mind you, it was a single Mac Quadra. We had just gotten a 60mb hard drive which really opened up audio editing capabilities. We would dump our projects onto a DAT and reload them in our next time booked in the lab. All that gear was really expensive. I ended up using the analog stuff more (Moog something or other, Otari 1/4" 4 track, razor blades, tape, etc.) because it was so much easier to get up and going and tape loops were a lot of fun.
It's a shame Commodore never properly iterated on the product to let it reach it's potential.
After Jack left Commodore in 1984, Irving Gould was in control and he seemed hell bent on pulling as much profit as possible out of the company for his personal benefit. He didn't leave R&D or Engineering enough money to do anything meaningful. That is what killed Commodore. The greed of one man.
My Mom english major (who made a “living” as a potter saw I a Mac 128K at the mall and went on a frenzy to buy and experienced the birth of desktop publishing at the age of 46.
I wrote a whole book about it: Inventing the Future (starts end of 1976)
The PARC people already had several independent efforts (Bravo, Markup, Draw) and Smalltalk to build them with. Our task was to unify all the functionality into a single GUI.
(and by the way, Jobs and his minions didn't sneak in and steal it, contrary to what you've read.)
Not everyone was sold on it, a fact which some people would love to forget now. "We'll just slap a GUI on all this great functionality" was something marketing-droids actually did say.
The book is written with a strict "no hindsight" rule, but a friend and I did reconsider it all, with hindsight definitely allowed:
> Younger people on here may not realize what a mindbending experience it was to see a GUI for the first time.
Indeed. When the Mac came out, my friend and I went to a local computer store to look at one.My friend sat there for several minutes, mostly just playing with the menu bar, watching the menus come and go, flickering as he went from one to another with the mouse. He just sat there slack jawed going “Wow!”.
Folks have to appreciate the simple marvel of having that beautiful, crisp display with square pixels. This was not routine.
It was amazing.
We both bought one with the student discount. $1480 for a 128k Mac. Over $4300 inflation adjusted today. At the time a box of floppies were $40. Man did we blow vast amounts of money on computer hardware back then.
And, yes, a 128k, single floppy Mac was borderline useless. We got back alley RAM upgrades and external floppy drives. There went another $800.
But with that we could develop in C on the machine.
It was my first usable C compiler with the awesome MPW Shell. I was developing commercially and had a hard disk :-)
Luxury!
I’ve been a commercial Mac coder ever since, going via NeXT which turned out to be lucrative a few years later.
I’m sure you’re aware that OSX and, indeed, the current macOS is a UNIX like operating system. Hence, it’s more BSD and Linux like.
To aid legacy Macintosh applications, such as Photoshop and MS Office to run on OSX, an emulator was provided in early OSX releases that would detect old apps and run them in the emulator.
These apps were later modified to use a Classic (mostly) compatible set of libraries known as “Carbon”. This allowed them to run native but still have their older codebases.
The Carbon UIs were often written with a framework called “PowerPlant”, part of the Metrowerks C++ IDE.
Essentially, it gave large old codebases time to rework their code to the NeXTStep frameworks called “AppKit” and “FoundationKit” (rebadged as “Cocoa” by Apple).
Carbon APIs have now mostly been deprecated or removed as nowadays applications are written in Objective-C, the preferred NeXT language, and mixed with C++. Or, if they’re trivial Mac only apps, they might be written in “Swift”.
I do recall at Agfa in Belgium we had an effort to get the existing Mac products out of CodeWarrior’s compiler and into the g++ tool chain. It was non-trivial as the compiler Apple had provided for C++ wasn’t up to date. We had an Apple tech guy on site and he relayed the issue back to Apple and fixes to the compiler we’re made. Still, the code was PowerPlant based so looked old fashioned.
At another gig, Macromedia ported Flash from “CFM” to “mach-o” format (Classic Carbon to g++) and it failed to run. It turned out the g++ linker couldn’t handle code segments more than 32mb. Flash was a single blob of a binary. Apple released a new linker.
I think much if the reason why Apple moved off g++ to clang was due to the difficulty in maintaining g++ on their fork.
My first exposure to a computer was in 1981, when I took a high school programming class. So I was certainly not intimidated by text entry or errors. While in college, someone came through and demonstrated an Apple Lisa in the big lecture hall, and of course the Mac came out soon thereafter.
What I can say is that complete familiarity with text based systems didn't make me any less impressed with the GUI. It might have made me in less of a rush to go out and get one of my own, but my bank account had its own opinion about that too.
A couple years later I'd get my hands on a TRS-80 at school. I loved it.
When I went to high school in 1983 we had the Apple IIe in our lab. I spent hours on them doing basic and later Pascal. By then I had a vic-20 at home. I was a part of a nerdy group that hung out in the computer lab any time we could and we had a pretty good relationship with the head of the match department. When they go their first Mac, that was the first time I saw gui in person.
We were given a few minutes to play with it and I really liked it. I don't think it was so shocking though. While the command line was my normal initial interface, I was used to playing games that provided a GUI of their own to interact with the game. The gui on the mac was another program like a game running on top of the OS.
After high school I didn't have a pc of my own for years. When I did get one is was a Packard Bell sx386 running DOS. I had that for a decent amount of time before I got my first copy of windows. For me it all just fit together and I would say to this day I think of a computer and it's operating system as the CLI it provides. The GUI is an abstraction that runs on top. I like a nice desktop ( KDE is my favorite ) but I get very frustrated any time I can't get underneath it and do what I want. My first instinct any time I hit a problem that feels difficult, is to drop to the shell.
I presume you speak metaphorically, but just in case. Macintosh wasn't like that, the GUI was the OS, which is what made it so different from Microsoft's attempts, no DOS to drop down to.
The next time I experienced that same feeling was during Jeff Han's TED talk where he gave his multitouch demo.
On the same computer I was using many MS-DOS programs which not only were much more useful than anything that existed for Windows at that time (1990), but they also had much more clever user interfaces, some of which have never been completely matched by the later GUI programs, e.g. Lotus 1-2-3, AutoCAD, XTree, the Brief text editor and many others.
While it is said that in GUI programs it should be easier for the user to discover the available commands and their behavior, that was not really a problem for the good MS-DOS programs, because they all came with huge well-written user manuals, where the user could find everything that was needed to know about the user interface, so even when the UI consisted of obscure keyboard shortcuts, learning it was not difficult.
The only thing that I really liked in the Windows 3.0 GUI were the included fonts (Times, Helvetica and Courier), which were much nicer than the SuperVGA bitmap fonts that were used by the MS-DOS programs.
Later, Windows 3.1 and 3.11 for Workgroups were still not competitive with MS-DOS or DR-DOS. Only IBM OS/2 2.0 looked like a real improvement over the operating systems with a text-based UI, but it lacked useful applications.
Everything changed with the launch of Windows 95, together with the availability of cheap 486 CPU clones, which finally enabled a decent speed for GUIs, which previously had been annoyingly slow for any professional use.
Only with Windows 95, both myself and everybody I knew have switched to using GUIs.
When the Amiga came out, there was no way I was choosing a Mac over the powerful Amiga with 4096 colors, stereo sound, multi-tasking and amazing games. While I had it I was able to digitize and edit images, record and edit audio, built and used a MIDI interface with it, and even went online with it even before the web was a thing. I also ran a BBS on it and was still able to use the Amiga for all these other things at the same time.
The Mac was the first commercially available bitmapped GUI but the Amiga was the first system with a bitmapped GUI that was like modern OSes with color, sound, and multi-tasking. The Mac didn't have those things until later.
Definitely recommended reading, especially if you weren't around in those days.
I like this passage which explains some computing terms we take for granted now:
> The model was, of course, the Lisa workstation with its graphic “windows” to display simultaneously many different programs. “Icons,” or little pictures, were used instead of cryptic computer terms to represent a selection of programs on the screen; by moving a “mouse,” a box the size of a pack of cigarettes, the user manipulated a cursor on the screen. The Macintosh team redesigned the software of the Lisa from scratch to make it operate more efficiently, since the Macintosh was to have far less memory than the 1 million bytes of the Lisa.
The article also reminded me how expensive computers were in those days. Accounting for inflation, the $2500 Macintosh would be about $7500 today, and the $10,000 Lisa with its 1 million bytes of memory would be $30,000.
Back in 1982, I decided the just-introduced IBM PC was not a real computer because it didn't have a front panel with lights and switches. So I bought an Ithaca Intersystems DPS-1, a high end S-100 system with a Z80 processor and a real front panel, for $8000. Later I added a 20 megabyte hard drive for another $6000. So I guess that $14,000 computer would be about $42,000 today!
Something about both that strikes me: that wonderful era when the engineers were (and pardon me) left the fuck alone.
To be sure there was management (and I am under no illusion that Jobs would have been in fact one of the more complicated managers to report to) but there was an attitude of "let me wire-wrap it up tonight and I'll show you we can handle the mouse interrupts without a dedicated driver chip." Or, "I think I figured out a fast way to handle rendering to overlapping windows within out memory constraints. I coded it up last night."
It strikes me that a device like the Macintosh (or the Data General Eclipse for that matter) will never be created again in the same way.
I'm also old enough to see parallels in how software development was done at the beginning of my career and how it is done today. I admit that I miss those days.
This, plus the 1984 Macintosh article showed me what was a great time back in the 70s/80s for computer engineers -- you could design the hardware with a very small team, and there were more diversity in computer manufacturers back then.
Nevertheless, thanks to people like Ben Eater, I can still live that era with much more information and hand-holding. His 6502-computer series is a gold mine. I'm going to watch all of the videos and build one afterwards.
In the back of my mind, sometimes I feel that building a customer computer, if possible to be a 32/64 bit one, is an essential skill to fight the dystopian future we are inevitably walking into.
Also see : https://www.folklore.org/ all kinds of short stories about the early days of the Macintosh
Cursors on sun workstations (not just because of their optical grips mousepad) or windows boxes always felt like “oh there’s an abstraction between the physical hardware and the pointer, that’s just the way it is” - and nobody cared to make it better.
That "tightness" you mention is quite tangible on my 500Mhz PowerBook G3 when it's booted into OS 9. It would be nice to get that back in its entirety somehow.
https://www.folklore.org/StoryView.py?project=Macintosh&stor...
My friend DJ and I both attended the summer program for highschoolers at CMU in 1981. She accepted the early admission that were were both offered (I couldn't see skipping my senior year), so she was starting her sophomore year when I started as a freshman. DJ was a woman - a very rare thing in engineering at that time. She was MechE, but my ECE class of over a hundred had just two women.
Anyway with that story background, DJ bought an IBM and I bought a Mac. They were probably the only two computers in our dorm for a while. We joked with each other about their respective pros and cons. Funny thing is that many of those same jokes and pros and cons are mentioned today.
I still have that Mac. Unmodified Mac early off that assembly line :)
https://www.businessinsider.com/steve-wozniak-thought-the-fi...
(sorry, can’t find the reference now)
Jobs certainly had no interest in making the Mac compatible with the Lisa, and intentionally positioned it as a competitor - effectively killing the Lisa line. Sadly, preemptive multitasking didn't become standard in mainline Mac OS until OS X in 2001 - some 18 years after it appeared on the Lisa.
Apple also didn't want the Mac and Newton lines to compete with each other, much as they are currently trying to differentiate MacBooks and iPads. Kind of a shame since the eMate seems like it was an interesting system, and pen-based and palmtop Macs could have been interesting as well. Today iPadOS still has many limitations compared to macOS, in spite of running on the same silicon.