Buy Books

Virtual Tours
Big Projects
Big Disasters

os war 2.0: part I
June 7, 2001

Mac OS X

Microsoft and apple are both scrambling to finish modern operating system projects for the masses. Their efforts, apple's mac os x and microsoft's windows xp, are about to go head to head in a new os war: version 2.0

part one: a decade of preparation

Much has been written about the so-called os war between apple computer and microsoft. And despite steve jobs' conceding that the os war was over, neither apple nor microsoft would be better off without one. After all, apple provided microsoft with much of their user interface and multimedia ideas, and microsoft has written much of the software that keeps apple's os relevant.

Competition creates change and demands improvement from both sides. And as software progress increasingly falls behind new hardware advances, a serious os war is just what microsoft and apple need to jumpstart new development. The treasure island experiment will be publishing a series of articles comparing the latest operating system offerings from apple and microsoft. As the first article in the series, part one will examine the history of apple and microsoft through the last decade, and how both have arrived at the very same point this year: that of delivering their first modern os for the mass consumer market.

Apple has just released the first consumer version of mac os x, an entirely new os for the company, based on next computer's nextstep product; microsoft is finishing windows xp, the final push to move dos users onto their nt based operating system product. Successive articles will examine how the two projects compare in terms of their legacy, design philosophy and architecture; network services and application support; hardware support; user interface and usability; application availability, portability and api support; and real world performance, capability and reliability.

the os war 0.9: beta developments

Up into the early 1990's, apple's system 7 was so far ahead in many respects of any other consumer desktop operating system that apple could charge a significant margin more for their hardware and software package. For the next five years, apple would demonstrate various plans toward an improved operating system, but microsoft would actually deliver new os actual products. As microsoft's progress began to erode away at apple's advantages, the differences became less apparent. Additionally, microsoft's aggressive alliances and pervasive marketing often eclipsed apple's efforts.

Microsoft and ibm had terminated their plans to co-deliver a new super-dos called os/2. Microsoft left ibm with the project and started work on their own post-dos in 1988 that they dubbed windows nt. Designed to replace their existing windows 3.0 product, which was little more than a graphic dos shell, this new operating system was planned to bring a level of power to the consumer desktop that only expensive unix workstations had achieved previously. Released as windows nt 3.1 in 1993, the new product was well received, but had very limited success on the desktop because of performance reasons. Without the ability to replace dos with their new nt operating system on reasonable hardware of the time, microsoft faced competition with unix on the high end, and other dos vendors, including digital research and ibm, on the low end. And in the middle, apple's very functional macintosh made the current windows 3 look very lame.

As a stopgap measure, microsoft bound dos together tightly with the old windows 3.0 product, and improved upon both. They copied the look of apple's macintosh and released the product in late 1995 as windows 95. The new product sank other dos vendors by making their products simply irrelevant, and it targeted sales, not in competition with high end unix workstations, but apple's successful, easy to use macintosh.

Well before the release of windows 95, apple had been planning an improved system 7. But apple faced the same problems in releasing a significantly new os as microsoft: their current customers were tied to their current product, and moving them to a largely new os would prove to be a herculean task.

The macintosh's operating system had been designed in the early 80's to to run an appliance computer, with no thought of support for multiple users or even multiple concurrent applications. Apple had retrofit the mac's os to support application switching, then system 6's multifinder mode, and finally, by system 7, multiple cooperative overlapping applications. Now they were faced with retrofitting in the ability to allow applications to run in their own memory space, and under system directed scheduling rather than running with full control of the system. Without that work being done, the macintosh operating system would never be more reliable the its weakest application.

Microsoft wasn't waiting on apple however. Shortly after the release of windows 95, microsoft released windows nt 4, which copied the interface of 95, and introduced system compromises that boosted performance, even though they reduced the system's stability and reliability. Microsoft, unable to figure out how to wean dos users onto nt, decided not to at all, and instead embarked on a dual-os strategy that allowed for a dos based windows to continue for the low end pc market, while the company's nt product provided a more reliable product for business users. Microsoft called both products 'windows,' and thus created a strong brand that rivaled apple's once untouchable macintosh. On the low end, windows 95 allowed dos pc users a work environment similar to the macintosh; on the higher end, windows nt provided both a workstation and server product that served as cheap replacements for unix workstations and low end servers.

os war 1.0: game over

Apple's copland project promised a revitalized mac os well before 95, but slipped each year until 96, when apple, nearly dead from years of bad management and chaotic planning, woke up and strangled the out of control project. Copland was nearly ripe for beta delivery, with a developer package ready to go, but apple's new management realized that copland was largely a fanciful network of pet projects with little tangible benefits for customers and a very unclear business plan.

There simply wasn't compelling reason for developers to spend the required time to rewrite their apps to take advantage of copland. First, copland offered few significant new benefits to users, even if new software was written for it. Second, the dwindling numbers of mac users offered little incentive to developers to create any new software. So copland was killed and apple went shopping for a new os outside of the company.

In the meantime, apple worked toward improving the mac's marketshare and repositioning development efforts to improve system 7. Apple signed agreements with vendors to sell copies of mac hardware, and rechristened system 7 as mac os 7 in an effort to invent a difference between apple and the mac market. Apple's small improvements to system 7 and cloning of the mac hardware did create some new excitement around the macintosh.

Unfortunately, excitement wasn't enough to sustain profitability, and apple began to hemorrhage massive amounts of cash. By the end of 1996, windows 95 was looking very successful, and apple, and its products, were looking weak. Apple had reduced itself to a research and development company for various hardware vendors, and was selling them, for cheap, an aging os with a uncertain future.

os war 2.0: what's next

Apple's acquisition of next computer changed things dramatically. First, apple obtained a high powered os that delivered exactly the things they lacked: a powerful, reliable foundation, a bright potential for future application development, new customers and new developers. Along with that, lots of associated talent and technology, including a former founder who successfully took control of the old apple and replaced much of the company with his own.

The new apple aggressively took back the macintosh hardware market by ending clone contracts and releasing their own compelling new products. At the same time, the new apple made the most of the current mac os, adding salvaged bits of copland in a release called mac os 8.

One slowdown on the road to apple's recovery was the hesitation of developers to make the dramatic changes needed to move to a materially new os. Apple's ambitious plans to release next's os as the new mac os were sent back to the drawing board each year, in a process that started looking like the copland fiasco. The difference was that in copland, apple wasted lots of time making elaborate plans, reinventing wheels and delivering nothing. This time around, the new apple made continual progress. Each year real software was released to developers that, even from the start, demonstrated real potential. As plans changed, the product continued to improve.

Microsoft wasn't waiting on apple this time around, either. In the interim, the dos based windows 95 was re-released as windows 98. At the same time, windows nt 4 was improved upon, and eventually released as windows 2000. Microsoft eagerly wanted to move their 95/98 customers to the 2000 product, but it still couldn't support the games and range of hardware available to dos.

Pressed to release a successor, microsoft improved upon 98 twice more, once as 98 special edition and again as millennium edition. Later this year, microsoft plans to release the first nt derived product for general consumption: windows xp.

With drastically different starting points, apple and microsoft are once again in a position to compete head to head. Not since system 7 and windows 95 has it been simple to compare the two companies' products. Now, both are offering new versions of modern operating systems with significant, yet very different, backgrounds. Both are similar in appearance to the products they replace, but very different underneath. Both are first steps into the next decade.

For microsoft, windows xp will be a first effort to get mass market customers to start using an nt derived system, while still providing support for legacy applications; for apple, the effort will be to move mac users to an entirely new operating system as comfortably as possible, while enabling as much new potential as possible.

How will apple's mac os x stack up against microsoft's windows xp, and how does it fair against the current windows 2000? The treasure island experiment's next article will examine how they compare in terms of legacy, design philosophy and architecture.

More on apple, microsoft, comparison pricing and history:

cover cover cover cover cover cover cover cover

Microsoft Windows 2000 Professional
Microsoft Windows 2000 Server (5-Client)
Microsoft Windows Millennium Edition
Mac OS 9.1
Mac OS X
History of Computing: An Encyclopedia
Fire in the Valley
Hackers: Heroes of the Computer Revolution

2001 tiexp.com The Treasure Island Experiment. All rights reserved.