After Three Years, It's Time to Begin
Apple's New Operating System Is a Mixture of Good and
Bad
By Jason Snell
It's been almost three years since I started covering the development
of Mac OS X. During the course of those three years there were times
it seemed that Apple's task was an impossible one, and other times when
OS X's arrival seemed like a sure thing, provided we expected it to
appear on our desktops at about the same time that the sun expanded
and turned the Earth into a cinder. As a result, it's a bit mind-boggling
to consider that I'm typing these very words in a real, shipping version
of Mac OS X.
Saturday marks the official beginning of Mac OS X as a "real" product.
And on Wednesday, we in the press got a chance to spend several hours
with Apple executives, hearing them give their final take on the software
they've slaved over for all these years before they open the gates and
release it into the wild.
Mac OS X is a complex piece of software that has many amazing features
and a lot of blemishes. Therefore, anyone who raves about it uncontrollably
isn't telling you the truth, but anyone who writes it off is not particularly
attached to reality.
Impossible Mission
As we start this new phase in the Mac's life, it's only fair to point
out what a massive accomplishment Mac OS X is, even in its current work-in-progress
form. Three years ago, when Apple claimed that this new, Unix-based
operating system, originating from Next's softwar, would run all old
Mac applications as well, we were all pretty skeptical. Mac OS X's Classic
mode may not be perfect -- in fact, it's quirky, a bit weird, and probably
something best suited for only occasional use -- but it really works.
Equally as mind-blowing is the fact that Mac OS X is a bubbling cauldron
of Unix beneath the surface -- and yet from above, all the users see
is serene, antialiased text on an extremely familiar interface. Apple's
executives made a point Wednesday: by the end of the year, Apple will
be the world's largest supplier of a Unix-based operating system . .
. even if most of its users don't know that there's Unix underneath.
That's really cool. It's also quite intriguing because perhaps the Unix
community will embrace Mac OS X as a suitable operating system and a
solid partner in the battle to keep Microsoft from controlling every
last computer in existence. We may see a flood of Unix software being
brought over to the Mac -- not just in the command-line interface you
can get to from Apple's new Terminal application, but with Mac developers
creating friendly Aqua interfaces to control the software running underneath
the surface.
Likewise, for all the complaining the Mac community did about Apple's
interface experimentation with OS X, the final result is recognizably
a Mac interface. I would suggest that it's not much more of a radical
step in interface design than the jump from System 6 to 7 from System
7 to Mac OS 8.
Sure, some interface pieces have come and gone, but it's still fundamentally
a Mac. And over time we'll grow into it like we have grown into the
Mac OS 9 interface.
It's Not Over Yet
Now for the other side of the Mac OS X story. It's not a finished operating
system and doesn't quite work right. Even Apple admits that, saying
that it will use its Software Update engine to roll out many bug fixes
and feature additions to Mac OS X in the coming months. (Among them:
DVD movie support, due this spring; and CD burning support, due in April.)
However, some of my issues with Mac OS X are ones that Apple hasn't
really talked about, and chief among them is performance. We've been
told that Apple had to move to Mac OS X because there was just too much
junk in the old Mac OS. And that's true -- with Mac OS X, Mac users
now have memory protection and preemptive multitasking, two great features
that add to the usability and stability of the operating system.
But another thing that Apple has always said was that the classic Mac
OS wasn't very efficient at taking advantage of the PowerPC processor,
and so as a result Mac OS X would be faster than its predecessor. When
it comes to my experience with Mac OS X 10.0, that's not true at all.
Running Mac OS X, my Mac -- and my 500MHz PowerBook G3 is not exactly
a slowpoke -- seems much less responsive much of the time.
Sure, dragging a window around on the screen happens in a flash. But
when I try to grab the bottom-right corner of a Finder window and resize
it, my cursor far outpaces the edge of the window. It's sluggish when
I try to increase the size of a Finder column through dragging, too.
In general, everything seems just a little bit slower than it does when
my PowerBook is booted into Mac OS 9. The one exeception is waking up
from Sleep mode: that happens in a matter seconds.
Over the next few weeks, users will find all sorts of peculiar bugs
or missing features, many of which will only manifest themselves to
the small fraction of Mac users who take advantage of a particular feature.
A colleague of mine complained to me just the other day about the fact
that you can no longer grab a window by its edge and slide it around;
either you grab it by the title bar or you leave it where it is. As
someone who always moves my windows via their title bars, I'd never
noticed it.
I am, however, one of those people who jumps to items in the Finder
by typing their names. You know the situation: you've got a Finder window
open, but it lists 200 different files and folders and you're at the
top. You know you want to get to one that begins with the letter t,
so instead of scrolling meticulously down to it, you simply type t and
the window automatically scrolls down to that point and selects the
first item beginning with t.
In the Mac OS X Finder's List view, you can type t and it'll select
the first t item, but it won't actually scroll the window. A little
thing, but my Finder productivity is now greatly reduced. Sure, this
bug may never affect you -- but another one will. That's how it's going
to be for a while for OS X pioneers.
Script Issues as a Metaphor
I rely on AppleScript a lot. And oddly enough, Apple's scripting language
is really a microcosm of Mac OS X. It's a remarkable technical achievement
that AppleScript is even present in Mac OS X, and it works much better
than anyone ever expected. But parts of it aren't there.
Almost every script I've moved over from Mac OS 9 applications to their
Mac OS X-native equivalents has worked without any trouble at all --
an amazing feat. But for people who use certain AppleScript features
in Mac OS 9, they'll find the new OS lacking. It doesn't support program
linking, folder actions, or network scripting (including AirPort, modem,
and remote access settings).
However, there's good news. When I asked Steve Jobs about AppleScript
on Wednesday, he answered loud and clear: "We're very committed to AppleScript."
And Apple marketing czar Phil Schiller agreed, saying that AppleScript
support in Mac OS X will only keep getting better.
That's pretty much the story for Mac OS X as a whole: it's here and
Apple's committed to making it better. It'll work for some people; others
will take a pass because it doesn't do something the way they want to
do it.
Mac OS X is the future of the Mac. It's where all Mac users who stick
with the platform are going to end up. But despite the huge amount of
work Apple's software engineers have done, and the amazing technical
achievements they've pulled off, this is only the beginning. There's
much more work to be done before they can rest easy and call Mac OS
X a success. For now, after three years, it's still a remarkable work
in progress.
JASON SNELL is the editor of Macworld.