Home > Apple > Pink and the PowerPC Pink and the PowerPC Submitted by Saad 2005-09-10 Apple 29 Comments Apple pinned its hopes on a project started in 1987 to provide a next generation version of Mac OS, called Pink. The operating system never shipped, but it prompted Apple’s adption of the PowerPC. Read the story here. About The Author Thom Holwerda Follow me on Twitter @thomholwerda 29 Comments 2005-09-10 8:03 pm could pink be opensourced? 2005-09-10 8:04 pm MikeGA Interesting stuff. Why do Apple seem to have had so many attempts at a brand new OS in the past, many of them ending failure due to management problems. Or is it just that everyone else has done the same, but more people are interested in Apple and so no-one ever really hears about them? 2005-09-10 8:40 pm Microsoft BOB, lol 2005-09-10 11:37 pm rm6990 Microsoft BOB, lol BOB was not an OS, it ran on top of Windows 3.1. 2005-09-11 1:53 am mini-me Windows 3.1 was not an OS – it ran on top of DOS :p 2005-09-11 1:54 am ma_d But I wanted the last word! 2005-09-11 1:54 am ma_d “Plan to throw one away” –Frederick Brooks 2005-09-11 3:17 am pcummins Or is it just that everyone else has done the same? Actually if you dig around you’ll find literally dozens of OS’es that were developed and later dropped as the larger OS’es subsumed most of their functionality or had significant inertia to override the superior but smaller OS’es. Many are still around, however they’re only used by the most discerning of computer users for specific tasks. People should remember things like Plan 9, A/UX, AIX, Hurd, NeXTStep, OpenStep, BeOS, QNX, Tron, CP/M, and the many others… for if you ignore the mistakes of the past, you’re doomed to repeat them. If many developers had investigated these OS’es, things would be a lot better in today’s current OS’es. Having said that, Plan 9, Hurd, Tron and QNX are still around. NeXTStep -> OpenStep -> MacOS X migrated well, however there’s still a lot to do before we really manage to catch up to some of the advances that the less popular OS’es have shown us. 2005-09-11 3:40 am JLF65 You missed a branch… NeXTStep -> OpenStep -> MacOS X & GNUStep GNUStep is still going strong. They even have their own live CD. 2005-09-12 4:08 am pcummins True – I’ve taken a few liberties with my comment, GNUStep is more of a rewrite from the OpenStep specs than a direct descendant like MacOS X (especially if you’ve seen Rhapsody and MacOS X Server 1.x). I’d hate to get into the descendant tree for Unix and GNU/Linux though! I’d be keen to give it GNUStep and MacOS X programming a go more in depth at some point in time, they look pretty interesting. I find the event driven programming like Delphi easier to deal with than the MVC paradigm they offer, but I suspect it’s just lack of familiarity getting to me. 2005-09-11 12:17 pm jonas.kirilla While Linux definitely is the juggernaut of this GNU Age, I don’t see it replacing the mature, industrial leaders of big iron and embedded realtime (Solaris, AIX and QNX) just yet. 2005-09-12 4:14 am pcummins While Linux definitely is the juggernaut of this GNU Age, I don’t see it replacing the… industrial leaders… just yet. No, definitely not. There’s still heaps of room for specialised OS’es like QNX, Tron and Solaris. It’s a matter of picking the tool (ie, the OS) for the job (ie, embedded systems). Trying to make one tool fit all tasks is usually not a good idea if you’re not very experienced in making it work properly. What’s scary is people who rote-chant to use Windows for everything, or the GNU/Linux guys who insist that GNU/Linux be used for everything under the sun without a proper understanding or awareness of other options. (Or they refuse to admit there are other options available). I think the scariest people of all are the ones that flip from side to side with a 100% love/hate relationship (so they love GNU/Linux or Windows, then hate it next year, then love it again the next year) as you can be sure they have no concept of being able to assess tools properly, or even close to it. 2005-09-10 9:39 pm Why do we keep rehashing Apple news from the 80s?! Have we run out of all current Apple news and had to reach back to countless stories about Apple’s Lisa and that spread sheet application I keep seeing on OSNews and Slashdot. 2005-09-10 9:58 pm Hey, a few of us are interested in the history of operating systems too. It may even help people learn from history. After all, a company like Apple doesn’t want to end up with another OS project as poorly managed as Pink while end users (such as ourselves) don’t want Vista to turn out to be Bob 2006! 2005-09-10 11:00 pm StephenBeDoper I hadn’t known that Erich Ringewald (former Be Inc. engineer) had worked on the Pink project. 2005-09-11 4:19 am You didn’t think Gasse, Erich, and Sakoman just happened to meet on the giant campus did you? Remember, around the time Erich was heading up Pink he would have been sharing a warehouse with the Newton project headed up (at that point) by Sakoman, all of whom were working for JLG, who was one of the people pushing for Pink to run on the Jaguar… a hardware & software combination that JLG would occasionally mention as the inspiration behind the BeOS and the BeBox. 2005-09-10 11:10 pm with dozens of engineers under deadline pressure, that is totally different from everything else out there. Just getting everyone on the same page is extremely tough, regardless of how many specs and meetings you have. That’s why the big new products that ship often are modelled after someone else’s success, often one that had the luxury of being developed at a more leisurely pace. So Linux was modelled after UNIX, Windows copied the Mac (which copied PARC), .NET copied a lot from Java, OpenOffice took from MS Office, etc. Then people can be told, ours is going to be like that over there, only a little bit better and up to date. Paul G 2005-09-11 2:00 am ma_d I don’t like your reasoning at all. At the level Windows copied the Mac a spec would have done just as much. Linux was modelled after Unix, but that wasn’t a large project at that point… It is not, it was not long after that, but that’s probably the single most important technical decision that made Linux so popular: It was 386 Unix, and it was here now not when Berkely finishes. I’m not sure how much .Net copies Java. My understanding is that Java is actually a full, and highly complex, machine that’s theoretically buildable. Is .net? Besides, .Net was like a 3-4 year project, and I image Microsoft had a huge mindshare working on it in any possible way: Because it was the technology behind their marketing. I think communication is the hardest part of a large project. But I don’t think having some other closed program to copy helps much. It’s not like they can say “hey, this function bar inside this class foo isn’t documented, oh let me check what it does in this other program!” Or “hmm, this behaviour is missing from the spec, what should I do if they hit cancel? Let me check what that other program does … oh, it doesn’t have a cancel button.” 2005-09-10 11:16 pm parrotjoe An interesting article – and I recall this ongoing process that ended up as Taligent. However, at the beginning the author states that the Mac had saved the company in the mid-eighties. Nothing could be further from the truth. The Mac was a money loser at first and it took some time for it to be really profitable. During that time, it was the profits from the Apple II that kept things going. 2005-09-11 2:01 am ma_d It saved the company from its own damage! 2005-09-10 11:23 pm mini-me For those of us who are students of historical computing it would be nice to note the hardware requirements for the OS and where to get it. I would love to give this OS a whirl if possible 2005-09-11 5:14 am The blurb is slightly inaccurate. A much crippled edition of Pink, built by a rotating door of developers shipped Pink on PPC workstations. The article itself says so However, there are no resources on Pink as a technical entity anywhere that I know of. No screenshots, no discussion from the developers, just the same story of Pink’s history, over and over again, like some piece of OS oral folklore. Obviously, legal trouble will keep Pink from being open sourced, even if the code still exists – any software project with two large companies involved is going to be so entangled that it is unlikely to ever go anywhere. But is there someone with firsthand knowledge of Pink out there willing to talk about it? 2005-09-11 6:25 am nimble What a sad story. Developing two OSs in parallel, whereby the new one didn’t care much about the existing one must have looked rather silly even back then. And outsourcing Pink into a joint venture with IBM who were already heavily invested in a similar effort with OS/2 certainly didn’t help. Nor did the PowerPC switch. It never really paid off, but it sure diverted Apple from developing a solid preemptively multi-tasked and memory-protected operating system (Copland). So they ended up buying Nextstep and it took until 2001 for OS X to be released, when Apple arguably could have had a stable and modern desktop operating system long before Microsoft came out with the half-arsed effort that was Windows 95. 2005-09-11 3:23 pm I remember reading an article about Pink in 1993 in Wired and below is a hyperlink. http://www.wired.com/wired/archive/1.02/taligent.html 2005-09-11 9:57 pm Actually, the Taligent and OS/2 debacle is a bit more complicated than that. There was a version of OS/2 (briefly) released for PowerPC which itself, like Taligent, was microkernel-based. In fact, the MK was Mach, the binaries were ELF, and the DOS box was a full-on virtual machine. IBM’s (public) intent was that Pink/Taligent would be able to run OS/2 as well as Mac OS and native apps, tough I’m fuzzy about where OS/2 PPC Edition fits in, in terms of timeframe. I’m pretty sure that the two were developed virtually independently, though there was obviously a view from somewhere that the two projects would interoperate (i.e., OS/2-on-a-Microkernel would sit atop Taligent’s MK; Mach just being a stopgap). Some Taligent documentation is still floating around on some websites, but almost all of it pertains to “CommonPoint on AIX” – CommonPoint being the programming environment (similar to Cocoa or Carbon) which made up Taligent. CommonPoint got rolled back into IBM, and handed over to the VisualAge team. IBM’s free International Components for Unicode are a descendant of CommonPoint, at least in part. 2005-09-12 6:21 am Just search for java taligent to see where some of that work ended up. For example, http://www.sun.com/smi/Press/sunflash/1997-05/sunflash.970527.1112…. 2005-09-11 10:09 pm Bnonn Thom, no slight intended, but is there any reason you can’t run a spellchecker over your blurbs? It seems like just about every article you post has a typo or spelling mistake in it. It just looks bad is all. 2005-09-12 1:19 pm g2devi Here’s the full scoop on Taligent: http://en.wikipedia.org/wiki/Taligent Taligent/Pink/Blue had some good ideas. Judging from the Taligent coding guide, it looked as if they were trying to redesign NextStep in C++. 2005-09-12 1:25 pm Why did apple have so many failed attempts at new OS’s. Primarily it was a loss of vision. During the late 80’s Apple’s board had forced Steve Jobs out. For all his many faults, he was the driving force in development at apple. Apple had by the early nighties, ceased to be the creative force that it once was. The company became mired in an attempt to try to hold on to both it’s profit margins and its dwindling market share. Without Jobs at the head to be the visionary, the creative efforts of the developers were not embraced and several false starts occurred. This continued until Jobs came back bring with him Next. Now the company is centered in development again and we are seeing the results. They once again view development as the driving force of the company and embrace radical change rather than kill it.