According to Hoyle...
Top 10 All-Time
by Jonathan Hoyle
This month, we take a trip down memory lane, across the history of
the Macintosh industry, to point out the Top 10 Fiascos from the perspective of
a Macintosh developer. Although
the history of the Macintosh is littered with blunders and boneheaded mistakes,
we are going to look at outright fiascos. And as this is a software development column, I have chosen those which
are from the perspective of a Mac developer.
So what counts as a fiasco? Well, one thing I do NOT count as a fiasco is a single failed product
within a successful line. Common
examples of such non-fiascos include the Apple III, Lisa, or even the Mac G4 Cube. Sure, each of these
products gave Apple some bad press, but ultimately they were just blips on the
radar scope. Let's look at IBM as an example. Almost any book
chronicling the historical embarrassments of IBM will point out the IBM PC
jr. But in the end, the PC
industry as a whole was not hurt by it. PC users stayed PC users, they bought different products, and IBM
endured a couple of less profitable quarters. No big deal. A
true IBM fiasco would be OS/2. OS/2 was not simply a failed offering to a community that could simply
choose another product. IBM's
failure here destroyed an entire platform, with an installed base of users
preferring to stay with OS/2, but cannot due to IBM's bungling.
For me, a Mac fiasco is one which has the potential of crippling an
entire user base (even if that user base is a minor one). Alternatively, a fiasco could be a
corporate decision to do something that even their most loyal customers disavow
and mock. They are the "New
Taste Coke" of the industry, openly ridiculed by both pro-Mac and anti-Mac
Some of the fiascos I mention will make you laugh (like #5), some
will make you angry (like #4). They include those from Apple, 3rd party collaborators, and even its
competitors. #5 is actually a
Windows fiasco that was indirectly caused by Apple (and thus a positive fiasco
from the Mac standpoint). Some
fiascos actually create new demand (such as #6). Many are specific to software developers (#7), while others
are fiascos to the general Mac user (such as #3). In the end, I wanted to pick the most eventful ones in the
eyes of Macintosh programmers.
The Missing 10th?
Although this column's title suggests 10 fiascos, I list only 9 of
them here. I will let you the
reader to decide upon a 10th. Email me at email@example.com a description of the fiasco you feel
belongs in the Top 10, and I will devote an entire column on the reader's
pick. For now, here are my 9 of
the Top 10:
9. Copland (1996)
By the spring of 1996, Apple was running a bit nervous. Microsoft had just introduced Windows
95 the previous August, and for
the first time Windows now had the technological upper hand over Mac OS. Before this point, the Mac operating
system had always been superior to its rivals. However, Microsoft's new Windows 95 was the first consumer operating system which
was preemptive multitasking and memory protected, while the Mac's System 7.5 OS
still lacked these features. For
two years, Apple was struggling to release their next generation OS, codenamed Copland, to get back on top in the operating system
game. At Apple's Worldwide
Developer Conference in May 1996, this new OS, now given the official name Mac
OS 8, was the focus. At this conference, Apple said that
Copland was very nearly ready to go beta, and a pre-release had narrowly missed
being released to WWDC attendees. Just a few months later, this was all revealed to be a lie.
Copland wasn't anywhere near beta-worthy. It was a bloated, bug-infested mess. It was so bad, Apple would eventually
drop the entire venture to start from scratch. Why Apple would knowingly mislead their developers about the
state of Copland is hard to say. Certainly Apple engineers knew how bad things were. Could it be that Apple marketing was
just blinded to the realities? Perhaps, but whatever the reason was, it caused Macintosh developers to
lose faith for the first time. Apple
was not only behind Windows in technology now, but this would be the case for the foreseeable future. Over the next year, a number of Apple
employees lost their jobs, including CEO Gil Amelio.
Epilog: Mac developers felt betrayed by the false promises
made at WWDC '96. 1996 would
represent the last of the "fun" WWDC conferences, as future
conferences would be forced to focus on content (rather than as a pep rally). The Copland fiasco left Apple's future in serious
question. Scrambling to recover,
Apple began searching for ready-made solutions for a next-Gen operating system,
with BeOS looking like the
likely candidate. In the end, it
was Steve Jobs and NeXT that Apple turned to for a solution.
Visit here for more information on Copland.
8. The Microsoft "Marriage" (1997)
At the 1997 MacWorld Expo in Boston, Steve Jobs shocked the Mac
faithful by announcing a deal between Apple and Microsoft. In this deal, various patents would be
cross-licensed, Microsoft will invest $150 million in Apple, and in return Office
97 will not be canceled and
Apple will make Internet Explorer its default web browser. At one
point, Big Brother Gates loomed large in a screen over Steve Jobs, leaving an
ominous impression as to what this relationship meant. Jobs described the deal as a
"marriage". In reality,
Steve Jobs was simply Bill Gates' bitch.
What was desperately marketed as a "deal" was nothing more
than a poorly veiled blackmail threat made against Apple. Microsoft was in a browser war with Netscape, and wanted Internet Explorer
to be the dominant product. Before this time, Apple remained
browser neutral, allowing people to choose for themselves between Netscape and Explorer. In
1997, Apple was still on the ropes financially with sales down and people still
recovering from Fiasco #9. If
Microsoft killed Mac Office at this stage, Apple might not have ever had a
chance to recover. Gates, knowing
this, took this opportunity to extort Apple to do its bidding. One can view a scan of a Microsoft memo
outlining the corporate blackmail attempt from court documents in a later
In particular, the smoking gun found in the memo:
"The threat to cancel Mac Office 97 is certainly the
strongest bargaining point we have, as doing so will do a great deal of harm to
Apple immediately. I also believe
Apple is taking this threat very seriously."
In the end, Steve Jobs caved. (He really didn't have much of a choice.) Apple was forced to make Internet Explorer the default browser on the Mac, and in return
Microsoft will deliver Office 98 for the Mac and invest $150 Million in Apple. As for the patent sharing, you can
guess who got the upper hand here.
years later after Apple had been truly revitalized, Steve Jobs no longer needed
Microsoft to remain successful and the "marriage" was annulled. Apple's default web browser became its
own Safari, and Microsoft
continued support for Office for its own financial reasons, not Apple's.
For on the blackmail, read this MacWorld article.
7. Rhapsody (1997)
On the heels of Fiasco #9, Apple was in desperate need of a modern
OS to compete with Microsoft's Windows 95. Apple eventually
decided to purchase Steve Jobs' NeXT
and use its operating system, NeXTStep as the path for the future OS. At
the 1997 Apple Worldwide Developer Conference, Apple outlined what the future
looked like in its port of NeXTStep to the Mac platform, called Rhapsody.
However, applications which wished to run natively on Rhapsody had to be completely rewritten from scratch in
Objective-C, in a completely new API called YellowBox. YellowBox was completely incompatible with the existing Mac
ToolBox API's, and there was no
transitionary mechanism for developers upon which to rely. Existing software could continue to run
in what was called BlueBox, a
primitive version of Classic. BlueBox lived inside its own window and had its own desktop and Finder,
much like the old Virtual PC or today's SheepShaver emulator.
Developers were pissed, and WWDC '97 went down as one of the most depressing
conferences Apple has ever held. As the development community had no intention of rewriting everything
from scratch, Apple found that the Rhapsody initiative was completely
Epilog: The following year, Apple came back with a winning
strategy at WWDC '98: merging NeXTStep and Mac OS into a
hybrid operating system called Mac OS X. In addition to the
Objective-C based NeXTStep APIs (now renamed Cocoa), Apple also provided the Carbon API,
a cleaned up version of the old Mac ToolBox, which allowed users to transition their
applications without having to do a complete rewrite. With Carbon in place for current projects, Cocoa remained
available for new development, and this dual strategy became enormously
6. The Round "Hockey Puck" Mouse
Anyone who was a Mac owner knows about this little fiasco. Yeah, the friggin' round mouse. Sure, we can laugh about it now, but
back then it was as annoying as fingernails down a blackboard. When Steve Jobs introduced it with the
iMac, he said, "I think this is the best mouse we ever made." It's hard to imagine the future genius
behind the iPod, iPhone and the Mac's revival could have slippped up like this so
badly. But he sure as hell did.
For those who missed out on the fun ten years ago, this mouse was
perfectly round, not oval-shaped like most. What's wrong with that, you ask? Well try using it. You find that you couldn't orient it correctly (since the shape is
perfectly symmetrical), and your pointer will go off in a different direction
that you expected. Everyone, I
mean everyone, hated it. But stubborn Steve wouldn't let
go. He perhaps thought that with
time people would come to his way of thinking.
This fiasco actually caused a boon in 3rd party mouse products. Most people just bought another mouse
when they bought a Mac, and threw the Apple mouse in the closet. Vendors started making money selling
mouse accessories to fill in the gap Apple created. Some vendors even created enclosures to give this mouse a
real shape. In any case, even
Steve Jobs stubbornly refused to accept what was obvious to all alike: this
round mouse was a fiasco.
Epilog: In 2000, Steve Jobs replaced the round mouse with
a normally shaped optical mouse. This new mouse was happily embraced by Mac users. Although he never admitted failure on
this product, Steve doesn't really bring the topic up either.
For more info on
the round mouse, read this article.
5. The Gates / Seinfeld Ad Campaign (2008)
After two straight years of getting beaten up by the "I'm a Mac
/ I'm a PC" ads, Microsoft was ready to strike back. The Apple ads remained overwhelmingly
popular, even by Windows users. Microsoft needed combat this marketing threat with an ad campaign of its
own. They decided to pay off long
time Mac aficionado Jerry Seinfeld $10 million to ditch his Powerbook and go on
television with Bill Gates to promote Windows.
What came out was a rather bizarre and very unusual television ad.
Unless you've been living in a closet for the past few months, you
all already know how bad they were. The press and bloggers were relentless. Ridicule and derision was heaped upon Microsoft, far worse
than the Apple ads had. Two weeks
and $300 million later, these dismaying ads were finally put out of its (and
everyone else's) misery. In their
place, a somewhat risky approach was taken: turn the "I'm a PC"
around and fire back at Apple. Make it a positive again by showing people who use the PC, and have them
say "I'm a PC". It was
actually quite a clever move for Microsoft. And the "I'm a PC" ads certainly performed better,
as Microsoft started to finally feel good about their expensive campaign.
That is until...
...until it was leaked that these new ads were made ... (you guess
it) ... using a Macintosh.
I have friends who had tears coming down their eyes laughing so hard
when this news broke. With egg
still showing on their face, Microsoft attempted to dismiss this in a rushed
press release, stating: "productions houses use a wide variety of software
and hardware...including both Macs and PCs". Hey, now Microsoft's press releases are advertising
Macs! Woo-hoo! The embarrassments continued further
when it was discovered that some of the paid spokespeople displayed in these
ads were actually Apple users.
Epilog: The ads continue to play, so it's too early to say how the "I'm
a PC" campaign will be viewed in the future, but you can read further
about it in its Wikipedia entry.
4. MacBasic (1985)
When Apple engineers were designing the Macintosh, they wanted an implementation of
the Basic programming language which was powerful yet easy to use. They wanted to truly empower the Mac
user, not just give him the same old console-like Basic, such as Microsoft's
implementation on the Apple II. The company's guru of Basic programming at the time, Donn Denman, poured
his heart and soul into the project, creating MacBasic, an advanced, object oriented version of the
language, one which could even control many aspects of the Macintosh GUI. It was a revolutionary feat of
engineering and received rave reviews from all who saw the demos. By the spring of 1985, beta testers
were using it heavily, and it was highly anticipated by all for a full
commercial release that summer. Unfortunately for Apple, Microsoft CEO Bill Gates happened to witness
one of these demos.
By this point in time, Microsoft had already ported their mediocre console-like
Basic interpreter to the Macintosh. As a product, MS BASIC couldn't hold a candle to MacBasic,
and Gates knew it. Gates was
determined to stop MacBasic from shipping as he had no desire for the competition. As it happened, the Apple II's Basic
(built into the computer's ROM) was written by Microsoft and its license
agreement with Apple was up for renewal that coming September. So Gates blackmailed Apple: If Apple
doesn't kill the MacBasic project, Microsoft would refuse to renew its MS BASIC license for the Apple II.
In 1985, the Macintosh was still a very young product, and the vast majority of
Apple's revenue was still coming from Apple II sales. Although everyone knew that the Mac would eventually
overtake the Apple II in sales, the company in 1985 still relied very heavily
on the health of its older Apple II line. Apple President John Sculley, knowing that he could ill afford to
jeopardize the Apple II community, capitulated to the demands. In exchange for Microsoft renewing the
Apple II Basic lease, Sculley was forced to sell MacBasic
to Microsoft for...(get this)...$1.
That day, Denman was told by Apple management that the MacBasic
project was being terminated immediately, and
that he had to destroy all existing copies, including the source code and
documentation. Apple refused to
give Donn an explanation, despite his desperate pleas to know why. The stunned Denman watched as several
years of his painstaking work got deleted. Heart-broken, he left the building on an emotional and
ill-advised motorcycle ride. On
his way home, he was involved in a terrible accident which totaled his bike
(fortunately, Denman survived with only minor injuries).
Epilog: Apple attempted to retrieve all the remaining
copies of MacBasic it had
distributed to beta testers, but when word got out what was happening, the beta
testers refused to turn in their disks. The MacBasic beta
became widely pirated and circulated amongst developers and remained in
underground use for some time. Surprisingly (and attesting to its continued popularity), two planned
books on MacBasic continued
into publication, and, despite all else, sold well.
As for Microsoft, the unimpeded MS BASIC became the dominant seller of Basic on the
Macintosh for the next several years. With MacBasic now the
legal property of Microsoft, many of its ideas and unique features were
cannibalized and reused for the creation of Visual Basic years later. But that is a story for another time.
For a more on the story of MacBasic, check out this article.
3. The Death of Mac Clones (1998)
In 1995, Apple took a close hard look at their rich rivals over in
Redmond and wonder how it was that they could be making so much money, whilst
Apple was constantly broke. They
quickly concluded that Microsoft's advantage was based upon the fact they had
no expensive low-margin hardware to deal with, only software licenses. While Apple, IBM and others had to
manage expensive hardware, Microsoft had only disks and manuals to worry about. So then CEO Michael Spindler came up
with a plan to follow the same path: license its OS, then System 7, to hardware
vendors. After defining a
standardized hardware platform called CHRP (Common Hardware Reference Platform),
Apple began attracting hardware vendors to join the Mac platform.
Mac users were extremely excited by the prospect of new hardware
vendors joining the Mac community. Most believed that the reason for PC dominance had less to due with
Windows' strength and more to do with the openness of hardware vendors to jump
on board. In 1996, the first Mac
clones hit the marketplace, all with rave reviews. Such names as Motorola, Power Computing, UMax, and even IBM
themselves, all jumped into the market to sell Mac clones. Mac OS marketshare began to spike,
reaching an astounding 13%. Many
of these companies, however, were just breaking even still paying for the
startup cost, hoping to build a customer base so that they could realize
profits the following years.
Although overall Mac OS marketshare was rising high, Apple's own
share of the pie was by necessity shrinking. After the Copland disaster was revealed (see Fiasco #9),
Steve Jobs was brought back in, and shortly thereafter bumped out CEO Gil
Amelio to take command. Sadly, one
of his first acts was to kill off the clone market.
To no one's surprise, Steve Jobs would rather be a big fish in a
small pond than a small fish in a big pond. He didn't want to grow the Mac market unless that meant
growing Apple as well. So very
quickly, he cancelled all future internal development in CHRP, and beginning
with Mac OS 8.5 and later, clone support would be dropped in the OS.
Mac clone hardware vendors felt betrayed by Apple. After all of the pushing to get them to
join the platform, Apple pulls the rug out from under them. Mac users themselves were also upset,
seeing the loss of hardware vendors as a serious downturn to the platform.
Epilog: As clone manufacturers discontinued their
products, Mac OS marketshare plummeted. Though Apple's bottom line improved (as it once again owned 100% of the
Mac pie, the pie continued to shrink). Free falling below 10%, then 8%, then 5%, eventually the Mac OS
marketshare hit an all-time low of 3%. Eventually though, Jobs found a way to bring Mac marketshare back up
(without licensing Mac OS). Despite Apple's current good fortune and record sales, Mac OS
marketshare in 2008 still pales to that of the heyday of the clones.
For an interesting look back at Mac clones, you might enjoy watching this half hour program.
2. Mobile Me (2008)
The most recent of all fiascos is one we are dealing with now: the
living purgatory of MobileMe. MobileMe is Apple's unfortunate replacement to its
popular .Mac (pronounced dot
mac) service. .Mac gave you email, hosting area for posting web
pages and a number of other goodies. MobileMe does
essentially the same thing, except more sluggishly and more poorly
implemented. The intention behind
moving to MobileMe is to
better support iPhone users. Sadly, Mac users were forgotten in this move.
Prior to WWDC '08, Apple's .Mac program had been fairly successful.
Huge numbers of Macintosh users embraced their mac.com address (including your's truly)
and take great
advantage of its iDisk and
the web features available. Although most of these features remain in place under the new name, it
has become quite painful to use.
Users attempting to access email over the web interface are stuck
with painfully long delays, tragically common outages, and unwelcome bugs, such
as lost and undelivered mail. Hundreds of formerly devoted .Mac are fleeing to
Google's Gmail or other alternatives, as they feel cheated by
Apple. Even Steve Jobs himself
realizes that this has become an unbearable situation and would not have made
this move had he known how bad things were,
and even today it remains in very poor shape.
Apple has extended users' terms of service to stave off this potential mass
Epilog: >It's too early to tell yet how this fiasco will
end. Most users, including myself,
just simply wish they could have their old .Mac accounts back.
1. Metrowerks CodeWarrior (2005)
#1 for Mac developers, with no close second, was the riches-to-rags story of
Metrowerks CodeWarrior, which overnight went from universal use to product
cancellation. Any Mac software
engineer who has been in the field for more than a year or two still feels the
pain of this fiasco.
Back in the days
of 68K Macs, Metrowerks was a small education market compiler developer, known
principally for its Pascal and Modula-2 products on the Macintosh and MIPs.
With the advent of the Power Macintosh, all of that changed. Metrowerks
introduced CodeWarrior at the
end of 1993, the only user-accessible development environment that could create
native PowerPC applications. CodeWarrior offered three front end languages:
Pascal, C and C++, and had two back-ends that could be compiled for: 68K and
PowerPC. Later versions of CodeWarrior would offer additional language and
processor support, such as Java and Win32 development. Metrowerks also
introduced the PowerPlant C++ class framework. In less than two years,
Metrowerks went from niche to market dominance, completely changing the face of
Macintosh software development. For the next decade, no other tools developer (including Apple) could
adding support for a number of new platforms, went public and was eventually
purchased by Motorola for its innovative compiler technologies. CodeWarrior became the overwhelming
marketshare leader in Macintosh C/C++ development. Over 90% of shipping Mac applications during this time were
built using CodeWarrior. Considered both easier and more powerful than the freeware Xcode from
Apple, CodeWarrior was the IDE of choice for the Mac community. In addition to the PowerPC compiler
inside CodeWarrior, Metrowerks also had an x86 compiler to build Windows
applications. Its x86 compiler had
been tested and grown for over 10 years, just as its Mac compiler had.
However, in 2003,
the face of Metrowerks began to change. Motorola was in its initial stages of
spinning off portions of Metrowerks to Freescale, a company which had no respect or understanding
of Metrowerks. Freescale's
mismanagement and blunderings were seen immediately. In 2003, for the first
time in 10 years, Metrowerks failed to even make an appearance at Apple's Worldwide
Developer Conference, the single most important conference for Macintosh
software development. Things went
from bad to worse in 2004. Not
only was Metrowerks absent again from WWDC, but for the first time in the
company's history, it failed to deliver its annual Macintosh release of
CodeWarrior. With corporate apathy ruling the day at Freescale, developers
began to rightly suspect that doom was on the horizon.
However, it would
be 2005 that would put the final nail into CodeWarrior's coffin. In an amazing
combination of stupidity and bad timing, Metrowerks sold off its Intel compiler
technology just weeks prior to Steve Jobs' announcement that the Macintosh
would transition its processors from PowerPC and to Intel. Bad decisions have killed businesses in
the past, but it is a rare case when a single bad decision turns a business
from monopoly to cancellation virtually overnight. The trickle became a flood as Macintosh developers abandoned
CodeWarrior to switch to Xcode.
By effectively shooting itself in the head, Freescale drove an overwhelming
popular development environment, a near monopoly in fact, into insolvency in
just two years. Metrowerks was
left with no choice but to abandon the Macintosh market, which is what it
announced at the 2005 MacHack conference. A final version 10 of CodeWarrior for the Macintosh was released in late
autumn as a download only and at a slashed price of $99. In the spring of 2006, all support and
sales of CodeWarrior for the Macintosh was officially terminated.
This sad end to a great product speaks volumes of how an incompetent CEO can destroy
a truly great product.
Epilog: For over a decade, CodeWarrior was the Mac developer's best
friend. Far superior to
Microsoft's offerings for Windows, Metrowerks' product line represented the
cream of the crop in the industry...at least for a time. Toward the end of its life, it began to
atrophe, and with its exit, Xcode is now the primary development environment for Mac OS X.
If you have any feedback, or would like to offer your own suggestion, please email
me mailto: firstname.lastname@example.org with your thoughts, and I will devote a column to your responses.
Coming Up Next Month: SheepShaver Update for 2009! See you in 30!
To see a list of all
the According to Hoyle columns, visit: http://www.jonhoyle.com/maccompanion