Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
Desktops (Apple) Apple

Andy Hertzfeld Shares His Thoughts on 25 Years of the Mac 142

blackbearnh writes "It may make you feel very, very old, but the Macintosh will be turning 25 in January. As we approach this momentous anniversary, O'Reilly News had a talk with Andy Hertzfeld, one of the original Macintosh designers, about the long and storied history of the Mac. Hertzfeld, who tells the story of the Mac in his book A Revolution in the Valley, shares his thoughts about how the Mac has aged over time, how life might have been different if Steve Jobs had stayed on at Apple, and the differences between working for Apple, and for Google (his current employer.)" Read on below for a bit of what Hertzfeld had to say.

"They're very similar in certain ways — essentially both Apple and Google want to rewrite the rulebook; they don't want to do things in conventional ways. They want to come up with a better way — for everything; that's not even just the technology but the work processes, the work environment, everything has to be unique and better, so they're very similar in that way. One of the ways that they're different has to do with essentially trust of employees. Apple is very secretive within the company; people working on Macs don't know anything about the new iPods, et cetera. Google is extremely open within the company; once you're a Google employee you have access to just about every piece of information there is."

This discussion has been archived. No new comments can be posted.

Andy Hertzfeld Shares His Thoughts on 25 Years of the Mac

Comments Filter:
  • I'm curious as to why people are still interviewing Mr. Hertzfeld, given that his most recent successful project was the Mac. Even more puzzling is that he continues to be able to raise funds, attract developers, etc., in view of his decades-long track record of failure.

    I don't know why people give him money, but as for an interview subject, he was a witness to history.

  • by Anonymous Coward on Wednesday August 27, 2008 @05:56PM (#24771045)

    It's hard to create a wildly successful product once, let alone twice.

  • by fermion ( 181285 ) on Wednesday August 27, 2008 @06:04PM (#24771123) Homepage Journal
    Apple, Inc is a mature company with products that generate constant profits. Google is new company that has benefited from the fact that few people have more money than they know what to do with, and just need to invest it anything that looks halfway honest, and needs breeds of financial instruments that help hide whether a company is profitable of not. That last statement may sound a bit harsh, but banks, expert in fraudulent financial instruments, were able to create the illusion of profit in what we now know was in fact was not the case.

    We don't know if Google will work in the long run. And in the long run I am thinking AOL. Google's success depends on the advertising market tolerating secretive and random marketize techniques which appear to be abuse of the near monopoly that Google now has in advertising. The success is also dependent on the ability of cheap commodity severs to provide six nines service, externalizing the majority of the cost of content creation to third parties, and externalizing the majority of infrastructure costs to the taxpayer. I am not saying that at some point their house of cards will fall al a AOL, but I am not quite sure how they are going to make money off cloud computing, other than selling personal information collecting from the love letters of their users to third parties.

    All Apple has to do is come up with the next cool thing that people will pay for. This is not a simple thing, but something that Apple has been doing with some success for quite a while. We now see a diversification outside of computers, so, when the Mac OS does become something that is not limited to any machine, and when, by the same rules, MS is not able to limit OEM versions to run only on the machine it was originally shipped with, Apple will be able to enter this brave new work of zero profit computer equipment with new consumer appliances.

  • by Bryan Ischo ( 893 ) * on Wednesday August 27, 2008 @06:22PM (#24771381) Homepage

    Perhaps so, but he does have interesting things to say and a very intelligent way to say them. That was one of the best interviews I have read in a while, because both the questions and answers were intelligent and interesting.

  • by Anonymous Coward on Wednesday August 27, 2008 @06:31PM (#24771537)

    Highly insightful. The Mac was like the old order, where one company made hardware, OS and software. The PC is part of the new order.

    Hard to say -- the difference is the Mac competed directly and enthusiastically with the openly licensed ecology. That's a real difference with any "old order" you might be referring to. Also, had they open-licensed they would have probably failed to retain the same polish, and become a pretty footnote like Amiga and Atari.

    And of course this "old order" is what's having a 25th anniversary here, and doing just fine. Probably best not to use terms like that.

  • by againjj ( 1132651 ) on Wednesday August 27, 2008 @07:08PM (#24772075)

    [Google's] success is also dependent on [...] externalizing the majority of infrastructure costs to the taxpayer.

    Where did that come from? I can not think of anything here that fits. Clearly you are not talking about Google-owned infrastructure like their data centers Perhaps you mean internet infrastructure, since this was funded in the past by the government. But then, that is true about the electrical/sewer/telephone/etc. infrastructure. At this point, the initial investment has been paid, and none of these are funded by the government. All services are paid for. Google pays well for good Internet connectivity, and users pay for connectivity; each Internet provider pays its providers recursively, funding all through that money. Where are the charges getting externalized to the taxpayer? Roads?

  • by RetiredMidn ( 441788 ) on Wednesday August 27, 2008 @07:10PM (#24772099) Homepage

    It was probably the decision to openly license it. The Mac--when the Mac came out and for two years thereafter it was at least four or five years ahead of Windows and possibly could have taken the place of Windows if it was openly licensed, but because the Macintosh was restricted to a single member, Apple, it never could become an industry rather than a single platform.

    Highly insightful. The Mac was like the old order, where one company made hardware, OS and software. The PC is part of the new order.

    I disagree with Andy's assessment. The Mac may have been years ahead of Windows, but it's real problem, IMHO, is MS-DOS was already pretty entrenched, and the Mac didn't offer a migration path. I was working for Lotus at the time (working sometimes on the Mac, sometimes on DOS), and we had a pretty large community of 1-2-3 users who would not leave behind their accumulated DOS spreadsheets and what-not for the Mac even if they wanted to.

  • Re:G6 dreams (Score:4, Insightful)

    by TheRaven64 ( 641858 ) on Wednesday August 27, 2008 @07:12PM (#24772127) Journal

    OS X currently supports three instruction sets; x86, PowerPC and ARM. Most software can use any with a straight recompile (the decision to use UIKit instead of AppKit on their ARM-based platforms makes porting GUI apps slightly harder, but this is orthogonal to the question of the CPU). Very little code directly depends on things like AltiVec or SSE. Code that does, often uses libraries for common algorithms (FFT, and so on) which just uses the correct code for the current platform. Other code uses generic vector support in GCC and LLVM, which is compiled to whatever instructions the host architecture supports.

    I wouldn't be surprised if PowerPC surfaces again at some point in the future. Freescale, in particular, make some very cheap parts that would make sense in successors to the AppleTV.

  • by InterruptDescriptorT ( 531083 ) on Wednesday August 27, 2008 @08:41PM (#24773277) Homepage
    I just wanted to take a few bytes of badwidth to say that Andy Herzfeld is one of my personal heroes and should be to any creative, true old-school hacker-type programmers/engineers out there. Among other things, he is the father of the desk accessory in the original Mac OS.

    The original Mac had 128K of memory, some 27K of which was used for the screen buffer alone, and although much of the OS was in ROM, it used a significant amount of the available RAM for itself. And this isn't even to mention any currently running application. A Desk Accessory, then, and the ability to invoke it while an application was running (many people forget that the original Mac OS was not multitasking at all), required some pretty incredible feats of programming to make it fit in the tiny amount of memory left. And he found a way to make it work.

    People often speak in awe of how the 512K Amiga did multitasking on its tiny memory budget, and while I also admire that effort (especially having been a Commodore kid from VIC 20 to C64 to Amiga), I still think the original Mac OS represents one of the most incredible feats of software engineering of the early microcomputer era. I get slightly down every so often when I think about how modern developers, including myself, have gigabytes of memory and ultra-fast processors to work with and don't often have to think about the resource consumption of their algorithms/designs. Must have been so cool to work that kind of stuff back there...

    Fawning mode off now...
  • Re:25 years of... (Score:2, Insightful)

    by LKM ( 227954 ) on Thursday August 28, 2008 @02:15AM (#24775631) Homepage
    The issues between the KHTML and WebKit teams did not last two years, and have long since been resolved. If you have to dig that deep to find some dirt, I count that in Apple's favor.
  • by peter303 ( 12292 ) on Thursday August 28, 2008 @10:02AM (#24778757)
    The key player that "saved the Mac" was Steve hiring Tevanian from CMU who understood modern software technology and software engineering (originally for NeXT). Andy's group did a great job for its time. And the Xerox PARC crowd, while making brilliant inventions, did not how to engineer something durable for the market (all PARC commercial software failed). But the early Mac code was pretty "cowboy" and brittle. Apple was desperately shopping for a "workstation generation" operating system in the late 1990s, considering where to go with Sun, BeOS or NeXT. All three would have been good choices from a technology point of view, but they got Steve back with #3.
  • Re:25 years of... (Score:3, Insightful)

    by Ash-Fox ( 726320 ) on Thursday August 28, 2008 @11:18AM (#24779867) Journal

    Apple is one of GCC's biggest contributors.

    For OS X platform specific code.

    Or maybe you're thinking of WebKit.

    Which took so many years for the original KHTML team to decipher and make it useful because of how Apple released the sourcecode initially.

    KDE will be using it as well.

    Indeed, there is no point for them have projects running in parallel, thus the KDE team's decision to unify efforts which Apple didn't do initially is quite beneficial.

  • by suck_burners_rice ( 1258684 ) on Thursday August 28, 2008 @01:10PM (#24781487)

    Apple is very secretive within the company; people working on Macs don't know anything about the new iPods, et cetera. Google is extremely open within the company; once you're a Google employee you have access to just about every piece of information there is.

    Well, Google is, after all, the company that wants to make all information transparent and available to everyone. Apple, on the other hand, is an often-imitated company that must get its product to market before someone else gets a mimicked product out there. Once its on the market, imitation is the sincerest form of flattery, but if something like that shows up earlier, it pisses Apple off.

  • by oblivionboy ( 181090 ) on Friday August 29, 2008 @02:28AM (#24790851)

    The third era is MacOS X. The big issue now is that low-end PCs can do everything I need, and low-end PCs are insanely cheap, so why buy a mac?

    Uuhh...because I want to?

    I've noticed on /. that there are people who like linux, and people who like Mac. And the reasons for this are different and have changed over time. I've used ubuntu, and even found recently the two floppy disks I used to install Linux 0.11 on -- well boot up anyways. In those days getting it to install on a hard drive was a real days work.

    I suspect that there is a segment of /. geekdom that doesn't care about whatever it is that OS X brings to the user. They don't care about usability, they don't care about how nice it looks, they don't care about how it all just works. Its simply not important to them. And they come up with posts like yours and talk about the price, as if this was the only thing to compare. As if PCs and Macs are all equal. And really, they're not. In fact I recently had the pleasure of using a Mac with Parallels, and was pretty delighted with how well Windows ran in Mac OS X. Pretty sweet actually.

    But I KNOW there is another segment of /. geekdom -- including me -- that just love the way the whole OS X thing works. We may or may not have been Mac fans before (and there's alot to dislike about the Mac lineup -- the entire iBook G3 line up and motherboard failures I'm looking at you, as well as the latest stuff about the Nvidia chipsets on the MacBook Pro), Mac OS X is just pretty sweet. For us. Those that like it. So why buy a Mac?

    Well because I want to. Thats why.

  • Re:G6 dreams (Score:3, Insightful)

    by Ilgaz ( 86384 ) on Friday August 29, 2008 @06:16AM (#24791977) Homepage

    You know what? If some stuff takes off such as ultra-ultra light devices which can perform as a laptop with comical power needs, they are always powered by FreeScale chips. As Apple policy/ideology is OS X should only run on Apple hardware, what does it mean?

    Perhaps MS keeps their PowerPC Windows maintained just in case? It wouldn't be surprise. If you look at Windows NT history, they had to deal with that horrible Intel RISC chip so they won't accidentally code x86 specific code in any case. MS still keeps saying "i386" on various directory and windows update sites for a reason.

    If there are people who gave up Altivec because Apple dropped PPC, don't forget, even POWER6 has Altivec now and no reason for FreeScale not to include it.

Can anyone remember when the times were not hard, and money not scarce?