The Weak Case For Microsoft

[ The opinions expressed by authors on Linux Today are their
own. They speak only for themselves and not for Linux Today.
]

By Mike Cornall

Standard Disclaimer: While I believe that Microsoft should be
prosecuted for individual acts of sabotage and fraud, I am opposed
to the DOJ anti-trust action. I consider it an improper government
intervention into the market, and a dangerous precedent.

That said, there’s no way that I can let Bill Gates’ latest
collection of half-truths and outright groaners pass unchallenged
(not that I’m the first to challenge it) . . .

In the Time article, The Case For Microsoft (1), Mr. Gates attempts
to explain why separating Microsoft’s application development, from
its OS development, would be a bad thing. Mr Gates goes on to claim
that the split would cause untold harm to consumers, citing
examples of past Microsoft “innovations” that might never have
occurred, had Microsoft been split at the time.

Note that, while Mr. Gates is arguing here that it is vital for
Microsoft’s application developers and OS developers to have close
contact, Microsoft argued in court that their application
developers gained no advantage from Microsoft’s ownership of the
OS. Ah well. Consistency was never one of Microsoft’s strong
points.

In writing for Time, Mr. Gates is aiming his article at the general
public. It’s a good thing too — for him — because only the
general public, lacking an understanding of PC technology and
history, can fail to see through the transparently thin
arguments.

Let’s examine the article on a point-by-point basis:

Our company could not have created the Windows operating
system if we had been prohibited from developing Microsoft Office
as well.

Apparently, Mr. Gates would have us believe that an OS-only
Microsoft would have stood back and let DOS fall behind, while
other GUI-enabled systems (Macintosh, Amiga, Atari-ST, UNIX, etc.)
took over the market. Considering how DOS and Windows have been
allowed to stagnate from time to time, it’s tempting to concede his
point, but I can’t — it’s ridiculous.

The fact is, before Microsoft ever produced a successful version of
Windows, there were already at least two GUIs in widespread use on
DOS, namely, Digital Research’s GEM, and GeoWorks. The latter,
especially, showed great potential. Released in 1990, GeoWorks ran
on an XT, in 640K, and provided a full GUI, pre-emptive
multi-tasking, and a basic Office Suite. Geoworks was developed by
a company just a fraction of Microsoft’s size, yet it had many
features that surpass Windows, even to this day. It was precisely
the desire to prevent the development of non-MS GUIs, such
as GeoWorks, that led Microsoft to tie the Windows GUI into
DOS.

The symbiotic nature of software development may not be
obvious outside the industry,

Actually, by providing knowledge of certain aspects of the OS to
Microsoft application developers, while denying that same knowledge
to outside application developers, Microsoft has prevented
the very symbiosis of which Mr. Gates speaks. If you want to see
what symbiosis can achieve in an open environment, just look at the
rapid development of Linux and various Linux applications. While
Microsoft has struggled to get out Windows 2000, Linux has spread
to a dozen platforms, from embedded systems, to IBM mainframes.

but it is a phenomenon that has produced enormous consumer
benefits.

I don’t consider the (near) destruction of competing applications,
such as GeoWorks, Ami-Pro, WordPerfect, Netscape, and so on, to be
beneficial. The close ties between Microsoft applications, and the
Microsoft OS, have served mostly to limit consumers’ choices.

Windows and Office — working together and drawing on each
other’s features and innovations — have improved personal
computing for millions.

In many cases, Microsoft has defeated competitors’ applications by
introducing incompatibilities into DOS or Windows, or hampered them
by limiting access to the full API set. In doing so, and thus
limiting competition on the Windows platform, Microsoft has denied
PC users the opportunity to take advantage of the often-superior
features in many non-Microsoft products. The list is endless, but
examples include the pin-able menus and flexible fonts in GeoWorks,
the function-key paragraph formatting in Ami-Pro, the “Show Codes”
facility in WordPerfect, the better virus protection in Netscape
and Java, and so on.

Furthermore, I am not aware of a single beneficial feature of MS
Office that results specifically from its close ties with Windows.
In every case where such a tie exists, the sole benefit seems to be
for Microsoft (by locking out competing applications), to the
detriment of Microsoft’s customers. Invariably, in such cases, an
implementation that avoids ties with the OS would be more more
stable, easier to implement, more standards-compliant, and better
all around for the consumer.

Take the tablet PC . . . A small, lightweight, portable
device, it will enable you to take notes, dictate, annotate and
then seamlessly transfer everything to a PC or any other
device.

That sounds a lot like a Palm Pilot, or a Psion.

Under the government’s plan, however, Microsoft’s tablet PC
simply won’t happen, because our OS and applications developers
will be unable to collaborate.

And yet, somehow, the people at 3COM managed to do it, having no
access to the Windows OS, and before Palm applications even
existed. Amazing.

Also, note the fact that there are a huge number of applications
for the Palm, none of which — beyond the half-dozen basic programs
that come included — were written by 3COM.

Almost every aspect of the tablet PC’s evolution —
starting with the design of handwriting-recognition applications —
requires real-time collaboration between OS and applications
developers.

And yet, every Palm application can use Palm’s pen input, despite
having been written by companies other than 3COM.

Note the hand waving here (no pun intended). Gates has chosen the
most borderline function he can, namely, handwriting-recognition,
which some would deem a function of the OS, while others would
consider it an application. It’s an old trick: defend the wrong
side of a black and white issue — the browser in this case — by
pointing to a gray issue.

Come to think of it, except for the driver to detect the pens’
motion, handwriting-recognition is probably better as an
application, because then I can replace it with a different
handwriting app. that is more to my liking (as, in fact, can be
done with the Palm).

Today that happens spontaneously, just as it does at IBM
and Sun Microsystems. Real-time collaboration is the cornerstone of
software development.

Really? So why didn’t Microsoft come up with the Palm Pilot?

Just as chassis developments at Lincoln (owned by Ford) are
shared with Ford’s other car divisions, Microsoft takes the best
thinking among its applications software developers and shares it
with Windows developers (and vice versa).

That’s how it used to work, but it proved to be an inefficient
model. Today, a car can be assembled out of components manufactured
by various independent companies around the world. It’s quite
common, for example, for an American car to have a Japanese
engine.

Just because Ford’s Taurus is an American best seller,
should the company be barred from sharing its innovative work among
its divisions? . . . If consumers’ interests are paramount, the
answer to each of these questions is clearly no.

Because of my position on government intervention, this is the one
point I would concede to Mr. Gates. I know that others will
disagree with me, arguing that since Microsoft has been declared a
monopoly by the court, Microsoft must abide by a different set of
rules.

Windows never would have gained popularity and reached
critical mass without the benefits of innovative, user-friendly
technologies developed by our Office team — technologies that
often then became part of Windows and further drove innovation
across the industry.

So what Mr. Gates is saying here is that Windows is such a bad OS,
that no one would have developed any applications for it, if
Microsoft had not done it themselves.

This would seem to be a unwise admission, especially now, when so
many PC users are looking at Linux as an alternative to
Windows.

For example, in 1991 software developers for Microsoft
Office introduced a new feature known as a toolbar. We now take
toolbars for granted . . . Had those toolbars been created
elsewhere, they no doubt would have been patented and never
incorporated into Windows.

It seems to me that I was using various toolbar-like features prior
to 1991. As to the patent comment, is Mr. Gates suggesting that
Microsoft has done us a great favor by declining to patent
toolbars? On the contrary, I would consider anyone who tried to
patent such a basic concept to be a complete jerk. So even assuming
that Microsoft’s choice not to patent toolbars was voluntary (i.e.
that they could have avoided the prior art claims), Microsoft would
have to do more than simply not be a jerk if they want to impress
me.

Once added to Windows, toolbars became available for use in
software programs created by Microsoft and thousands of independent
companies. That is the great efficiency of innovation in platform
software.

Yes, and by tying it to Windows, Microsoft killed any further
innovation in this area.

On Linux, where the toolbars are provided by separate GUI
applications, I have a choice between half a dozen different styles
of toolbars, yet each one works fine with other Linux applications.
In fact, sometimes I run more than one toolbar at the same time
(ICE for its compactness, plus a minimized Gnome Panel for its
on-the-fly reconfigurability).

Had the proposed plan to dismantle Microsoft been
implemented 10 years ago, such innovations might never have found
their way to broad consumer availability.

DesqView provided DOS multi-tasking before Microsoft killed it.
GeoWorks provided a full GUI environment and basic Office Suite
before Microsoft killed it. Norton provided a taskbar and “start”
menu. The list is endless.

Had Microsoft been just an OS developer, with no reason to sabotage
competitors’ applications, we would have a great many more choices
available under Windows (or DOS) today. Who knows what GUI
developments and other innovations might have occurred?

They never could have moved from the “applications” company
to the “OS” company that the Justice Department envisions.
Consumers and developers would have been harmed.

Since, so far, all of the examples have been of things that
should not have been moved into the OS, and since
consumers have, in fact, been harmed by moving them into the OS,
Mr. Gates does not have a very strong point here.

Also, though I agree that government involvement can stifle
innovation, Mr. Gates still does not have a strong point, because,
to the best of my knowledge, Microsoft has never innovated (2).

The DOJ plan reflects a profound hostility to Microsoft’s
efforts to make products that work well with one
another.

While I have my own disagreements with the DOJ, Mr. Gates statement
here is blatantly false.

For example, the plan would effectively prohibit the new
Windows and applications companies from engaging in technical
discussions to develop new versions of Windows and
Office.

I’m sure the DOJ would not object to a conference that included
many different application developers, one of which was the MS Apps
group. I’m also sure there would be no objection to the MS Apps
group putting in a request for a new feature, so long as the
request is made in an open forum. Who knows — with input from a
variety of application developers, it might even make Windows
better.

The DOJ scheme permanently prohibits any further
improvements to the Internet software in Windows. It would mean no
improvements in browser technology and no support for new standards
or technologies that would otherwise have helped protect your
privacy or the safety of your children online.

This too is blatantly false. There is nothing to prevent browser
enhancements, or improved standards support.

In fact, as demonstrated by the recent “ILoveYou” virus, it is the
close ties between the Windows OS, and its applications, that
presents the greatest security risk. Furthermore, as demonstrated
in the Halloween Document (3) by Microsoft’s threat to
“decommoditize protocols”, those same close ties also represent the
greatest threat to standards.

Reading further, note the implied threat in the above quote. Mr
Gates is saying, in effect, “If you don’t let us do what we want,
then your children will be at risk”.

When it comes to protecting children, by placing limits on their
computer use, parents are much better off with Linux. In Linux, OS
functions have been carefully separated from application functions.
Because of this separation, a child can be given his own secure
account, where he only has access to the directories and
applications specified by his parents. Under Linux, a child would
not be able to disable filtering by simply running his own
program.

The DOJ scheme also effectively imposes a ban of up to 10
years on the addition of any significant new end-user features to
Windows.

No it doesn’t (I’m ignoring the intentional ambiguity in the words
“end-user”).

New features must be provided on an a la carte basis and
priced separately to computer manufacturers.

Only if those features are, in reality, independent
applications.

However, good luck to the DOJ trying to make this part stick. As
with all government regulation, it’s impossible to word this
section in a way that clearly defines where an application ends,
and the OS begins. In other words, this provision will be toothless
— Microsoft will always be able to find a way around it.

Provisions like these would kill innovation in the OS–and
impair the livelihoods of the tens of thousands of independent
software developers who depend on constant innovation in the OS to
make their products more attractive.

Strange, Mr. Gates didn’t seem to care about the livelihoods of the
developers whose products were sabotaged by Microsoft. Where was
his concern for the DR-DOS developers? What about the Netscape
add-on developers? Or how about the potential Java developers whose
careers were cut short by Microsoft’s decision to “pollute”
Java?

Updates to Windows and Office technologies that could, for
example, protect against attacks such as the Love Bug virus would
also be much harder for computer users to obtain.

Let’s think. We could limit a user’s access to system files, as
does Linux. Or we could display a strongly worded warning before
running an attached executable, as does Netscape. Or we could use a
scripting language that runs in a secure sandbox, as does Java.

Mr. Gates must be showing his age here, because his system design
skills seem to be failing.

The effect of this lawsuit will be to punish Microsoft no
matter what harm this does to consumers, software developers, the
industry that has driven America’s remarkable growth–or, indeed,
the entire economy.

The lawsuit had one benefit, which was to bring out evidence, for
all to see, of the corrupt nature of Microsoft. From this point on,
I predict that the case will have little effect on Microsoft. The
case will drag on into appeals, any sanctions will be watered down
and twisted, and Microsoft will carry on as they always have. The
private lawsuits to follow, however, may have some merit, and,
based on the evidence gathered by the DOJ, may achieve some
success.

As to the consumers, Mr. Gates is crying crocodile tears — he has
never hesitated to harm consumers when it was to Microsoft’s
advantage. I would like to hear him explain how it benefits
consumers when Microsoft tries to “pollute Java”, or “decommoditize
protocols”. In my opinion, consumers are harmed by Microsoft’s
existence, and, along with the entire economy, would benefit from
Microsoft’s disappearance.

That is why Microsoft plans to appeal the district-court
decision, which is at odds with a decision of the U.S. Court of
Appeals and with antitrust law.

It won’t matter, because it’s not the DOJ that’s going to restrain
Microsoft’s behavior, and benefit consumers. That honor lies with
Microsoft’s newest, and toughest competitors — Open Source
projects, such as Apache, Mozilla, BSD, and Linux.

References:

(1) Time: The Case For Microsoft:

http://www.time.com/time/magazine/articles/0,3266,44557,00.html

(2) The Microsoft “Hall of Innovation”:

http://www.vcnet.com/bms/departments/innovation.shtml

(3) opensource.org: The Halloween Documents:

http://www.opensource.org/halloween/