Slashdot Mirror
How the Lisa Changed Everything
Sabah Arif writes "The
Lisa, started in 1979 to provide an inexpensive business computer to
Apple's lineup, enjoyed little success. With its advanced
object oriented UI and powerful office suite, the computer was priced
well above the means of most businesses. Despite its failure,
the Lisa influenced most user interfaces, and introduced many
features unheard of in earlier systems (like the Xerox Star or VisiOn).
Read the story of the development and demise of the Apple
Lisa at Low
End Mac."
/. already posted this story http://apple.slashdot.org/article.pl?sid=05/07/31/ 1346224&tid=190&tid=3 a few months ago. In their defense, the old article was hosted at Braeburn.ath.cx (but looks like they've redone their website and braeburn resolves to lowendmac.com).
The Apple Lisa was way too expensive and rather slow. The Mac was much cheaper and worked much better. Hence why you buy an Apple Mac not an Apple Lisa when you go to an Apple store now.
It was Steve Jobs who brought us the Mac too, he recognised it was the right product and better than the Lisa. Much of the work on the project until Jobs took over was done by Raskin and his team.
I kid, I kid.
Anyway, here's a picture of the orgional ad: http://www.jagshouse.com/lisabrochure.html
I don't get it.
When Jobs brought technology in from Xerox PARC, and Adobe, he had the keys to the kingdom handed to him on a silver platter:
1) A tokenized Forth graphics engine.
2) Smalltalk.
The Forth graphics engine was originally intended to grow from a programmable replacement of the NAPLPS videotex graphics protocol, into a silicon implementation of a stack machine upon which byte codes, compiled from Smalltalk would be executed. At least that's the direction in which I had hoped to see the Viewtron videotex terminal evolve when I originated the dynamically downloaded tokenized Forth graphics protocol as a replacement for NAPLPS in 1981 and discussed these ideas with the folks at Xerox PARC prior to the genesis of Postscript and Lisa.
If Charles Moore could produce an economical 10MIPS 16 bit Forth engine on a 10K ECL gate array on virtually zero bucks back then, why couldn't Jobs with all his resources produce a silicon Postscript engine with power enough to execute Smalltalk?
Somehow a Forth interpreter made it into the first Mac, as did Postscript, but Smalltalk just didn't.
The Motorola 68000 family just didn't have the power. It may have been better than the Intel 86 family, but that really isn't saying much, now is it?
Seastead this.
The site seems to be here: http://emulation.victoly.com/ Must admit I couldn't find any Lisa emulators on the site, but did find this link: http://lisa.sunder.net/
Jumpstart the tartan drive.
Objective-C's object system and general philosophy is _very_ smalltalk-ish.
Java is much more "C++ with some warts removed" than an Obj-C derivative. Obj C _is_ a "dynamic, late-binding programming environment." C++ and Java are not.
Self is no more a son of Smalltalk than Java is a son of Obj C. They (Self and Smalltalk) both came out of PARC, but they are very different.
I suspect you have no more idea about what went on at Apple than you do about programming languages, but I can't speak to that myself.
My father, an early adopter-type, had a Lisa for his office, and it was the Lisa that I first learned how to program on.
One of the most maddening things about programming the Lisa was that you couldn't make programs that integrated well with the Lisa office suite. Why? Because there was no API for the GUI. None. If you wanted a window drawn, you fired up QuickDraw and drew it yourself. Want a scroll bar? Do it yourself. Menus? Right.
I ended up only using the development environment's console for my programs' interfaces. The development environment was also console based, probably for the same reasons. A couple of years later, Apple released the Lisa Toolkit that had all that stuff, after they had announced they were going to discontinue it.
So in my opinion, it was the lack of software that killed the Lisa, not its high price. I mean, people were paying for it, and they wanted more. The ability to use proportional fonts was the killer feature to end all killer features.
It's worth noting that Apple learned its lesson about making developers happy - the developer support program for the Macintosh has been one of the best.
I had a Lisa, and Apple made the same mistakes with the Lisa as Xerox had made with the Star: it was too expensive, in particular for the limited hardware and completely incompatible software you got.
Claims that the Lisa represented significant technological innovation seem dubious to me. You need to compare the Lisa to the totality of R&D efforts around at the time, not just the Star. Xerox alone had Alto, Star, Smalltalk, and probably others. The GUI of the Lisa was an evolutionary change, and not always for the better; what was under the hood of the Lisa can charitably be described as pedestrian. It took Apple 20 years to catch up and finally adopt system software that even is in the same league as Smalltalk-80 (that's "80" as in "1980"; Smalltalk-80 is the language and platform that Objective-C and Cocoa are modeled on).
Lisa's main significance was to be a prototype for, and cannibalized for, Macintosh (and it served as the main development machine for Macintosh apps for a while), but I can't think of any significant new technology it introduced.
Interestingly, the article also refers to the 8088 as a 16-bit processor, which is an 8-bit processor if one uses the same criteria that you'd have to in order to call a 68000 "16-bit".
68000: 32-bit registers, 24-bit address bus (linear addressing), 16-bit data bus
8088: 16-bit registers, 20-bit address bus (segmented addressing), 8-bit data bus
I frankly don't consider the 8088 and 68000 even remotely comparable - it's far easier to program for (and design hardware around, IMHO) the 68K. The only difficulties that I knew of anyone really experiencing when moving to the 68020 and other full 32-bit variants was that people had gotten into the really bad habit of using the upper 8 bits of the A registers for general storage, which would break things on a '020 horribly. Even so, it was certainly nothing like the EMS/XMS hell that PC programmers had to go through just to use memory above 1MB because of the limitations of the 8088 memory architecture.
Please stand clear of the doors, por favor mantenganse alejado de las puertas
Two years later. And yes, the graphics and multi-tasking ability of the Amiga kicked the Lisa's ass (and Macintosh's) all over the shop at a fraction of the cost, no doubt about it.
But, as a former Amiga user, I'll still say that the OS 1.x interface wasn't the best GUI ever. They improved quite a lot with 2.x onwards, but that was five years or so later.
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
The lack of a hard drive was the killer. By the time the Mac came out, IBM PCs had a hard drive, so Apple was playing catch-up. Apple had tried building hard drives (the LisaFile), but they were slow and crashed frequently. But at least the Lisa had a hard drive. Third parties added a 10MB hard drive to the Mac in early 1985, which brought performance up to an acceptable level. Some people say that third-party hard drives saved the Mac. But Apple fought them tooth and nail. Apple finally came out with a 20MB external hard drive for the Mac in 1986. This was very late; IBM PCs had been shipping with hard drives for five years.
Sales for the Mac were well below expectations. Apple had been outselling IBM in the Apple II era. (Yes, Apple was once #1 in personal computers.) In the Mac era, Apple's market share dropped well below that of IBM.
What really saved the Mac was the LaserWriter, which launched the "desktop publishing" era. But that required a "Fat Mac" with a hard drive and 512K. By then, the Mac had reached parity with the Lisa specs, except that the Lisa had an MMU and the Mac didn't. The Lisa also had a real operating system, with protected mode processes; the Mac had "co-operative multitasking" in a single address space, which was basically a DOS-like system with hacks to handle multiple psuedo-threads.
The MMU issue was actually Motorola's fault. The 68000 couldn't do page faults right, and Motorola's first MMU, the Motorola 68451, was a terrible design. The Lisa had an Apple-built MMU made out of register-level parts, which pushed the price up.
Apple might have been more successful if they'd just stayed with the Lisa and brought the cost down as the parts cost decreased. They would have had to push Motorola to fix the MMU problem, but as the biggest 68000 customer, they could have.
You have to remember that the microcomputer hardware was sloooow and expensive back then. Compiled "static" languages simply run faster, even if they take longer to program with.
Personally, I feel that GUI's should be mostly declarative based such that one stores descriptions and attributes of windows and widgets rather than use boatloads of "new Window(...)" and "new Widget(...)" commands in code. Events are then bound to a programming language of choice. Declarative approaches are usually easier to adapt to multiple programming languages. Why does the world still want to hard-wire GUI's to one and only one programming language? Java still made this same mistake in 1996, almost 20 years after Job's alleged "big mistake". I see no reason why it *must* be this way. It does not make sense to reinvent GUI engines for each of the 100 or so common languages. They like to talk about "reuse", but don't practice what they preach.
Table-ized A.I.
Another good site full of first-hand descriptions of how early Apple development was done is http://folklore.org/.
I've never owned a Mac, and am too young to have been involved in earlier developments - but that site does make it all seem very impressive.
The Lisa, like other computers of the day, had rectangular pixels. The Mac's introduction of square pixels allowed true WYSIWYG, and was crucial to desktop publishing and computer art. The Mac's still strong position in the graphic arts industry is a direct result.
Q: What does the "B." in Benoit B. Mandelbrot stand for? A: Benoit B. Mandelbrot
Holy crap, are you misinformed. Read some books on the history of the personal computer industry in the late 70s and the 80s. Start with Cringely's "Accidental Empires."
Actually, the original PC BIOS wasn't so much reverse-engineered, as it was simply duplicated. IBM published the full annotated assembler listing in the original IBM PC technical manual.
Yes, that was done purposely in the hopes that its existence could be used to quickly shut down any cloners via copyright infringement lawsuit. The grandparent poster was right on about Phoenix. They actually took out a huge insurance policy from Lloyd's of London that would protect them if IBM sued them, so IBM couldn't just use a nebulous lawsuit to legal-fee them into submission (as Hughes and the RIAA like to do today).
IBM for its part didn't seem to care one way or the other.
The hell they didn't! Aside from addressing shortcomings in the ISA bus, one of the main reasons IBM developed the very proprietary Micro Channel Architecture was to try to cram the cloning genie back in the bottle. IBM told the cloners they'd be happy to license the MCA, but demanded extortionate fees to do so, and apparently also wanted to be paid for every prior IBM PC clone that had been produced-- an offer that IBM had to know would be completely unacceptable.
The cloners told IBM to pound sand, and went on building ISA-based IBM PC compatibles without paying IBM a dime, until PCI came on the scene. MCA went nowhere, and IBM's PC business continued its downward slide that started when Compaq was first to market with an 80386-based machine, and finished just last year when they sold the division to Lenovo.
~Philly
Time to bust out the holy wars.
p resentations/MC680X0OPTAPP.txt
6 8k.htm
2 14.asp#ADC
I like the 68000 because it has so many registers but I think all in all in the 80386 is the better CPU.
For reference, consider:
http://www.freescale.com/files/32bit/doc/reports_
http://www.df.lth.se/~john_e/gems/gem0028.html
http://linux.cis.monroeccc.edu/~paulrsm/doc/trick
http://www.csua.berkeley.edu/~muchandr/m68k
Right off the wheel, we notice that the 68000 did not support 32 bit multiplecation at all. Doesn't sound too much like a 32 bit chip to me. Compare that to Intels quirky IMUL, which I believe puts the result into EAX, EDX to get a real 64 bit result.
Integer math was faster clock for clock on the 386. Compare things like 68K register addition to Intel register addition. There's no comparison.
Compare
http://www.gamedev.net/reference/articles/article
to
http://www.df.lth.se/~john_e/gems/gem0028.html
Whenever you did any 32 bit pointer math on a 68k, you paid a huge, huge performance penalty. It was always more efficient to do things in 16 bit PC relative addressing.
The 68K had no concept of isolated memory or tasks. So systems like the Amiga and the Macintosh would run without any isolation between processes. I was an Amiga fan boy and I used to get that GURU meditation error so much that it was not even comical.
The tragedy of the 386 architecture was actually Microsoft and not Intel. DOS and Windows did not use even the 386 chip to its fullest capability for memory management. MS users would have to wait until Sept 1995, almost 10 years after the 386, for a true 32 bit operating system.
This is my sig.
The Fat Mac had neither a hard drive nor cooperative multitasking (unless you count desk accessories, but the original Mac had those too). There was Switcher which gave the ability to switch between apps, but there was no multitasking -- the background apps were completely suspended. Cooperative multitasking didn't come 'til Multifinder with System 6.
Internal hard drives didn't come 'til, I want to say Mac II? Was there one for the SE?
-- It only takes 20 minutes for a liberal to become a conservative thanks to our new outpatient surgical procedure!
There is a huge 6 year gap between the two. I could say a powerpc G5 processor is alot better than an 8086. Well of course it is. But does that mean a fast Athlon64 is slower than a G5?
By the time the 80386 came out, Motorrola had 60020's and perhaps 68030's.
PS the 68020's and I think the 68000's could run Unix because of built in memory protection and other features. Could 8088's, 8086's, 80186's or 286's do that? No I do not Consider early SCO XENIX aka Openserver a real unix with built in memory protection and primptive multitasking until well after the 386.
http://saveie6.com/
OS/2's Workplace Shell is generally considered the best example.
Things are starting to recover a bit with "AJAX" due to Javascript's more dynamic character. However, Gates can't get the garbage collector to work properly with IE for some reason so it can't be used as an application platform. Whether this is deliberate or not it certainly has helped protect his monopoly position by preventing web browsers from being becoming a viable cross-OS platform for network service applications. We'll have to wait and see if Ray Ozzie fixes the garbage collector before passing judgement. Thus far, it appears IE7 beta's garbage collector is still shit.
Seastead this.
High-end PC:
High-end PC: 640 KB RAM. No further expansion possible
LISA: 1 MB RAM, expandible to 2 MB.
High-end PC: One 360K floppy, one Hard Drive
LISA: Dual 860K floppies, 5 MB Hard Drive.
High-end PC/Hercules Graphics: 720x348 bin-mapped display (plus 80x25 chararcter)
LISA: 720x364 bit-mapped display