LisaOS to be open-sourced and donated to Computer History Museum | Hacker News Reader

LisaOS to be open-sourced and donated to Computer History Museum

twitter.com

259 pointsmpweiher84 comments

In the earliest days of the Mac (1984) you had to have a Lisa to develop for the Mac, because that was the machine the Pascal compiler ran on. I bought a Lisa and a Mac for this reason. It'll be fun to see the source code.

Wasn't that enormously expensive? Was it purely for fun or did you recoup your investment?

It sure was. The Lisa originally sold for $10,000 in 1983. The price dropped to about $5k (IIRC) for the Lisa 2 which came out the same day as the original Mac, but that was still very expensive.

128K Mac ($2500) + Lisa 2 ($5000) = $7500 in 1983 (~$18,600 in 2017 dollars).

To put that into perspective, in the early 1980s the average new car was around $7,000 to $8,000 and the median new home price was ~$75,000 in 1983. It'd be like a personal computer costing $30,000+ today, in relation to those other major purchase items.

True. But the first time I saw a Mac in person, I knew it was going to change the world and I had to develop for it. I spent all my savings on a Lisa and a Mac. It's hard to describe today just how radically impressive that first Mac was.

Jeff Han's 2006 Ted Talk featuring a multitouch demo was the only other time I've seen a user interface demo and instantly knew a UI paradigm was going to be a very big deal.

https://www.youtube.com/watch?v=ac0E6deG4AU

In comparison, the Xerox Star was a mere $16,500, which would be a bit more than $40,000 today.

Jesus, that's as much as I paid for my brand new car recently...

Yes! But I was a student at UT Austin, which had a special student purchase agreement with Apple. Steve Jobs even came to campus and gave us a pep talk about it. So I got my whole setup for something like half price.

In the earliest days of the Mac (1984) you had to have a Lisa to develop for the Mac

Looks like the concepts of strictly separating users and developers, as well as paying extra to develop for the platform, have been with Apple for a very long time. From what I've read, even developing for and on the Lisa itself was a huge hassle.

The other concept which Apple might've pioneered with the Lisa is forced obolescence --- the software as well as the custom RTC chip were designed to not handle dates beyond 1995, less than 15 years after it was released: https://images.techhive.com/images/article/2013/01/lisa_cloc...

(In contrast, while the original IBM PC didn't have an RTC, DOS was able to handle dates up to 2235. Plus, for better or worse, users and developers were considered simply different extremes of the same demographic for much of the PCs early history --- computer magazines would print Asm listings or hexdumps to be entered in DEBUG, essentially letting users gradually become developers too.)

I certainly agree the Apple Lisa was "ahead of its time"... but not in a good way.

>the software as well as the custom RTC chip were designed to not handle dates beyond 1995, less than 15 years after it was released

Why did they do that? Did they have an explanation that doesn’t just boil down to “we wanted customers to keep upgrading”?

The year was stored as a 4-bit integer

> we wanted customers to keep upgrading

After 15 years? Who was going to be seriously using a 5 MHz Lisa in 1995 (yes you can spare me the "this broke school district is still using an Amiga to run their HVAC!" stories)

The C64 was introduced in 1982, discontinued in 1994, and remained in widespread use for several years after that.

If it means anything: 2^29 seconds is 17 years almost exactly

I didn't know about the RTC issue, but the separation of users from developers was strictly because of hardware limitations. The Lisa cost $10k and came from a separate division of Apple from the Mac, which sold for [around] $2500 and only had 128k of RAM and no hard drive. They were completely different target markets. Nobody at Apple at that time knew how to make a Pascal compiler run in the limited environment of the Mac. (But as mentioned above, the Forth vendors had no problem doing it.)

Does anybody know how to make a Pascal compiler run on such a system? The executable for Turbo Pascal for x86 was 39kB and likely needed more to run on a character-based screen.

After subtracting video memory and sound buffers (combined about 22kB) and what the OS used, that 128kB Mac had, IIRC, about 28kB available for applications to use.

I think the biggest issue was probably the IDE support and the Object Pascal features, given the existence of Pascal compilers on 8 bit OSes.

I owned a few ones for ZX Spectrum, but they were basically standard Pascal + Z80 low level extensions.

And plain single pass compiler with zero optimizations.

Reading https://archive.org/details/bitsavers_applelisahManualSep82_..., that clock runs on a custom CPU that also handles the keyboard, the mouse and software power off.

”The keyboard/mouse control is performed by a dedicated slave processor, known as a COPS. It provides communication with four peripheral devices via the A port of the keyboard 6522 VIA port controller.

[…]

The COPS receives power from the backup supply. This voltage is available at all times, whether the system is powered down or even unplugged. The only time the COPS will cease functioning is if the battery is allowed to run down by having the system unplugged over a long period.”

Given that and the time, it doesn’t surprise me they weren’t over-provisioning bits for that clock.

Funny enough the Lisa ended up being used to develop a competitor's OS as well... Digital Research used the Lisa to develop its initial versions of GEM (the GUI that ended up on the Atari ST and on some MS-DOS systems and was the underpinning behind the Ventura Publisher DTP system) and GEMDOS and I believe also CPM 68/K.

As the sources for GEM were open sourced years ago, so was the build environment and images for getting it running on the Lisa. I wasn't able to get them to boot on any of the Lisa emulators out there though.

GEM on Amstrad 1512 takes me back

Of course, part of the reason is the only 128K of RAM. It was only one year later that 64K and 256K DRAM prices fall fast, leading to US makers leaving the market and dumping charges.

A friend of mine started a company in 1983 to develop Mac software. They didn't want to go the Lisa route for a couple of reasons. First, they wanted to write in C, not Pascal. Second, they wanted several developers working simultaneously and could not afford to get a Lisa for each.

They knew that MacFORTH was going to be available pretty much concurrently with the release of the Mac. They knew they could use that to write programs on the Mac to explore and reverse engineer things.

They bought a Callan Data Systems Unistar 200 Unix workstation [1], which was a 68k based all-in-one non-graphical Unix workstation. They also got a bunch of serial terminals (ADM3a's, if I recall correctly) and hooked them up to the Callan, and gave each developer a Unix account on the workstation so that several developers could be working at once.

The idea was to write in C on the Callan, use its compiler and assembler, and then somehow turn that object code into code that would run on the Mac.

They had the Callan connected to the Mac over a serial port, and a program in MacFORTH that would accept object code from the Callan and invoke it. With this setup, their developers could start working on parts of the product, with test harnesses in MacFORTH to let them test on the Mac.

My memory is getting fuzzy, but I think somewhere in there they wrote a Mac linker on the Callan that could generate Mac executable files, which they could send to MacFORTH on the Mac and have it write them to floppy.

[1] I was a systems programmer at Callan, and was able to get my friend a nice discount. I think it came to $5000 for a nicely configured workstation.

I used Mach II Forth as well as Pascal. Mach II was a great resident compiler on the Mac and it was a joy to use. IIRC, MacForth was interpreted but in any case yes, several other languages were resident on the Mac before the "official" Apple Pascal was available.

I think a better name for this would be Mac OS series 0.

Going back to being serious seeing the source code for LisaOS sounds exciting . Apple changed my life when I used an Apple 2 in grade school. My first computer I got for Christmas was a G5 iMac. In college I used MacBooks and they replaced one with a newer model. I’m not sure I would have my current job if I wasn’t in Steve’s paradise .

Actually there was a patch for the Lisa that allowed it to run Mac OS and software (maybe called a Mac XL?). I believe it was a combination of ROMs and disks but it's been a day or two since I dealt with that so I may be mis-remembering.

I believe you're correct; Mac XL sounds right to me, too.

Yup, Macintosh XL (https://en.wikipedia.org/wiki/Macintosh_XL). I remember a company (forget the name, likely long gone) at a Macworld Expo (in Moscone!) in the late 80s or early 90s selling refurbished Lisas and Mac XLs.

Sun (something or other)... they were an Apple reseller that not only did the refurbs but also wrote the software that allowed the Lisa to run 128K ROM Mac stuff.

SunRem I think

ResEdit changed my life as a kid. I basically made a full game conversion at age 9 using ResEdit a basic draw application, and a Midi keyboard over serial on the family LCIII

LisaOS really was a different beast than Mac OS, and very little code was shared between the two from what I've read. Just some QuickDraw code is my understanding. LisaOS was multitasking and more object oriented and more like a true OS from the start.

I'm super excited to be able to read the sources now.

Yep. In some ways LisaOS was ahead of where MacOS is now. It had stationery pads, for example, and it just felt snappier. On a processor 1/1000 as fast as a current Mac.

macOS actually has stationery pads! https://www.macworld.com/article/1161619/software-utilities/...

But Lisa OS was amazing, and was ahead of the original macOS in many ways: preemptive multitasking, way nicer error messages, hierarchical files system, a more fully fleshed out desktop/document metaphor, etc.

I think the Lisa OS's error messages are still better than macOS's.

The GP quoted "MacOS" when the correct capitalization would indeed have been "macOS". What you're referring to is most correctly termed the Macintosh System Software, which is way too long and could also be referred to as the System Software or System.

The camelCase capitalization is a 2017 thing, not a 1984 thing.

</Neurotic pedantry>

I'm not sure what point MacOS got its name.

It was around the time of System 7.5 when the MacOS bootscreen was added I think. But names like “System 7.5” were also used. A confusing time!

Pretty sure it was System 7.5, and then Mac OS 7.6.

Thanks.

Thanks for the nitpick :). I’m usually really picky about that stuff too. I have Macs from the 128K running System 1.0 though macOS 10.13 High Sierra.

Wow, awesomeness. :)

(I need to get myself some classic Macs at some point.....)

Wow, thanks for sharing that link. It always amazes me that after so many years with macOS I can still regularly learn of new features.

I feel that calling it "Mac OS series 0" comes too close to a touchy subject for it to be considered a better name.

Raskin's original vision for the Macintosh project was a text-based UI. (The Canon Cat was a descendant of that design.) Only later did Macintosh gain support for the Lisa-style graphical programs.

I don't know the timeline, but I can point to https://en.wikipedia.org/wiki/Macintosh which says:

> Smith's first Macintosh board was built to Raskin's design specifications: it had 64 kilobytes (kB) of RAM, used the 8-bit Motorola 6809E microprocessor, and was capable of supporting a 256×256-pixel black-and-white bitmap display. Bud Tribble, a member of the Mac team, was interested in running the Apple Lisa's graphical programs on the Macintosh, and asked Smith whether he could incorporate the Lisa's Motorola 68000 microprocessor into the Mac while still keeping the production cost down. By December 1980, Smith had succeeded ...

Then at https://www.folklore.org/StoryView.py?project=Macintosh&stor... you read that "the very first prototype of the very first Macintosh sprung to life early in the first month of the new decade, in January 1980" with the 6809E. That would be before the 68000 was added.

If that prototype - or any prototype before the 68000 was added - had an operating system, then that should be called "Mac OS series 0."

Wow, the 6809 was the Chrysler Cordoba of 8 bit microprocessors! It was tastefully luxurious and highly orthogonal, sporting a powerful hardware multiply instruction, and soft Corinthian Leather.

https://en.wikipedia.org/wiki/Motorola_6809

https://youtu.be/tfKHBB4vt4c?t=10s

But I'm sure glad they ended up using the 68K for the Mac.

Using a 6809 they could have made a far cheaper, far more accessible machine. Moving to the 68000 meant moving to more expensive support chips, and required more RAM, and the like.

But it did mean they could use the QuickDraw code from the Lisa. And in the end the Mac interface ended up being a kind of dumbed down ripped off version of the Lisa, instead of the Jeff Raskin thing with hot keys and the like of early prototypes.

I just wish they'd worked instead on making an actual cost reduced Lisa, using the same Lisa OS, but without the crappy copy protection and the overpriced hardware...

Looking back, I wonder how much the switch from the cheap 6502 to the expensive 68000 killed the original 8-bit companies. Both Commodore and Atari had higher market shares than the Apple II. It seems like playing in the higher price bracket moved them too close to the PC. It feels like people making the opposite decision of the 6502 over the 6800.

Plus, I did love the 6809 and its ISA. It was fun to program.

Tramiel's Atari Corp did manage to keep 8-bit-ish pricing on its ST line, despite the 68k switch.

But yeah, if the 65816 had come out a year or two earlier, I wonder if Amiga and Tramiel and friends would have chosen to develop on that instead. An 8/16 processor that could address larger memory regions but still could use 8-bit memory and 6502 support chips, and had the cycle efficiency and simplicity of the 6502. Apple came out with the IIgs based on it, but overpriced it.

Well, the 520ST was $799.99 for the monochrome computer, so it was quite a bit more expensive than the Atari 65XE at $99.99. If I remember correctly the 68000 itself cost north of $200 in quantities, so it was impossible to stay in the old price range.

I remember a story of Commodore / Amiga trying to use the Amiga chipset with a cheaper processor. Of course they also tried it with a PA-RISC.

I thought the 68000 was down to something like $10 by that time? Given that, the price discrepancy is surprising, since the ST was intentionally a rather simple design. What exactly made the ST have to cost that much?

Indeed this article seems to indicate that the 68000 was $15 by 1984: http://www.nytimes.com/1984/06/29/business/motorola-s-powerf...

But 512KB or 1MB of RAM was really expensive then. Atari cheaped out a bit by running their RAM a bit over its spec, tho. There were a few custom chips in it too (the Shifter [video], the MMU, the GLU [logic], and DMA chip.

Those who put down the 6809 have never programmed it. It was (and still is) a joy to work with and way more powerful than the other 8 bit chips of the era.

This is a 6809 at 1 MHz and no special graphics hardware: https://youtu.be/yrkx6vuiMrE

Two stack pointers! Great fir Forth impls. In the early 80s UK home micro mkt the Dragon was 6809 based.

The Vectrex had it.

Geh, in reality the Mac didn't get an operating system in the true sense of the word (proper process separation, memory protection etc.) until like... System 7.6 or maybe even later in OS X. Classic Mac OS was more like an overgrown API than a proper OS environment. A real step backwards from the Lisa...

It's baffling to me that instead of cost reducing the Lisa, Apple ran two parallel product lines (Mac and Lisa), similar to each other and competing with each other.. and from reading histories it sounds like it came down to territorial turf wars and Job's ego that made this the case. Lisa wasn't Jobs baby... so it was killed.

Lisa wasn't Jobs baby

True in one sense... false in another!

If I accept your "True sense of the word", then many early operating systems, like the "SHARE Operating System" from 1959, are not, actually, operating systems.

I prefer to believe that you are using a stricter definition of the word/phrase, rather one which encompasses how that word/phrase was actually used by practitioners in the field.

Sure, fine, I agree that back then the definition of OS was more fluid. Though in reality I don't think that in those early days they even called what was running on the Mac an operating system at all. The term "Mac OS" didn't arrive til much later, and it wasn't until there were later revisions that they used the phraseology "System" for it.

But to the original point, my understanding is that the Lisa had a more classic OS architecture when compared to the Mac: multiple processes, dynamic memory allocation, etc. But until I can actually read the source for it (soon!) I'm just repeating hearsay. I've run the OS on emulators, but that doesn't tell you much about its inner workings.

"in reality I don't think that in those early days they even called what was running on the Mac an operating system at all"

I must live in a different reality.

The first Apple operating system, with the phrase "operating system" in the name, was Apple DOS, starting in 1978. See https://en.wikipedia.org/wiki/Apple_DOS . It did not have the features you require in your definition of an operating system.

Here's the summary from the bibliographic entry for an article from "Popular Computing", Aug 1984, quoting from http://www.worldcat.org/title/whenever-you-use-a-computer-yo... :

> Examines design, features, and shortcomings of eight disk-based operating systems designed for general use that are popular or most likely to affect the future of microcomputing. Included are the CP/M family, MS-DOS, Apple DOS/ProDOS, Unix, Pick, the p-System, TRSDOS, and Macintosh/Lisa. (MBR)

Several of those have the term "operating system" in the name, and as you can see "Macintosh" is described as an operating system.

Here are internal development notes from the Macintosh project, dated 1983, which shows that Apple had a "Macintosh Operating System Manual" already by that time: https://archive.org/stream/MacintoshDevelopmentNotes/Macinto... .

Chapter 1 of those notes is titled "Overview of the Operating System". https://archive.org/stream/MacintoshDevelopmentNotes/Macinto... The introduction is "There are two main groups of routines that may be called by an application program -- the MacIntosh Operating System and the MacIntosh User Interface Toolkit." (The capitalization in "MacIntosh" is theirs.)

I know what your intent is, and I think you are correct. What I disagree is your use of phrases like "True sense of the word" and "in reality" in expressing what seems much more like a personal opinion based on a shallow understanding of the topic.

> Geh, in reality the Mac didn't get an operating system in the true sense of the word

Indeed.

Apple's operating systems were completely outclassed by Windows 95, AmigaOS and BeOS (and by that I mean both at the OS level and also the GUI).

Even the "merger" of NeXt and Apple didn't give Apple a credible OS for many years after that.

This is why I can't help chuckle when I read comments such as "Lisa was programmed with planned obsolescence because it couldn't support years past 1995".

This is not planned obsolescence, it's just incompetence from Apple's software engineers. Back then, Apple was a lot better at product and hardware than software.

And when I see iTunes, I still think it's still kind of true today.

NeXtStep was an excellent BSD-based OS, and Mac OS X, released (I believe) first in 2000 (3 years after Apple bought NeXT) was basically NeXTStep with a new UI, new graphical apps, and new APIs to bridge the gap to OS 9.

> Geh, in reality the Mac didn't get an operating system in the true sense of the word (proper process separation, memory protection etc.)

What? No. So DOS wasn't an operating system either then?

Lots of higher level OS features just weren't really possible with the extremely limited RAM, ROM, and complete lack of hard drive of the original Mac. That doesn't mean it wasn't an "operating system"

Lisa was the really expensive Apple aimed at businesses right? I think I've seen it mentioned on the 8-Bit Guy's show.

I wish I had visited the CHM when I was out on the west coast. I'll have to add it to the list for next time.

I have been to the Living Computer Museum in Seattle. For those of you travelling there, I highly recommend it and Recycle PC (in the same area). They've got a ton of old working systems you can actually play with; everything from TI-99s to HP-UX machines to a NeXT system to a fully working Alto.

+1, the running Alto was the highlight of my trip there.

Apple /// was their first stab at a business computer. Too late as the 16 bit revolution was already underway.

IIRC, it was around 10k.

The Living Computer Museum is incredible. Worth noting in the context of this article is that among the many machines they have working and available to play on is a Lisa 2 running Lisa OS.

I bought a Lisa system and lots of accessories at a garage sale when I was a kid. The guy had kept all the receipts for everything, and the total, including a couple of 5MB and one 10 MB hard disk, was over $20,000. He used it for business, and it had some business apps that I don't recall. I sold the whole pile of stuff for a few hundred dollars (and I paid less than that for it, I don't remember how much, but I was usually spending less than $100 on 8 bit stuff or PCs that needed work, so it couldn't have been much), and it's probably the machine (of many really cool/weird/rare things) that I wish I'd somehow kept through all those years. Such an interesting piece of history. But, I was a Commodore fanatic at the time, and didn't really get excited about Apple products (and still mostly don't, kinda for the same reason, summed up by an old Tramiel slogan "for the masses not the classes").

And, yes, the CHM is well worth a visit. I used to take everyone who came to visit me in Mountain View to see it, even if they weren't nerds. It is an enjoyable and educational afternoon, I think, for anyone with even a little curiosity about the history of the devices we all rely on.

Have you seen the TV series "The Americans", which is set in the early 80's, and has spot-on era props and scenery? They have a Commodore Pet (with the adult-sized non-chicklet keyboard ;) in their real estate office! It makes sense that deep cover KGB spies like Phillip and Elisabeth would have the computer for the masses not the classes, comrade!

The kids Paige and Henry even have the exact same Star Wars blanket I had (which their parents tuck them into at bed time, then sneak out and skulk around stealing Star Wars secrets from the US Government).

I haven't, but I've heard good things. I was watching Halt and Catch Fire which is also fantastic bit of history on several fronts (in addition to the computer history on the show, I happen to also love punk rock from that era, especially Texas punk rock, like Big Boys and the Dicks, etc., and so the soundtrack is fantastic), but I kinda drifted away from watching TV a while ago. I'll probably get back to it at some point, as I've heard the later seasons improved.

Pro tip, never keep receipts for computer gear you buy. If my spouse knew how much I spent in "then" dollars for IMSAI's, S100 systems, Apple II, Mac's, Lisa's, etc. instead of investing them in some kind of bond/stock/retirement they would kill me (for the insurance money).

It was expensive to be a pioneer, I think it was worth it, since it helped a long successful IT career. But other may not. There are posts later on going "I paid that for a car!" but to pick up an extra $100-300 a week makes it good in the long run.

Cool. I see there's a Lisa emulator that exists and works on the raspberry pi: http://lisa.sunder.net

Sounds like access to the source code might allow for some more progress.

That's pretty neat - from a computer history perspective. I assume the museum has a Lisa. It's one of the few Apple products I've never seen in person.

They not only have a Lisa, they have a Lisa 1, which is very rare. The last one that sold at auction (in the past few months) sold for around $50,000. I've seen estimates for the number remaining in existence as low as 30. It didn't sell well originally, but more importantly, Apple offered all Lisa 1 owners a free upgrade (via modification) to a Lisa 2 when it was released, and most took them up on it.

Lisa 2's are much easier to come by, especially as the "Mac XL" which sold decently as a high-end Mac. I've got a Lisa 2 running Lisa OS myself.

Wasn't this called the Lisa Desktop Manager and never 'LisaOS' ?

I'm looking forward to the Lisa hardware clones this will spawn...

This is so cool. I can’t wait to comb through it looking for great code comments from a bygone age. The folks who wrote the code for the lunar lander have set a high bar.

I remember when it was released the Hecht's department store at my local mall was selling them, at least trying to, and had one on display that I played with a few times. They also later briefly sold the original Mac.

I’m reading Jaron Lanier’s new book on VR and he mentions coding some of the earliest haptics on a Lisa.

This is a great Christmas present!

Duplicate of https://news.ycombinator.com/item?id=16000606

This one has more comments already so maybe that one is the dupe

Except the other one was posted two hours earlier...

Is the difference due to promotion or randomness? Reminds me of business competition.

Likely the descriptiveness of the title. This one has three “things” in the title - Lisa, open source, computer history museum. The other has just one - the Lisa. Nothing about the license, nothing about the museum.

Submitter here.

Getting to the HN front-page has a very large random component. That doesn't mean you can't improve the odds, in this case by providing an appropriate title. To me this seemed like a story that deserved attention, and the auto-generated title from the twitter post was sufficiently useless to warrant editing the title (which I usually don't do).

Also, this happens all the time and sometimes you will be lucky and sometimes somebody else will be lucky. I've had dups of my earlier submissions go to front-page so many times I can't really keep track. But it really isn't a big deal, this isn't exactly doing years of research and getting scooped by someone else filing a patent first or publishing more quickly. It's just a submission to a web-site, effort was a few clicks.

¯\_(ツ)_/¯

This reminds me of this debacle: https://news.ycombinator.com/item?id=10638220

Reply on news.ycombinator.com