Dr. Dobb's Journal February 1998
This issue of DDJ focuses on scripting and alternative languages. So who uses these "alternative" languages, anyway? Alternative programmers? Alternative. Hmm. In the music biz, "alternative" may be a good word, but in most contexts it carries a subtle stigma. But Mom, it's just an alternative lifestyle. Support Betamax, the alternative format. WB, the alternative network.
People who use "alternative" languages must be some sort of a fringe group; you know, those people, the crazy ones, the ones tolerated with amused forbearance by the homogeneous majority. The RC Cola drinkers. Alternative language users probably are attracted to alternative operating systems, too. Maybe they hang out at the OSNews site at http://www.osnews.com/, where they read about NOSes, Javoses, Linuxen, Bees, Newts, Infernos, and other fringeware. Or the rumored German MacOS8compatible operating system COS.
Many of them are probably left-handed, like DDJ reader Richard B. Talley, who writes:
Since lefties are [a large] minority one would think they might have some rights! I've instructed number-two son, who is left-handed, that one of the accommodations he will need to make to a right-handed world is to read "outside button' for "right button" and "inside click" or "index click" for "left click."
Another reader points out that Xerox PARC called those two mouse buttons "point" and "adjust," but apparently Microsoft, receiving its Xerox wisdom through the Apple one-button filter, lost track of the functional names.
As you might guess from these samples, I get a lot of mail from these alternative types. Imagine: Obsessing over what Microsoft calls its mouse buttons. As if anyone really cared. Alternative language users would probably buy a book about Apple Computer, like Apple: The Inside Story of Intrigue, Egomania, and Business Blunders, by Wall Street Journal reporter Jim Carlton (Times Books, 1997, ISBN 0-8129-2851-2). And be offended. And you just know they'd defend their alternative platform fervently.
Back to the subject at hand. Alternative languages, alternative platforms, alternative paradigms. Alt dot anything.
Parallel programming is certainly an alternative paradigm. Several paradigms, in fact, depending on whether you're talking about multiple instruction streams, multiple data streams, or both. And there are many approaches to each of these different models, making even more paradigms, branching off to incomprehensibility. Maybe standardization will finally bring some sense to the parallel paradigms.
Digital, IBM, Intel, Kuck & Associates, and Silicon Graphics have defined OpenMP, a new industry-standard API for multiplatform shared-memory programming on UNIX and NT platforms.
OpenMP defines a set of standard compiler directives that let Fortran programmers (and soon C and C++ programmers) express shared-memory parallelism. The idea is to let programmers introduce parallelism at any level of granularity, making it easy to introduce at least some parallelism into a program without the fundamental rewriting that, say, message-passing parallelism would require.
New topic. HyperCard is an alternative something. No, wait a minute; it fits the issue theme by virtue of having a scripting language. Yeah, that's it. Anyway, I never miss an opportunity to write about HyperCard. I haven't missed any opportunities to write about it over the past year or so, because there's been nothing to write. That will change this year. The integration of QuickTime and HyperCard that is underway should have some interesting consequences. Apple hasn't told me exactly what and when, but I would expect significant performance increases, radically improved control of static and moving images, and cross-platform playback for MacOS, Windows, and Rhapsody.
I'd also expect HyperCard to become an even more compelling platform for multimedia development. The developers of Myst use it, and they have wondered for years why everybody in their field doesn't. Next year, maybe everybody will.
New topic. Only Steve Jobs could turn his failed venture and eight-year-old operating system into the alternative platform of the year 1998, but that's what Rhapsody is. I mean that in the Time magazine Man of the Year sense -- as a news comment, not as a judgment on its quality or ultimate significance.
I've been beta-testing Rhapsody and will honor my beta-tester agreement by not complaining about the performance of the beta release, but I will say I'm impressed by the robustness of the OS. And developers are certainly bringing out the apps. There were 77 confirmed commercial applications announced for Rhapsody as of October 21, 1997, according to MacOS Rumors (http://www.macosrumors.com/). UNIX users who grumble at how little attention good UNIX software gets will be pleased, I think, to know that the Apache Webserver is one of them. You can find it at ftp://ftp.apple.com/devworld/Rhapsody/unsupported_tools/apache-1.2.4-605.tgz.
New topic. Eventually, you even need alternative enemies. After declaring Bill Gates no longer Apple's nemesis, Steve Jobs went looking for a replacement. He understands that you can't rally the troops without a common enemy. In 1982, it was IBM. In 1998 it's Mike Dell. "We're coming after you, buddy." Jobs said, facing Texasward, on announcing Apple's Dell Computer-like plan to sell computers directly to customers at its online store.
Dell brought it on himself, when some reporter asked him what he would do if he were president of Apple and he said he'd shut it down and return the investors' money. This was a little cheeky considering that Dell was, at the time, running its highly lucrative online sales operation using Apple's WebObjects software. Hard to feel sorry for a guy who stands in Steve Jobs's sights and sticks out his tongue. Maybe he thought he had the alternative platform. The alternatives have the right, nay the obligation, to be cheeky.
But everyone knows that the premier self-defined alternative platform is the MacOS. It's "for the rest of us," a label that describes a shriveled but loyal constituency more recently defined as "the crazy ones."
But the real crazy ones have to be the Apple executives as portrayed in Jim Carlton's aforementioned Apple. Is this book a must-read? Is it a page-turner? It is a page-turner, and for followers of the Apple story, it's a must-read, but I do have some reservations, which I'll get to shortly. The story moves along briskly, the style is authoritative but very readable, and the events are, to any follower of this soap opera, fascinating. And there are stories here you've never read elsewhere. Like the never-before-published mid-1985 memo from Bill Gates to John Sculley, arguing that Apple should license the MacOS and detailing why and how, even offering to connect Apple with some likely partners. It was really Jeff Raikes (currently Group VP, Sales and Marketing, and a member of the Microsoft executive committee) who drafted the memo, spurred by a comment by product manager Chris Larson. But Gates recognized its wisdom, added significantly to it, called up some potential licensees to sound them out, and sent the memo to Sculley. You know what Sculley did. Or rather didn't do. But reading the story is intriguing. And the Raikes/Gates memo argued the case for licensing back in 1985 as well as anyone has since then.
Another story that isn't well known is the Startrek saga (briefly described by DDJ's Jonathan Erickson in his June 1997, "Editorial"). In that highly secret effort, Apple and Novell came that close to releasing an Intel box that ran the MacOS. And then killed it.
I don't know how many times Apple launched a project to port the MacOS to another platform, or how many times poor, patient, long-suffering Dan Eilers made presentations on the advisability of licensing the MacOS, but Carlton does. And he recounts every canceled project, every missed opportunity, until you just want to kick an Apple exec.
Carlton's book does a thorough job of chronicling the dumb mistakes that Apple executives made between 1985 and mid-1997. It's hard to understand how a company that could make so many bad decisions could inspire such fanatical loyalty from its employees and customers. It's impossible to understand it based on Carlton's account alone, and that's one weakness of the book.
Most of the bad decisions were missed opportunities. WebTV has been in the news a lot lately. Wall Street loves it. Microsoft bought it. It made Steve Perlman a millionaire. Back in 1990, Apple kicked it out the door.
Apple did a beautiful job of porting the spaghetti code of the MacOS from the 68000 to the PowerPC on a tight schedule with the company's future hanging in the balance. Unfortunately, nobody bothered to put together a reasonable PPC development environment or porting solution. Apple was about to introduce an exciting new line of machines that, because they could only run 68K applications in slow emulation mode, were slower than the current line. MetaTools stepped in on that occasion and saved Apple from disaster.
Pink, Taligent, Copland...the list goes on and on. Granted that it doesn't rank with the real stupidities of Apple's management, Gilbert Amelio's keynote at MacWorld Expo in San Francisco in January 1997, will stick in my mind forever. He had spectacular news to relate: the acquisition of NeXT Software and the return of Steve Jobs to the company. And somehow he managed to make it all incredibly soporific. He put people walking by on the street outside into a catatonic stupor. Cars were piling up on Howard Street because their drivers were dozing off as they drove past Moscone Center. So I'm glad to see someone finally say in print that it was "one of the most long-winded, disjointed presentations ever heard in the history of corporate America." Right on, Jim.
Carlton does a thorough job of exhaustively recounting all the stupid decisions ever made at Apple during the years he covers, but this hardly results in a balanced picture. Apple has done one or two things right, which is why it has enjoyed customer loyalty that every other company in the industry would kill for. Leaving out the successes makes Apple's recent troubles seem more inevitable, but it isn't balanced reporting. And there are other problems with the book.
Carlton is a Wall Street Journal reporter writing for the general public. He sometimes oversimplifies technical issues, and sometimes just gets them wrong. This irks me more than it would his target audience, I suppose, but it does irk me.
The problems include tired metaphors inconsistently applied...
...the microprocessor acted as the "brain" of a computer and the operating system was like a central nervous system....
...the microchip is the "brain" of the computer...the operating system is the rest of the body....
Then there's sloppy terminology, like referring to the whole range of content development areas in which Apple still maintains a stronghold as "desktop publishing."
And there are questionable assertions and basic misunderstandings. He simplistically characterized CISC as "an older technology" that "had a finite life," versus RISC with its "much longer life span," ignoring recent market realities.
He claims that, on its release in 1995, Windows 95 rarely crashed, while the Mac "was prone to crashing several times a day." He accuses Apple of deceptive advertising in listing prices for CPUs without monitor, and so on, but this is common practice. He presents Rhapsody as simply an alternative to NT, and dismisses its chances to displace NT. But Apple would claim that Rhapsody's object-oriented development tools and cross-platform deployment capabilities give it an edge with developers. You or I might disagree about the importance of these features, but Carlton seems to be completely unaware of them.
He speaks of Jobs having derided IBM's entry into the market with the "Welcome IBM. Seriously." advertisement, but there was nothing derisive about that ad.
And then there's his treatment of people.
There is a cartoonish exaggeration of the personalities of the Apple executives that doesn't do justice to the real human beings who suffered through the craziness he describes. And unfortunately, he often paints characters in such a way as to make heroes or villains of them. Mostly villains. "There was Loren, lording it over underlings..." "And there was Sullivan...eyes darting and nervously coughing in his office as he solidified a power base..." "There was Gassée, the strutting cock of the walk..."
Before reading the book I thought that Michael Spindler was a clever and capable man who should never have been put in charge of Apple Computer. I felt the same way after reading the book, but I knew a lot more about his bodily fluids. Carlton's Spindler is always sweating or spitting. Or sometimes spewing. "Known for a towering intellect but an inability to articulate well, he occasionally paused to spew his theories on global strategy at whatever poor sap happened to be breezing by." A lot of poor saps were just breezing by when they got sideswiped by Carlton's colorful characterizations.
Then there's the ending. As of the date his book went to press, Carlton thought that Apple's days were numbered. It would either be bought or would merge with another company, probably Sony, Philips, AT&T, or MCI. He mentions Oracle, but doesn't seem to consider an Apple-Oracle merger likely. And he thought it would happen very soon. If he believed that, he quit writing too soon. He would have had a stronger book if he'd waited for the fat lady to sing. Then he could have actually told the end of his story, rather than just hinting at it. But book release dates depend on a lot of forces, like the imminent publication of competing books and the limited duration of sabbaticals from the Wall Street Journal.
As a result, Carlton overreaches at the end. He tries to wrap up a story that isn't finished, and is forced to fall back on speculation. His short list of likely candidates for Apple CEO looks nothing like the list the board actually considered, and his guesses about possible merger or takeover candidates are mere speculation. All in all, it's a weak way to end a book that is otherwise very fact-based. And fact-based it is. Despite my concerns about Carlton's proclivity to find villains and to paint Apple's past decade as grand tragedy, with deep flaws leading inevitably to a catastrophic fall, the book is a fact-filled account of a remarkable era, and it manages simultaneously to entertain and enlighten.
But there are other books about Apple coming out soon, so maybe you want to check them out before you buy. After all, it's nice to have alternatives.
DDJ