Skip to Content
Skip to Table of Contents

← Previous Article Next Article →

ATPM 2.09
September 1996




Download ATPM 2.09

Choose a format:

The Personal Computing Paradigm

by Michael Tsai,

The Operating System Trickle

MacWorld Expos always seem to have themes. Two years ago, everyone wanted to see the first PowerMacs. For the first time in ten years, Apple had made completely different machines. That was back when people seemed more interested in the 680x0 emulation speed than in the new applications the PowerMac would make possible. Now, it is not uncommon to see applications that *require* a PowerMac, and few people talk about emulation speed anymore.

Last year, the unofficial theme of MacWorld Expo Boston was clones. I remember hearing people saying, "That doesn't look like a Mac!" when referring to the (now discontinued) Radius System 100. This year things were different. Everyone was accustomed to seeing PowerMacs. In fact, I'd be hard pressed to find a booth that *didn't* have one. The same goes for clones. Whatever the reason, it seemed as though most booths were giving demonstrations with clones. People are now so accustomed to seeing Power Computing machines that I don't think I heard one person say, "Is that a Mac?"

This year, the new kid in town was OpenDoc. OpenDoc and Cyberdog have been available to the public for quite a while in one form or another. But MacWorld Expo Boston was the first time that the public really got to see product demonstrations rather than technology proofs of concept. The majority of Mac users won't get these technologies until January, when Harmony (Mac OS 7.6) is released. If all goes as planned, (and I certainly hope it does) next year, MacWorld attendees will be as accustomed to OpenDoc as they were to PowerMacs and clones this year. Of course, people were still wowed by the *latest* clones, and I suspect that this will be true next year for the latest OpenDoc parts.

Perhaps the most encouraging aspect of the Expo was the amount of developer support for OpenDoc. Unlike Publish and Subscribe, Apple's first attempt at allowing different applications to work on parts of the same document, OpenDoc seems to have generated tremendous enthusiasm among both developers and users. Even better, it seems to have attracted the eye of small software companies. The Macintosh platform has always been about innovation, but in today's world, with virtual monopolies such as Microsoft and Adobe, it's hard for even an innovator to get their foot in the door. If someone wrote a specialized graphing tool that was the best in the world at what it did, it probably wouldn't stand a chance of competing with Excel or DeltaGraph because it wouldn't be able to compete as a complete graphing solution. This is going to change soon. OpenDoc will invigorate application development because products will no longer need to attempt to do everything. A spreadsheet container won't necessarily need to contain a graphing component because graphing parts will be available. Word processors will no longer have to include bare bones equation editors because specialized products will "plug in."

This is the promise of OpenDoc, and it stands a good chance of actually begin fulfilled. Digital Harbor demonstrated WAV, their new "work processor," at MacWorld. Because Digital Harbor used Apple's OpenDoc Framework (ODF), WAV was created in a very short period of time. In general, OpenDoc containers and parts should be much easier to write than today's applications. And Apple's OpenDoc Framework makes cross-platform versions easier. Digital Harbor doesn't worry about the fact that WAV will not compare well feature to feature with Microsoft Word, because WAV is an OpenDoc container. This allows it to act as a word processor, page layout, and web browser, all rolled into one. And when more OpenDoc parts become available, WAV's capabilities will expand without any rewrites or bugs tests on the part of Digital Harbor. If you want to find more information about WAV, check out the Digital Harbor web site at

I think that OpenDoc has the power to "save" Apple and the Macintosh way of computing. The reduced development times, will make developers happy. It will also allow applications to have a very clear purpose. To successfully compete in an OpenDoc world, applications (or parts) will need only to do one thing very well. It also promises the end of "bloatware". You see, OpenDoc will make it possible for new companies to form niches, and allow them to compete with the already established products on their own turf. Since (even though it's technically cross-platform) OpenDoc is primarily a Mac technology, it will be easier for developers to write Mac programs in conjunction with, or even before their Wintel counterparts. Component Integration Labs will validate all OpenDoc parts to ensure that they will work well together.

Unfortunately, OpenDoc is a critical mass technology. Once enough people start using it, it could really take off and change the entire computing paradigm as we know it. Until then, you probably won't hear much about it. The main question, is if major companies will introduce OpenDoc versions of their products. ClarisWorks 4 is already a good example of integration, and the soon to be released version 5.0 will make it an OpenDoc container. Many other companies have also announced OpenDoc plans. However, neither Adobe nor Quark have announced plans for OpenDoc support in their page layout programs, and at least in the beginning, it will be difficult for OpenDoc to go head to head with their programs.

The Old and New Deals

I remember years ago when Apple announced that it was working on two next generation operating systems - Copland and Gershwin - that would take Macintosh technologies to the next level. The Copland release was supposed to coincide with the delivery of the first PowerMacs, creating an entirely new machine, with a brand new operating system. Application developers thought that they would have to rewrite their software once, adding both PowerPC native code, and optimizing it to run in Copland's new memory and tasking environments. Obviously, Copland isn't here yet, and instead of yet another release date, Apple announced at MacWorld that it was no longer going to ship a Copland in a single release, but would instead ship it piece by piece as individual parts were completed. The press had a field day with this announcements, and the Usenet newsgroups were filled with messages titles "Copland is Canceled."

The reality of the situation is that nothing has really changed. The part of Mac OS 8 that we've come to refer to as the guts of Copland (the preemptive multitasking, and the protected memory), are also the pieces that surveys show people most interested in. This part will still not reach customers hands until at least January 1998.

In reality, Apple abandoned the monolithic release strategy years ago. As Copland's release date slipped farther and farther into the future, more and more features were stacked on top of System 7.x. Most of this started with the release of System 7 Pro. And finally, with System 7.5 new components such as QuickDraw GX, PowerTalk, and AppleScript became available to everyone. For the past several years, all new components that Apple develops have been available on internet sites for public use downloading. The new versions of QuickDraw 3D, QuickTime, and PlainTalk can be downloaded as soon as they are completed without waiting for the next complete OS overhaul.

This is exactly the new operating system release strategy that Apple has announced as its "New Plan." So at least in my mind, there is still a question about what has changed. New components such as Cyberdog and the Copland Appearance Manager will still become available to the public when they are finalized, and delivery of the interdependent guts of Mac OS 8 is still far in the future. Aside from a new official attitude at Apple, the only difference to consumers is that they will be able to buy incremental updates in addition to downloading them from the internet. This is of course, unless Apple decides to charge net savvy users for the updates too.

Leaving new software components on web servers seems to have worked in the past, and I suspect that this is generally because the people who could find them were experienced enough Mac users to know their purpose and how to install them. These people will continue to get OS updates online. Other users may not enjoy seeing continual advertisements for the latest version of the operating system, and won't be pleased to find out how minor some of the revisions are. They won't like being nickel and dimed for bug fixes or integration of preexisting components. To save people the trouble of ordering upgrade CD's when they become available each quarter, perhaps Apple should institute an update subscription plan. A customer pays a flat rate up front, and automatically has one year's worth of software updates sent to him as they become available.

Hopefully, new updates will contain more complete documentation than the current ones do. I have yet to see a complete manual for OpenDoc or Cyberdog, and betas of the AppleGuides are only now being circulated. If continual updates are truly meant for the mainstream, this is one of the areas that will need to be improved.

When Will Mac OS 8 Actually Arrive?

Apple hasn't said when the version number will change to 8, but it makes sense to do this with the release of the new microkernel and application environment. So far, Apple has been very vague about how the incremental updates will be numbered. But if it announces system updates every quarter, how is the public supposed to know which one is the "big one?" And will all these updates be accompanied by complete OS installers, or will we have to install System 7.5.3 and then apply several updates to get the latest version? The updates must be easy to install, and it must be clear what they do. Apple chose to withhold the release of "Son of Buster," System 7.5.4, for about a week because they noticed a bug in the golden master. And sure enough, an improved version (7.5.5) was released today. This is a very good sign. It shows that even though Apple intends to ship updates more often, it will continue to wait until they are ready. No matter when updates are delivered, quality is the first concern.

Apple thought that rather than announcing another Copland delay, it could announce a strategy of continual improvement. In reality, this will make things even more complicated for Apple, and its customers, and the added resources needed to test and launch updates on a regular schedule are likely to postpone the final product even more.

With Copland still off in the distance, I think that OpenDoc's success or failure will, in a large part, dictate the overall success of the Macintosh platform. The quality, quantity, and timeliness of OpenDoc parts will determine whether users once again wowed by a new paradigm and user experience from Apple, or whether OpenDoc is seen as yet another cool technology that died.

Next month, I will discuss an alternate path towards a modern OS, a mirror universe if you will, in which Mac users gain power and elegance without sacrificing compatibility. At least on paper (or on the screen), there may Be a better solution than waiting till '98 for Mac OS 8. For all those Mac users out there, let's hope so.

"The Personal Computing Paradigm" is ©1996 by Michael Tsai, [apple graphic]

Also in This Series

Reader Comments (0)

Add A Comment

 E-mail me new comments on this article