have these boneheads even even *SEEN* NeXTStep?
last paragraph sums up my point of view exactly though.
~~~
Apple's OS Double Feature
Apple came to a fork in the Mac OS road -- and decided to take both paths.
By Henry Bortman and Jeff Pittelkau
Copland is dead. Long live Rhapsody.
Several months after scrubbing its plans for Mac OS 8, code-named Copland,
Apple last January unveiled its new
strategy for delivering a modern OS, code-named Rhapsody, to Macintosh
users. In a surprise move, the company
bought Steve Jobs' NeXT Software and announced that it would base its new
operating system on NeXT's OpenStep
technology. OpenStep is a set of object-oriented APIs (application
programming interfaces) that enable developers to
build robust applications quickly. It currently runs on several modern
operating systems, including Sun's Solaris and
Windows NT.
Although Apple had other choices for delivering an OS that met all the key
modern-OS criteria, OpenStep offered what
Apple considered to be the most compelling combination of benefits. Like
Apple's other alternatives, OpenStep can take
advantage of underlying modern-OS features such as preemptive multitasking,
multithreading, symmetric
multiprocessing, and protected memory, which improve performance and
stability. (For an explanation of these and other
modern-OS features, see "Plan Be," January '97, page 64.)
In addition, OpenStep's being object-oriented simplifies the task of
building and revising applications and makes the OS
well suited for distributed systems that run across networks. NeXT also
provides developers with an innovative
development tool, InterfaceBuilder, a kind of construction kit for building
application interfaces by dragging and
dropping buttons, menu items, and other controls. And perhaps most
important, it has been battle-tested. The NeXTstep
OS, from which OpenStep is derived, has been around for nearly a decade and
is considered reliable enough to be used in
many mission-critical Fortune 100 custom applications. Apple is undoubtedly
hoping that this legacy will give the Mac a
shot at establishing a beachhead in Corporate America outside the
publishing department.
Apple's OS Two-Step
You might think that, given its inherent strengths, Apple would simply port
NeXTstep to the PowerPC and be done with
it. However, such a strategy would leave millions of Mac users in the
lurch, without benefit of any backward
compatibility with current Mac OS apps.
That's where Blue Box comes in. Apple's new Rhapsody OS plan will give the
Mac OS a path to a thoroughly modern
future as well as eventually carry forward compatibility with System 7.x
applications through a Mac OS-compatibility box
code-named Blue Box. With Blue Box, your existing applications will
continue to run as they do now, albeit more slowly
(how much more slowly has yet to be determined). However, Blue Box will not
appear in Rhapsody until sometime in
1998, well after the Premier Release of the new OS.
As a result, many users, satisfied with their current software, may choose
not to upgrade to Rhapsody, at least not at
first. Others, because of hardware limitations -- to date, Apple has
committed only to supporting Rhapsody on currently
shipping and future PowerPC-based Mac OS systems -- will be unable to
switch without acquiring new hardware. Which
is why Apple will continue to support and enhance System 7 for several
years to come (see the "System 7 Forever"
sidebar).
Perhaps a bit of terminology is in order here. What you have known and
loved as "System 7" for lo these many years is
now being referred to by Apple as "Mac OS." (Apple's latest release, for
example, is called Mac OS 7.6, not System
7.6.) Apple's future System 7 releases will continue to be branded as Mac
OS upgrades. Meanwhile, Mac OS 8, the
official name of what was originally code-named Copland, has an uncertain
future. Apple hasn't said whether its new OS
will be called Mac OS 8 or not. For now, the new operating system has only
a code name -- Rhapsody. So, when we talk
about the Mac OS, we mean the System 7-based Mac OS, and when we talk about
Rhapsody, we mean the
OpenStep-based modern OS, with System 7 compatibility.
Apple has laid out an ambitious timeline for how and when it intends to
deliver successive versions of these two operating
systems (see the "Apple's Dual-OS Road Map" figure). Rhapsody will be the
OS in which Apple first offers its new
modern OS and object-oriented OpenStep technology, but innovation on the
human-interface front will take place first in
the Mac OS and then be folded into Rhapsody.
=46or example, the interface enhancements Apple builds into Tempo, the
version of the Mac OS due in July 1997, will form
the basis of the interface used in the Rhapsody Premier Release, which will
follow several months later. And the interface
enhancements in Allegro, the Mac OS release due in January 1998, will be
rolled into the Rhapsody Unified Release, due
out in mid-1998.
Stone Soup
If you're one of those who thinks the Mac OS' user interface is the best in
the world, you'll be pleased to know that
Apple agrees with you. Rhapsody's UI will be based firmly on the Mac's and,
as we just mentioned, will be derived
from work that first appears in the Mac OS. Only in those cases where the
NeXTstep UI offers features that the Mac's
doesn't -- such as its application dock and inspector panels -- will Apple
incorporate elements of the NeXTstep user
interface into Rhapsody.
So you can expect Rhapsody to include features such as drag-and-drop
printing, close and zoom boxes in familiar places,
a single unified menu bar at the top of the screen, and icons on the
desktop. (And, if we're lucky, Apple will take this
opportunity to get rid of the Chooser.)
This =96 la carte approach to constructing the new OS extends beyond the
arena of user interface. Wherever Apple and
NeXT both offer similar functionality, Apple is likely to opt for its own
implementations. OpenStep, for example, has a
3D Graphics Kit, but Apple is unlikely to favor it over its own QuickDraw
3D. NeXTstep provides support for Pantone
color matching and color-monitor and -printer calibration, but again we'd
be surprised to see Apple adopt NeXT's
technology over its own ColorSync, which has found wide acceptance in the
desktop-publishing industry.
One critical area in which Apple will go with NeXTstep middleware instead
of its own is in how the operating system
draws things to the screen. Since the first release of the Mac OS, back in
1984, the Mac's graphics-display model has
been based on QuickDraw. NeXTstep, in contrast, utilizes Adobe's Display
PostScript. Here, Apple has chosen to break
with tradition and incorporate Display PostScript or its more contemporary
cousin, Adobe's Bravo, into Rhapsody.
Although developers may have to do some extra work to get used to the new
graphics model, Apple's choice is likely to
be a boon to desktop publishers. Display PostScript is a variant of the
same language -- PostScript -- that is used widely
in high-end printers and imagesetters. Using the same language to draw
images on the screen that is used for printing will
allow desktop publishers to preview the results of their work on-screen
with greater accuracy.
Many other Apple technologies that are key benefits of the Mac are likely
to find their way into Rhapsody as well. The
QuickTime Media Layer -- which includes QuickTime, QuickTime Conferencing,
QuickTime VR, and QuickDraw 3D --
is a prime example of Apple technology that will make it into Rhapsody as
soon as possible. Apple has also committed
unequivocally to including and expanding OpenDoc, its component-software
architecture, in the new OS. Apple may also
look for an opportunity -- at long last -- to integrate the extended
character sets and line-layout-manager functionality of
QuickDraw GX into Rhapsody. Game Sprockets, text-to-speech and
voice-recognition technologies, and AppleScript are
other likely candidates for integration.
In other areas of the OS, it remains unclear which direction Apple will
turn. In networking, for example, Apple has put a
great deal of effort into such features as plug-and-play networking, Open
Transport, remote access, and file sharing.
NeXTstep, on the other hand, has some robust network-management
capabilities. Another example: Both operating
systems offer tools for localizing applications for languages other than
English, including 2-byte languages such as
Japanese that are not based on the Roman character set. In these cases,
Apple's choice will most likely be based on an
assessment of which technology delivers a better user experience -- and
which can be integrated into Rhapsody more
quickly.
Developer Dilemma
Mac developers today typically support more than one platform. On the Mac
alone, most applications still ship in both
680x0- and PowerPC-native versions. Many developers also generate
applications for Windows from the same code base
they use to create Mac apps. With Rhapsody, the picture will become still
more complex.
The economics of the marketplace will pull developers in two directions. On
the one hand, they will want to quickly port
their Mac OS applications to Rhapsody to tap into the rich
application-upgrade market that will no doubt accompany the
new OS. On the other hand, until a significant majority of Mac users switch
over to Rhapsody, developers will still need
to deploy new versions for Rhapsody as well as for the Mac OS. If Apple
wants to guarantee the success of its new OS,
it would be wise to (a) commit both time and money to ensuring that
development tools exist to ease the burden of
supporting both OSes and (b) make it possible to develop off a single code
base.
Of course, this may be a challenge. Most Mac and Windows application
development these days is done in C++.
However, NeXTstep and OpenStep both base their development on the more
fully object-oriented Objective C. Although
developers may like what Objective C offers them in terms of new
functionality and reduced coding time, some may be
reluctant to learn another development language and maintain a separate
code base to support Mac efforts. To ease this
burden, the most aggressive developer of Mac OS development tools,
Metrowerks, is modifying its C++ compilers to
support Objective C objects in NeXTstep and OpenStep. However, the
Metrowerks compiler will only port C++ to
Objective C; additional work will be required to take advantage of
Objective C's unique features.
And then there's the challenge of packaging and marketing two versions of
the same application into the relatively small
Mac market. Some developers may choose to support only one OS. Which they
choose will depend largely on what is
more important to them: gaining access to Rhapsody's greater functionality
and performance or going with the Mac OS'
larger market share.
Keeping Promises
The next two years will be decisive ones for Apple. In committing to
deliver the Rhapsody Premier Release by year's end
and the Unified Release by mid-1998, Apple has set itself a Herculean task.
=46rankly, we're a bit skeptical that a company
that was unable to deliver Copland, despite a three-year effort, can pull
it off. But Apple did make good on its promise to
ship Mac OS 7.6 in January. And it seems well on its way to a repeat
performance with the Tempo release in July. And, it
should be noted, the team tasked with delivering Rhapsody is not the same
group that labored with Copland -- in many
respects, Apple's past performance in the OS arena cannot be used as a
predictor of how the newly reinvigorated OS
organization will perform.
Still, combining bits and pieces of two distinct operating systems,
together with a sprinkling of completely new software
technology, and ensuring that the result will run properly not only on
Apple's own Power Macs, Performas, and
PowerBooks but on clone vendors' systems as well is a daunting task.
Apple's ability to fulfill the commitments it has
made about Rhapsody could well determine the fate of an operating system
that even Windows users will tell you, in
private, has the potential to be just plain better than anything else around=
=2E
The Verdict / We Need a Bohemian Rhapsody
In laying out its plans for Rhapsody, Apple appears to have taken its own
marketing slogan -- "Expect the Impossible" --
to heart. Barely back on its feet after scrapping the Copland project,
Apple's OS management has constructed an intricate
obstacle course and has promised to negotiate it with a deftness yet to be
demonstrated by any other OS vendor.
Whether they'll succeed remains an open question. Perhaps what matters more
at this point is how that success is
defined.
The success of Rhapsody hinges entirely on what Apple delivers when. In
laying out its two-year plan, Apple has
(perhaps wisely) left itself quite a bit of wiggle room in this department.
=46or example, the Rhapsody timeline calls for
"some" Mac OS compatibility in the Premier release. How much is "some"? At
press time, Apple wasn't saying.
Perhaps more important, Apple has yet to specify exactly which of its own
technologies will be integrated into the
Premier release. As we see it, this is the critical question. Why? Because
customers who are most likely to adopt
Rhapsody first are publishing, Web, and media professionals. These power
users are the ones who need Rhapsody's
performance and stability most. But they are also the ones who most rely on
middleware such as the QuickTime Media
Layer and ColorSync.
What concerns us most is that in its eagerness to deliver the Premier
Release within a year as promised, Apple will opt to
leave out technologies that developers desperately need in order to deliver
groundbreaking apps -- apps that give these
media professionals a reason to adopt the new OS.
But Apple faces a far more serious challenge than having to implement
technology and deliver it on schedule. It has to
convince loyal Mac users that Rhapsody will be worth the wait. It's not
going to be enough just to repeat the mantra "The
Mac is great. Stick with us."
Windows, particularly Windows NT, is nipping at the Mac's heels. Microsoft
knows that while Mac users are waiting for
Rhapsody, many of them will be tempted to make the switch. More and more
Windows machines are showing up in
prepress houses and Web-authoring shops. And by the time the Rhapsody
Unified Release ships, it will be competing not
against today's Windows NT but against NT 5.0.
Apple may well meet the challenge of delivering Rhapsody on time. But it
needs to do more. It needs to clearly articulate
how Rhapsody will serve the needs of its target users -- better than what
Redmond will have to offer. And it needs to start
articulating that message now. We're happy that Apple moved away from the
"hard stop" strategy that would have tried to
force users to adopt its modern OS and dump System 7. However, with the Mac
OS safety net firmly in place, it becomes
even more critical that Apple give users compelling reasons to move to the
new OS.
--I got two turntables and a microphone...
<> tbyars@earthlink.net <>