# United States v. Microsoft Corporation; Public Comments; Notice (MTC-00030631 - MTC-00032327)

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3AX02-140503

## Record

- **Collection:** Federal Register
- **Document type:** Notice
- **Published:** May 3, 2002
- **Citation:** 67 FR 29276

## Text

[Federal Register Volume 67, Number 86 (Friday, May 3, 2002)]
[Notices]
[Pages 29276-29794]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: X02-140503]

TO APPENDIX TO MEMORANDUM OF AMICI CURIAE IN OPPOSITION TO
PROPOSED FINAL JUDGMENT IN CIVIL ACTION NO. 94-1564
SIGNED BY GARY REBACK
2ND STORY of Level I printed in FULL format. Copyright 1994
Network World, Inc. Network World July 25, 1994 94-1564
FILED FEB 14 1995
SECTION: TOP NEWS; Pg. 4
LENGTH: 724 words
HEADLINE: Microsoft free at last?;
Ruling still lets firm incorporate apps in its OSes.
Clerk, U.S. District Court
District of Columbia
BYLINE: Michael Csenger and adam Gaffin
BODY:
Washington, DC
The antitrust settlement Microsoft Corp. reached with the
Justice Department skirted an issue central to network users, paving
the way for the software giant to continue integrating applications
with its desktop and network operating systems.
The consent decree, announced July 16, focused almost entirely
on the way Microsoft sold operating systems to hardware vendors. But
it does not prevent the company from integrating applications into
the operating system itself.
Competing software vendors such as Lotus Development Corp. had
long alleged that Microsoft's applications division received unfair
information from its operating systems division that gave the
company a leg up on the competition.
Some analysts and users said the decree, which also poses
stricter controls on the royalties Microsoft can collect from
personal computer vendors, leaves the path clear for Microsoft to
mop up competitors that sell stand-alone applications, resulting in
more limited user choice down the road. SKEPTICISM
But others said Microsoft has yet to prove to the market that it
has operating systems and networked applications worth betting a
business on. ``A lot of its networking products are either
futures or first-generation products,'' said Jamie Lewis,
president of The Burton Group, a Salt Lake City consulting firm. The
company faces entrenched and growing user bases for both Novell,
Inc.'s NetWare operating systems and Lotus'' Notes groupware
applications, he said. Users also expressed skepticism.
``Microsoft promises Chicago and Cairo and a whole lot of
networking, but the question is, will it work before they run out of
cities to name these things after?'' quipped a network manager
whose major brokerage house network runs on Unix.
windows NT is not a truly open environment, he said,
``Because if Gates doesn't have it then neither do you, and I'd
rather not put myself in his hands. That's why we've standardized on
Unix for our trading floor.''
Frank Caro, technology transition team leader for Otis Elevator
Co. in Farmington, Conn., cited interoperability problems with
Microsoft's current Windows implementation of Transmission Control
Protocol/Internet Protocol as an example of the company's network
shortcomings.
``We've been trying to get into the networking capability
of Microsoft's products and find there's one con, non theme:
NETBIOS,'' Caro said. Microsoft does not yet support native
TCP/IP, but uses NETBIOS or NETBEUI encapsulated within TCP/IP, he
said.
``we're totally uninterested in any approach like this; it
can't handle a network of more than 50 users and is terrible over
the wide area,'' Caro said.
And Windows NT has proved unable to handle the applications that
Otis wants to take off its mainframe system, because Windows NT is
not a multiuser environment.
But Caro respects Microsoft's ability to change course as
necessary and awaits the promised native TCP/IP support in Chicago.
``That one feature alone is going to cause dramatic change
in network connectivity,'' said Nick Lippis, principal at
Strategic Networks Consulting, Inc. in Rockland, Mass., referring to
Windows'' TCP/IP.
Native TCP/IP support for Chicago could help Microsoft cut into
Novell's installed NetWare client base by providing an alternative
to Novell's Internetwork Packet Exchange (IPX) protocol. If the
desktop operating systems supported TCP/IP directly, ``why
continue with IPX?'' Lippis asked.
NOVELL NOT WORRIED
``I laugh when I hear people say it's all over for Novell
now, we should pack up and go home,'' said David Bradford, vice
president and general counsel for Novell.
``Microsoft has come against Novell [several] now with
their networking products, and we've beat them every time,''
Bradford said.
Bradford also noted that this consent decree does not close
Microsoft's books forever. ``They will be monitored, perhaps
even more so than before,'' he said. ``The industry and
consumers have an ally in the Justice Department.''
Frank Dzubeck, president of Communications Network Architects,
Inc., in Washington, DC, agrees that the case may not yet be closed.
``If Microsoft gets very aggressive and starts burying
things in their operating systems, then this whole issue will be
revisited, he said. But it will require that another company first
go bankrupt.''
LANGUAGE: ENGLISH
LOAD-DATE-MDC: July 27, 1994
TO APPENDIX TO MEMORANDUM OF AMICI CURIAE IN OPPOSITION TO
PROPOSED FINAL JUDGMENT IN CIVIL ACTION NO. 94-1564
SIGNED BY GARY REBACK
Business Day
M??
The New Work Times
Microsoft's Barely Limited Future
By JOHN MARKOFF
??
SAN FRANCISCO, June 17--Rath?? than reining in the
Microsoft Corporation, the consent deeree that the Justice
Department announced over the weekend with Microsoft. the world's
largest software publisher, frees the company to define the computer
mudstry's ground ?? through the rest of the decade.
The agreement leaves ??intouched what many computer in??justry
executives say is Microsoft's ?? advantage--that it devel??ps
both the basic operating-system ??oftware that makes personal
com??ters run, known as MS-DOS. and ??pphactons software, like word-
pro??essing programs or spreadsheets, ??nat perform spec?? ??
``Microsoft s whole empire is based in the interlocking
nature of their ??perating-system and application oftware.''
said William Joy, a ??ounder of Sun Microsystems, and the ??uthor of
one version of the Unix Perating system.
??Vol a Central Issue Microsoft officials said Saturday ??nal
issues related to the relationship ?? their operating software and
their ??ppicaons programs had not been ??ocus of their ??
nego??anons with us??ce Department officials.
MS-DOS and the Windows proram, which makes DOS easier to ?? are
installed in millions of com. ?? worldwide White the Jusuce
??epartiment has decided that Micro?? does have a monopoly in
opera?? ?? systems, it ?? that the ?? changes the c??unsent decree
spells ??ut provide a remedy.
Yet many Microsoft compet??nors ??ce a broader problem, as well:
the ??ne between where the operating system ends and the
applications pro??las start is increasingly being ??lurred by
advances in technology. ??Smaller compe??tors with innovave ideas in
businesses as diverse as ?? man. ?? compression,
?? creates more storage space on disk, and screen savers, which
pre?? ent damage to mounors, are finding ?? their business is
evaporating because Microsoft keeps adding such programs to ??
operating system as ?? periodically brings out an updated version.
A Microsoft's operating system scheduled for release next year,
called Chicago, will acceler?? the process The program will mer??e
DOS and Windows and will include electronic mail, remote access,
filesearching functions and screen savers. Since introducing MS-DOS
in 1981.
Microsoft has continually campaigned to expand the ?? of what
computing functions belong inside the computer operating system.
The early vers?? of DOS were small programs that did ?? more
than control the storage and ?? of data and start and stop
applications programs. But in the 14 years that followed,
Microsoft's ??rating systems have greatly expanded the servtees they
provide to users and programmers The other important issue not
specifically addressed in the consent decree is whether Microsoft
has been able to leverage us virtual monopoly ?? operating systems
into domination of applications software--a far bigger and more
lucrative market This matter is of great concern to companies like
Lotus Development.
Boarland International and Novell, and its recently acquired
Wordperfect--which specialize in applications software. About
half of the 50 million computers that run Windows, for example use
Microsoft's word processor, called Word, and its spreadsheet, Excel.
It was for that reason that lawyers at the Federal Trade
commission toyed two years ago with the idea of breaking Microsoft
into two companies, More recently, Justice Department investigators
are believed to have

[[Page 29277]]

studied ways of creating some sort of ``Chinese wall''
that might limit the information traveling between the two sides of
the business. Anne K. Bingaman, Assistant Attorney General in charge
of annt??rust matters, refused to comment on the issue. But in
response to a question whether, the department had considered trying
to split Microsoft, she said Sa??day that her lawyers, bad looked at
``every possible legal th?? ory ?? Linkage Is Seft-Pedaled??
In an interview today, Ms. Bingsman: acknowledged that the
decre?? was silent about any linkage ?? ?? Microsoft's power in
operating systems and its, growth, in applications software. But she
also said the Justice Department had decided against pursuing a
``second range of issues'' that had been raised by the
F.T.C.'s earlier investigation.
``All I can tell you is we filed the complaint based on
what we decided were the problems that needed to be
corrected,'' she said.
What the consent decree announced on Saturday did achieve was
this: Microsoft agreed to change the way it deals with the companies
that make the hardware for personal computers, freeing them to offer
customrs a choice of operating systems.
Microsoft will also alter its softwarelicensing policies and the
way it gives information to software developers.
The expectation is that personal computer makers like Compaq.
Dell and others will now be more receptive to the operating systems
made by Novell, international Business Machines and Sun
Microsystems.
Software companies will be able to develop versions of their
programs for Microsoft's operating systems without making exclusive
commitments to Microsoft, leaving them free to create applications
for operating systems that other companies have designed.
Yet while the consensus is that Microsoft's influence will
continue to increase, computer industry executives are divided over
whether its power and influence will be good or bad for consumers.
``Microsoft has become the I.B.M. of the 1990's'' said
J. Paul Gravson chairman and chief executive of Mr. crogra??, a
software publisher it Richardson. Tex ``There are issues for
anyone who wants to participa?? in this market because of their size
and scope. Anything the Government does to slow them down would be
welcome.''
Believes Bigger Is Better But others in the industry believe
that Microsoft's strategy is benefiting consumers.
``If you really care about improving the personal computer,
you wan Microsoft to take over all the pieces of the pre,''
said Stewart Alsop, edito?? of Infoworld, a weekly computer-in
dustry newspaper.
Competitors like Novell, which were otherwise pleased by the
agreement obtained by the Justice Department, said they were
disappointed that the Government had not forced Microsoft to
disclose ??formation about new versions of its operating systems in
ways that would level the playing field for developers who are
competing with Microsoft applications.
The company's competitors have argued that Microsoft has gained
a special advantage for its applications programs by using hidden
operating-system features and providing earlier access to technical
information for its programmers.
Microsoft officials said the Government had found no evidence
that such a special advantage existed. ``We don't think this is
market power in the traditional an??trust sense.'' Said William
h. Neukom, the company's vice president for law and corporate
affairs. ``Anyone can come in and upset you with better
technology. We think it's a ferociously competitive business.''
While the agreement may aid some companies like Novell, which
makes a Microsoft-compatible operating system, it will not affect
Microsoft's power with respect to smaller software developers.
``Microsoft will continue to be very powerful,'' said
Martin Goetz, a cofounder of Applied Data Research, the nation's
first software company ``The Justice Department hasn't ?? to
the cries of the software companies.''
Michael J. Miller
The World According to Microsoft FILED
If you think Microsoft is too dominant in today's computer
industry, a quick look at where the Bill Gates juggernaut is headed
may prove disheartening. Already the leading provider of operating
systems and office productivity applications, Microsoft wants to
carry its success over to other areas, ranging from interactive
television to financial services. With its recent announcements,
acquisitions, and introductions. Microsoft is making its goal clear:
It aims to become a ubiquitous part of tomorrow's information
infrastructure.
THE RIGHT TOOLS
While Intel seems to face more competition than ever,
Microsoft's position in the operating-system market has gotten
stronger. The reason for this continued success is twofold.
Confusion and a lack of focus from OS competitors--such as IBM
and Apple--certainly helped, but Microsoft also gave itself
quite a boost by developing tools like Visual Basic and Visual C++.
Not too long ago, Borland surpassed Microsoft in the quality of
its tools. But more and more, the big firms I talk to are moving to
Microsoft tools. This kind of support gives Microsoft the ability to
decide which technologies to push and which platforms to support, as
well as which technologies to license and which to keep for itself.
For instance, Microsoft was first on the market with products that
really supported OLE 2.0. Now that it wants OLE 2.0 to be widely
supported, it has done a very nice job of making OLE support easier
by providing Wizards in its Visual C++ package.
Microsoft wants OLE to be the object standard, and wants to
establish it before OpenDoc or Taligent gets off the ground.
Microsoft even wants to control object standards on other platforms,
hence its introduction of tools that make it easier for developers
to take Windows applications and move them to other platforms, such
as Macintosh, with built-in support for OLE. Not only does this I by
Mans Bishofs kind of accommodation push Microsoft's APIs. it also
makes it easy for vendors to use Windows as their primary
development platform, regardless of what their target system might
be. This will, of course, lead to code that is optimized for
Windows. (Okay. Microsoft is a bit confused here. This is because
part of the company wants to protect the rights of its Word and
Excel teams by insisting on special terms for using the cross-
platform code for people who write word processors or spreadsheets.)
TIE RIGHT NETWORK
The dominance in tools, applications, and operating systems may
be just the beginning. Consider Microsoft's recent announcements,
such as Microsoft Network, a new on-line service that will be
bundled with Windows 95.
Microsoft Network, once code-named Marvel, may well be the first
thing users see when they start the new operating system and it may
be the best way to get Microsoft support. If users choose to
subscribe to Microsoft Network, the company could wind up getting a
steady stream of $4 to $5 a month from everyone on its operating
system, and that could mean several hundred million dollars a year.
Microsoft isn't the only one with this idea. IBM is doing the
same thing with OS/2 Warp by bundling in Internet access through its
Advantis service, which then sets up a continuing monthly fee. In
fact, you can almost view these two operating systems as loss
leaders for their suppliers'' on-line services. Since Microsoft
is in a position where its operating system is dominant, however,
users will be more likely to try its network service first. In order
to be successful. Microsoft Network doesn't even have to be the best
on-line service; it just needs to be good enough and the most
convenient. And including Microsoft Network with windows 95 will
certainly help.
Now take Microsoft's recent plans to acquire Intuit with its
Quicken personal finance program (which links to a check-paying
system), and add that to the likelihood of Microsoft Network's
success. Because of its size. Microsoft is in a better position to
work out relationships with large banks and other financial players.
Imagine how Microsoft could extend electronic banking onto an online
service such as Microsoft Network.
Microsoft could require just a small service charge on each
transaction. Or it could make money on the float--the interest
in the few seconds it takes to move money from one place to another,
or both.
Microsoft's success in one area helps it extend its success in
other areas. Because Windows is so successful, developers must
develop for it. If Microsoft Network becomes successful, more
developers and content publishers will support it. The same
reasoning will apply to Microsoft's Tiger system for delivering
video and other content to set-top boxes, or even to the far-off
plan of developing wallet PCs with access to financial information.
UNCHARTED WATERS
All this may sound inevidtable, but it isn't. First of all, no
one--not even Bill gates--is successful with every product
he introduces. Just think about Microsoft Money. And does anyone out
there remember the first Microsoft Access, the abortive Crosstalk

[[Page 29278]]

competitor? Not too many folks, obviously, or Microsoft couldn't
have recycled the name for use on its database.
Microsoft still has a lot of strong competitors who envision a
different future. Novell, for instance, is still the clear leader in
network operating systems and has recently announced plans with
General instruments, the leader in cable set-top boxes.
To date, Microsoft's track record in communications products is
less than stellar, Lotus's cc: Mail and Notes have a larger
marcompetitor in the world to come. Micro?? is getting into areas
where it will face ?? competition, in addition to its ?? software
competitors, from banks to tel?? sion and cable companies. in many
ca?? these firms have unique relationships ?? customers or content
that Microsoft ca?? easily duplicate.
This more Microsoft focuses on pushi?? its existing platforms
and operating system?? the more likely it is that there will be so??
outside force, some new technology, th?? Microsoft either won't see
or won't comm?? to quickly enough. This would leave roo?? for new
competitors. Remember, it was?? too long ago that IBM. Digital
Equipme?? Corp., and Wang were the dominant infor?? mation
companies, and look what happene?? when the technology changed.
Still, if you're worried about Micros ?? dominance today, you
have good reason. ?? may foreshadow a future where Microsoft has a
hand in every area of your life--from communications to
entertainment to pay?? ing your bills. the road to this future woo??
be easy, but Microsoft is very determined and is certainly in a
better position tha?? inevitable storms.
There are 3,462 chances to make a mistake in this document.
(Typing it in is number one.)
If the ?? office is here. how come people keep handling you
pieces of ?? paper, ?? faxes. photo-cop?? and newspaper ?? for you
to ?? in your report? Even ?? to type if out in is a mistake. But
it's got to be done one way or ??.
That's where WordScan Plus 3.0 from C??ro, can help. You see.
WordScan Plus uses the ?? 32-bit Adaptive Recognition Technology so
its accuracy rate is unparalleted-- ?? when coupled with the
enhanced image capabilities of any Hewlett-Packard scanner.
In fact, Hewteff-Packard's AccuPage 2.0
technology--including ?? image ?? that lets you read text on
colored backgrounds, small text support, and ??-zoning--makes
WordScan Plus ?? for complex mixed-media Input ?? well as straight
forward type recognition And WordScan Plus's ?? features like de-
skew, the Pop-up Proofer. ?? ?? defina?? page set-up and One?? OCRTM
?? its as simple to use as it is accurate, it even integrates ??
with your Windows ?? ?? thanks to ?? Chameleon Tool ??Tm, e-mail and
direct fax Capabilities.
So stop by your local ?? to see for yourself just how quick. ??
and ?? WordScan Plus is.
It could stop mistakes for good.
PCWEEK
THE NEWSPAPER OF CORPORATE COMPUTING ?? JULY 25, 1984 VOLUME 11
NUMBER 29 S3.05
DOJ accord fosters ``too little, too late'' perception
NEWS ANALYSIS ``Chinese wall'' sidestepped, but some
see new opportunities
BY JANE MORRISSEY
The justice Department and the European Commission won ??
concessions from Microsoft Corp., but ?? doubt the consent decree ??
agreed on will ?? much effect on the company or ?? competitors. The
?? got Microsoft to ?? up per-processor ?? and other business ?? ??
and will ?? us compliance for ?? and a half years, bus left ?? its
ability to ?? share opera?? ??lern ?? with its application ??.
The consent decree will be open for public comment within the ??
60 day, after which a federal judge will offer a final ruling. Legal
experts expect the court to uphold the decree.
Although the government could take further action and Microsoft
could face lawsuits from competitors, mo?? observer?? said both are
unlikely because of the time and expense involved. Microsoft com??
ma?? ?? to live with the outcome, ?? mans are not ??.
``Anyone who said this decision went far enough isn't in
touch with the industry.'' said Ed Zander, president of SunSoft
Inc., Sun Microsystems Inc, software unit. ``Of the three or
four issues [the DOJ] could have worked on. they picked the least
contentions. The `Chinese wall'' is more
subtantive.''
But Microsoft officials, citing legal precedents to back them
up. said ?? were able to convince the government that such
exclusionary sharing ?? ?? is managers take sides; desplts?? the
agreement, the government's Anne Bingaman and Microsoft's
Will??am Neukom still don't sea eye-to-eye; Microsoft financ??als,
meanwhile, are strong. in their rights. ``We encourage our
systems people to talk with the apps people about potential new
operating-systems fea??'' said Chairman Bill Gates
Operating-system makers such as IBM. Novell Inc.. Taligent inc..
(?? and Sun Soil said they were were encouraged that Justice took
the actions it did on per-processor licensing practices,
``We're going to jump all over Otis,'' said Lee
Reiswing president of IBM's Personal Sof??are Prod?? division, in
Austin. Texas. ``It means a level playing field for us for the
first time. We have the op?? to hit the OEMs.
``It will help us in the future in not disadvantaging us
with a pricing mechanism,'' said .Joseph (??. chairman anti CEO
of Taligent, in Santa Clara, Calif.
But some said it is too little, too lair''. ``To the
extent [Microsoft's behavior] prevented other operating systems from
succeeding. that war is over,'' said Mitchell ??, chairman of
Powersoft Corp., in Concord, Mass. ``DOS is it and Windows is
it: The ?? has close to zero impact
Novell. one of the insugaors of tile government ??. the decree
is a good first step m addressing its concerns. The Provo. Utah.
firm will discuss at an upcoming board meeting whether to submit
objections or ?? litigation.
``Sure. I am somewhat disappointed.'' said Novell (??
Counsel David Bradford. ``Nevertheless. I understand how the
justice Department and the EC got to where they did .... They did
all in their power, wen the political and legal environment.''
Bradford expects the decree to help Novell fight the
nextgeneration operating-system battle. ``The 32-bit OS market
has not been won by anybody.'' he argued. ``This decree
will al. low for freer competition.''
A major disincentive, to bringing its own charges against
Microsoft is Novell's recent desire to forge a better relationship
with Microsoft. Novell CEO Bob Frankenberg met earlier this month
with (ales to re-establish ties that had broken off under Novell
Chairman Ray Noorda.
``Noorda called us Nazis and. so far. Frankenberg hasn't
engaged in that type of thing.'' (axes said. declining to ??
rate on any new accords. ``We're not going to conduct this
phase in a fishbowl.''
Additional reporting ?? Mary Jo Fol??. Norvin Leach. and Sam W??
OEM licensing practices
?? no per-processor licensing deals
?? no minimum volume commitments required from OEMs
?? no contracts longer than one year: no penalty for non-renewal
?? no restrictions on OEM's licensing or sale of non-Microsoft
operating systems
?? no requirement mat OEMs license DOS to gain a license for
Windows
Non-disclosure agreements
?? duration not to exceed the products release, public
disclosure by Microsoft, . or one year, whichever comes first
?? cannot restrict third parties from developing software that
runs on competing operating systems.
THE CONSENT DECREE DOES NOT ADDRESS:
?? Microsoft benefiting from operating-system knowledge to
develop applications, such as Microsoft applications group getting
advance notice on operating-system advancements, and the use of
undocumented APIs
?? Microsoft acquiring technology from third Datives under guise
of making a deal
BUSINESS
Jesse Berst
Berst
Mode
Behind the smoke, Microsoft wins again
I know you've all heard about the settlement between Microsoft
and the justice Department. But I thought I'd tell ?? some made
information that hasn't made it into the press releases and official
statements.
?? ?? MICROSOFT REALLY DECIDE TO SETTLE? Because the Justice
Department and the European Commission both said they would ??ue
unless Microsoft agreed by July 11.
W?? DID THE JUSTICE DEPARTMENT REALLY DECIDE TO SET- TI.E.?
Because it got to wave the flag and talk in its most grown-up voice
about protecting consumers without the risk of lengthy
litigation--litigation it probably would have lost.
DOES THE AGREEMENT REALLY CHANGE ANYTHING? No, Microsoft has
always let hardware manufacturers make other kinds of deals. But the
price for those deals was so much higher that no one could afford to
use them. Everybody ended up

[[Page 29279]]

making per- processor arrangements whereby they ultimately paid
Microsoft rovalues for every machine shipped. There were always
escape clauses. It's just that nobody could afford to take them. Now
those escape clauses have been codified into the agreement. Because
of the economics, however, few will use them, at least not in the
short term. As for non-disclosure agreements. Microsoft was in the
middle of creating a new standard agreement an??.
How pathetic to see Janet Reno prattling on about ``lower
prices immediately.''
WILL CONSUMERS REALLY SEE LOWER PRICES? How pathetic to see At-
Torne?? General Janet Reno prauling on about lower prices
m??ed??cly. If the decree had come five years ago, when there were
viable MS-DOS clones, it might have had some immediate impact. Now,
in a world where MS-DOS is on the way out and Windows has no real
clones, it will have no short-term effect.
WHAT CHANGES WILL REALLY COME ABOUT BECAUSE OF THE SETTLEMENT?
Very few. It will be slightly easier for computer firms to sell
Net- Ware-rea?? s??ers without incurring financial penalties from
Microsoft. In the long term, it may be slightly easier for a firm to
introduce a new operating system.
WHO'S THE REAL WINNER? Microsoft. It gets two governmental
bodies off its back. And it does so without admitting that it was
wrong, without being forced to divest or break up, and without
paving a cent in fines or restitution.
Best of all, it has the opportunity to restore us mage just when
it needs it most. Microsoft wants to be a dommant plaver in the
c??terpri??e market. To do that, it must convince global
corporations that it is a trustworthy long-term partner. That job
would have been much harder it governments on two contments were
filing lawsuits. The company might as well have changed its slogan
to ``Microsoft--the most antitrusted name in the
business.''
HOW DO MICROSOFT'S COMPETITORS REALLY FEFL ABOUT THE SETTLEMENT?
They feel like schoolboys who complained about a bully stealing
their lunch money and the teacher let the bully keep taking money
for four more years while ``investigating''--and then
?? him off with a token promise to be a good boy from now on. And
the even got to keep the money he had collected.
Still, I think the announcement will ??mately benefit the rest
of the m??ustry. It frees them from their silly fantasy that the
government was going to come riding to their rescue. Now they can
get back to competing on the basis of better products and features,
not better lawyers and lobbvis??s. ??
?? ??SSF BEFST IS DIRECTOR OF THE WINDOWS SOLUTIONS CONFER-
?? EXPOST??. IF YOU WANT ?? TO YOUR ?? CREES. CONTACT ??
(JBERST?? OR ??
PC WEEK JULY 25, 1994
NOVEMBER 7, 1994 PC WEEK NEWS ??
Microsoft's Marvel beta leverages Win 95 desktop ?? ?? ?? AND ??
??
The Microsoft Network. Microsoft Corp.'s new on4ine service. is
taking the first steps toward ??ing the ranks of more established
services such as CompuServe and America Online by tv- ??g itself
into Windows 95's navi- ??tional tools.
Also known by the code name Marvel. Microsoft Network will reach
beta testers in large numbers as part of the sec- and major beta
version or Windows 95. due this week PC Week L??bs took a look at
the on-line service on a late-release candidate of the second beta.
Microsoft Network's on-line services are well-integrated into
the Windows 95 user interface. The content is very sparse at this
??ge. but once populated with ??rmation service providers.
Microsoft Network may prove to be a valuable information source
for Windows 95 users. The information that is available is well-
organized into a hierarchy of folders and icons.
Navigating discussion groups and chat areas was similar to
navigating local titles and folders. Windows front ends to America
Online and CompuServe, in contrast. are separate applications. With
Microsoft Network, we were able to create a link (called a Shortcut)
to a discussion group and place the link on the Windows 95 desktop,
where it appeared like any other folder. When we double-clicked on
the discussion group. Windows 95 automatically re-established our
connection before opening the icon.
Shortcut icons can be embedded as Object Linking and Embedding
2.0 objects, allowing usors to distribute them.
Messaging services are just as well-integrated. We could use the
standard Microsoft Exchange E-mail client included with Windows 95
to compose and send messages. ??
PCWEEK
THE NATIONAL NEWSPAPER OF CORPORATE COMPUTING * NOVEMBER 21,
1994
VOLUME 11, NUMBER 95 ??.95
PC vendors allege undue pressure from Microsoft
?? IBM, OEMs contend strong-arm tactics
BY MANY JO FOLSY AND LISA DICARLO
LAS VEGAS--IBM and other major hardware OEMs are
complaining that Microsoft Corp. is unfairly pressuring PC vendors
to refrain from bundling OS/2 and PC-DOS with their PCs.
Also last week. Microsoft disclosed to hardware OEMs at Comdex
here the Windows 95 MDA (Market Development Agreement), outlining
proposed licensing fees, incentives, and compliance criteria.
Concerning OS/2, the hardware makers claimed that Microsoft
officials threatened to delav, if not withhold entirely, delivery of
Windows 95 code: reduce market-development funds: and withhold sales
and support training for vendors that offer IBM's OS/2 or PC-DOS
preloaded on their systems, sources said.
Sources said IBM and the hardware vendors have held periodic
discussions with the Department of Justice about the alleged unfair
Microsoft practices. IBM, the Justice Department, and the vendors
declined official comment.
``The [Justice Department] has turned into a Better
Business Bureau for anvone who wants to shoot off a complaint
against Microsoft.'' said David Williams, group manager of
Microsoft's Personal Operating Systems Division, in Redmond, Wash.
``We've got some salespeople who sometimes can go too
far.'' Williams said he was unaware of any new filings
regarding Microsoft with the Jusuce Department.
``The playing field is not level.
SEE BUNDLJNC, PAGE 138
Bundling from page ??
and we have a problem with that.'' said an executive with a
hardware maker, who requested anonvmity. Other hardware vendors,
fearful of reprisals from Microsoft, also requested an?? nymity.
One Microsoft customer said further complaints to the Justice
Department against the company would not affect any business
dealings. ``We've been through this DOJ stuff with the ?? IBM??
said Pete Bavoso, vice president of information systems with The
Darby Group Co., a medical supplier and PC Week Corporate Partner in
Westbury, N.Y.
As for the MDA, several hardware makers complained about the
high rovalties that could hike PC prices as well as the stiff
provisions for preloading.
However, they also said the licensing figure is a mere trial
balloon floated by Microsoft, with Windows 95 not scheduled to ship
until mid-1995.
Also at Comdex, several PC vendors claimed to have been
discouraged by Microsoft from demonstrating IBM's OS/2 Warp at the
show. Hewlett-Packard Co. and Packard Bell were among the companies
that decided at the last minute against showing OS/2 as a result of
implied and suggested retaliation from Microsoft, according to
several sources close to the companies.
Officials with HP, of Palo Alto, Calif., and with Packard Bell,
in Chatsworth. Calif., declined to comment. Dell Computer Corp. and
Toshiba America Information Systems Inc. showed OS/2 Warp in their
booths.
``Microsoft has been very aggressive about staving off the
IBM assault,'' said another OEM source. ``There were
indications that the smoothness and flexibility of bundling Windows
95 would have been jeopardized'' if the vendor showed systems
running Os/2 Warp, said the source.
``There's about 15 things in there where you get $3, $2, or
$1 off if you do things like put the Win 95 logo in national
advertising,'' said another OEM.
``There are strong merchandising incentives [in the
MDA],'' said Steve Lair. Toshiba vice president of marketing,
adding that he didn't see anything in the agreement that overtly
demanded exclusivity to Microsoft's products.
In the weeks leading up to Comdex. Microsoft made it clear to
OEMs that it could make the transition to Windows 95 a costly and
bumpy move, according to one of the sources.
Hardware and operating-system vendors complained privately that
despite the proposed justice Department consent decree--which
required Microsoft to alter its OEM licensing and non-disclosure
agreement practices--Microsoft has done little to modify its
behavior.
With the MDA, ``we are not doing per-system incentives for
OEMs. That would be in violation of the consent decree,'' said

[[Page 29280]]

Microsoft's Williams. ``Instead, we're offering incentives for
OEMs who go that extra mile in marketing Windows 95,'' he said,
specifying financial, training, and joint promotional incentives. ??
Additional reporting by Neal Boud??. Dan Farber, and John Dodge
BUSINESS
Jesse Berst
Berst
Mode
Microsoft's on-line rivals could end up in `cyberia''
Microsoft has promised to bundle an on-line service called The
Microsoft Network ?? Windows ?? next summer on tall. If that
occu??s, I pre?? that competing on-line services will be sentenced
to a long, cold w?? of discontent. Microsoft's service will
have an ??beatable edge over Comp??Serve. Prod?? America Online, and
other rivals.
I'm no an export ?? an?? law, so I don't know whether this ??
??. But I do know it feels unfair. It feels like Microsoft is ??ing
a monopoly in one area to gain a monopoly in another, Microsoft may
change its terms and conditions before the final tele??. But as I
understand it right now. OEMs will be ?? to include MSN What's more
?? will not be informed they have a??
Let's ?? XYZ Co. makes a deal to ?? ??pecial Pro?? package w??
?? computer. It even goes to the trouble of ??ing for a Pro?? sign-
on screen to appear the first time the ?? book up.
When XYZ ships its Win 95 PCs, it will have to include The
Microsoft Network sign-on XYZ may not ?? to ?? MSN. It may have
given money ?? consideration to Pro?? in ?? not the bundle. Yet, as
far as I know, XYZ won the able to turn off the built-in MSN screen.
In essence. OEMs will be forced to distribute MSN if they want
to access Windows 95--even it that distribution is to the OEM's
detriment.
I also worry that consumers won't real??e they have options.
It's as if your local phone company were to automatically sign you
up for AT&T's long-distance service without letting you know
that you have other choices. And I worry that Microsoft will use the
MSN ``registration'' procedure to read information about
customers computer configurations and send that information to a
Microsoft da??base. At least one other compa??s (Delrina) has used
on-line registration to scan and store configuration info.
Now, that would be a competitive advantage--if Microsoft
knew the names of millions of Windows users and knew exactly what
hardware and software they owned
GULAG ??IBROGLIO. I have no evidence that Microsoft intends to
secretly capture and store contiguration info. But the fact that I
worry about it points up how Microsoft creates problems for itself.
These fears are feeding the mounting opposition to Microsoft's
Int?? purchase and to The Microsoft Network. The Justice Department
is being press??ed to open another investigation--pressured by
the same competitors that Microsoft cavalierly dismisses is
``wh??ers'' ?? quote a Microsoft exec). Luck??ly for
Microsoft, it has no much money in can alford to waste millions in
legal fees. It looks like it will get a chance to do just that very
soon.
RESPONSE OF THE WEEK: From system Anal??st Jim Ga??nor of
Columbus. O??io:
``The likelihood of a Big Crash on the Internet decreases
??ail??. Links between one portion of the net and another may
tempora??y go down, but the Internet is genes??s was in a Department
of Delense project to create a data network capable of withstanding
a ??clear attack. Truly crashing the Internet for an extended period
would require a bankrolled effort on the level of the most
professional modern terrorism. However, I agree that the tourists
will start leaving. While Mosa?? may be pretty, interaction requires
both action and thought, foreign concepts to the pas??ve??
entertained masses.'' ??
?? JESSE BERST IS THE ?? ?? FOR
OF W?? WA?? ?? ??
C??
?? MCI (JBERST OR ??
??66), Go?? (713372032), ??
FAX ??
Put Your Ideas to Work...
The Easy Way.
With Microsoft Word and Microsoft Excel
Microsoft Word 6
Microsoft Word 6 offers you an easier way to do your day-to-day
work. It has designed to make routine chores go faster and make
complex tasks simpler. Word features Intellisense
technology--build-in intelligence that senses what you want to
do and produces the desired result.
Microsoft Excel 5
Microsoft Excel 5 sets a new standard for spreadsheets. Built-in
intelligence and innovative features let you focus on your analysis,
not your data. And rich custom development tools help you build the
solutions you want. It all adds up to an intelligent spreadsheet
that works the way it should--the way you want.
And best of all, Microsoft Word and Excel are part of the
Microsoft Office programs that work alike and work together, so when
you learn one, you're on your way to learning them all.
Buying Software the Easy Way
Microsoft gives you lots of great ways to buy Word 6 and Excel5.
And Software Spectrum has them all at great prices.
MS Word 6.0 for Windows Upgrade (comp/version) S89*
MS Word 6.0 for Mac Upgrade (version) S89*
MS Word 6.0 for Windows NT Upgrade (comp/version) S125**
MS Excel 5.0 for Windows Upgrade (comp/version) S89*
MS Excel 5.0 for Mac Upgrade (version) S89*
MS Excel 5.0 for Windows NT Upgrade (comp/version) S125**
For over eleven years, Software Spectrum has been providing
superior customer service and value to companies just like yours.
Call Software Spectrum today.
SOFTWARE SPECTRUM 1-800-824-3323 7am to 7pm
(Central) In Canada call 1-800-624-6224
``Software Spectrum price S119, S89 after S30,??in rebate. ??
exp?? 1/31/95 **Upgrade from most mo??or word processors at ?? ??.
Call for det??, Proof of p?? for all upgrades. All prices subject to
change. Other restrictions may apply.
All product names are the property of their respect?? owners.
??1994 Software Spectrum.
Top of the News
Microsoft Settles: Business as Usual
Now that Microsoft's licensing agreements for MS-DOS and Windows
have been deemed ``unfair'' and ``monopolistic''
by the Department of Justice, will other operating systems have a
fighting chance on the desktop?
According to computer manufacturers. industry analysts, and end
users, the outtook is grim for Novell's DOS and IBM's PC-DOS and OS/
2. They say there's not much motivation for PC manufacturers to
preinstall a competing product, since Windows has millions of users
and thousands of software applications. And since Microsoft's
upcoming version of Windows Code-named Chicago) won't require DOS,
the demand for all flavors of DOS is likely to plummet.
Has the train for Chicago already left the station? ``I
think the world of OS/2.'' says Jerry Williams. vice president
of data operations for Eglin Federal Credit Union in Fort Walton
Beach. Florida. ``It's a good operating system. However, I
think the momentum has swung in Windows favor. If you go with OS/2.
you're kind of stepping off the ladder.''
``DOS is starting to go away and Windows is taking over
everything.'' says Gary Shurman, president of the New Orleans
Personal Computer Club. ``Unless somebody comes up with
something earthshattering. I don't think there's a serious
challenger to Microsoft.''
Despite the skepticism. Microsoft's competitors may have their
best chance in years to challenge Bill Gates's desktop domination.
After a lengthy investigation by the U.S. Department of Justice and
the European Commission (the executive body that governs the
European Community), Justice Department officials announced in July
that Microsoft had agreed to end its ``illegal monopolistic
practices'' and stop using ``unfair contracts that choked
off competition and preserved its monopoly'' in the PC
operating system market.
Terms of the Decree
Under the terms of the consent decree, Microsoft must change its
licensing contracts with PC manufacturers (called OEMs). It can no
longer make ``per processor'' agreements that require OEMs
to pay a royalty to Microsoft for each PC shipped--regardless
of whether the preinstalled operating system is from Microsoft or a
competitor. The company also can't require OEMs to purchase a
minimum number of Microsoft operating systems or sign a license with
terms longer than one year (although the OEM can renew the license
for an additional year).
Perhaps most optimistic about the Justice Department ruling is
IBM. Which little success in convin??ing OEMs to preinstall its OS/2
operating system. ``This has really opened the door. We've ??ut
proactively, contacting, hundreds of PC manufacturers
already.'' says John ?? detector of IBM's Personal Software
Products division in Austin. Texas. While So??ing expects some
``major North American manufacturers'' to pre??stall OS/
2?? so far Big Blue's ??tories have been in Europe. Soyring says
that

[[Page 29281]]

German PC makers Vobis and Escom already preinstall OS/2--and
Escom expects to ship 440.000 ??tems with OS/2 over the next 12
months.
Despite the ruling from Justice, Micro soft's influence over PC
manufacturers remains immense. Most of the leading OEMs contacted
for this article had little or no comment on the Justice Department
ruling, other than to say that their relationship with Microsoft
would stay the same (in other words, they'd still preinstall
Windows). And many industry pundits sec the consent decree as a weak
slap on Microsoft's wrist. ``I think Microsoft is thrilled with
the settlement.'' says Tim Bajarin. president of Creative
Strategies in San Jose, California. Of course, if Microsoft is too
aggressive, it is likely to find itself in the sights of regulators
once again. That's a position even Bill Gates wants to avoid.
Jeff Bertolucci
Jan ??rancisco ??ronicle
THE ?? DAILY CIRCULATION IN HO??THE?? CA??FORMIA
MONDAY, JULY 18, 1994
Microsoft Unscathed By Settlement
Antitrust pact a slap on wrist for software giant
By Da??id E??ste??
Ch?? Staff Writer
Although the government claimed victory in its antitrust battle
against Microsoft, it appears as if the world's largest software
maker suffered little damage and in fact should continue to
steamroller the rest of the industry.
By agreeing to halt some supposedly monopolistic practices, Bill
Gates'' giant company has left the door open ever so
&lightly for competitors to grab some piece of the market for
operating systems that run moat of today's personal computers. It is
a market Microsoft dominates with its MS-DOS and Windows programs,
currently installed on more than 120 million computers worldwide.
But sometime late this year of early next. Microsoft intends to
brush away its rivals once again when it introduces Chicago, the
next generation of Windows. If PC users flock to Chicago as
expected.
Gates actually could increase hit hold on the industry he helped
create in the early 1990s.
There had been speculation that the Justice Department, which
took over the Investigation from the Federal Trade Commission last
year, might have gone so far as to break up Microsoft Just as
AT&T was split up In 1984.
But ns the government closed the case late Friday, however, it
was with a mere slap on the wrist. Microsoft admitted no guilt over
allegations of monopolistic practices, and faces no fines or
financial penalties. Its revenues, now over $4 billion a year,
probably will not suffer.
No wonder Microsoft officials were happy with the terms of the
settlement. ``It preserves our ability to do business In a way
that IS effective,'' said Bill Neukom, vice president of law
and corporate affairs.
But Attorney General Janet Reno professed satisfaction with the
outcome of the first major antitrust case of the Clinton
administration, saying the settlement ``levels the playing
field and opens the door for competition'' by curbing
Microsoft's ``monopollstic practices.''
Reno talked tough, adding that ``while the company fairly
and lawfully climbed to the top of the industry ladder, It used
unfair and illegal practices to maintain its dominant
position.''
But the settlement did not address what many competing companics
consider the real antitrust issue. Microsoft, they say, has used its
control of DOS and Windows to extend lb hold on the software sector.
In fact. during the nearly four years the government
Investigated Microsoft, the Redmond, Washbased behemoth managed to
be the major player In the market not only for operating systems,
but also for major applications such as word processing and
spreadsheet software. And even as the consent decree goes Into
effect, Microsoft is trying to tighten the screws on its major
competitors by asking smaller software developers to adopt a
standard that would make their programs dovetail with Microsoft's
best-selling ``Office'' suite of applications. U??
Friday's consent decree, which steers the company and the
government clear of the courts, includes an agreement to change the
way Microsoft licenses its operating system. That Issue the
government felt was its best chance to beat Microsoft had the
antitrust case gone to court.
Microsoft no longer will offer PC makers steep discounts on
volume purchases of DOS and Windows In return for royalities from
every PC said--whether or not the Microsoft system was actually
Installed on them. That ``per-processor'' licensing
strategy had discouraged manufacturers from buying rival products
such a Novell's version of DOS or IBM's OS/2 operating system, since
they already were paying for Microsoft's version.
Novell's general counsel, David Bradford, saw the consent decree
as a clear victory. ``This has been a long effort by many
companies for many years.'' he said, ``and this decision
will provide consumers with Increased choices and more innovative
products.''
But the euphoria may wear off quickly. Microsoft's Chicago
program reportedly will not require an underlying operating system,
leading Industry experts to predict the irrelevancy or death of DOS
once Chicago catches on. Industry Standard P??ts
Competitors may benefit more from Microsoft's agreement
to'' loosen restrictions on its nondisclosure
agreements--Industry. standard pacts that software companics
must sign to get advance copies of new products such as Chicago.
Microsoft In the past has forced companies to agree not to work with
other operating systems in return for access to lb programs.
That may help large companies like Novell, which is updating its
popular WordPerfect and Quattro Pro programs. But smaller companies
still may find themselves ti??t- ing at windmills in trying to take
on Microsoft.
Ernie Simpson. president of The Wizard Co. in Denton, Texas,
called the settlement ``a waste of time.''
``Microsoft will continue to do as they have been doing,
only they'll word their contracts a little differently.'' said
Simpson, whose company develops software for some major Windows
programs. ``Microsoft Is the de facto industry standard for
operating systems, and they will continue to control the industry to
the advantage of Microsoft and the detriment of everyone
else.''
Microsoft had Insisted It would never settle antitrust charges
out of court. Gates was positively adamant about it, complaining
that the Justice Department was hounding him unreasonably, in the
end, however, with antitrust charges looking more and more possible,
the company decided to cut a deal. Judging from the first reviews,
Gates appears to have done quite well by It.
Chronicle wiry ser??c?? to this report.
Sunday, July 24, 1994 C-5
San Francisco Examiner
COMPUTERS & TECHNOLOGY
Microsoft deal: too little, too late
A few days after the Department of Justice announced the
settlement of its antitrust investigation of Microsoft, Bill Gates
told the Wall Street Journal, ``I intend to defy
gravity.''
Thanks to the nature of that settlement, it is likely that he
will.
The Justice Department press release announcing the settlement
quoted Attorney General Reno as saying, ``Microsoft's unfair
contracting policies have denied other U.S. companies a fair chance
to compete, deprived cop- ??mers of an effective choice ??ong
competing PC operating systems, and alowed innovation.'' True
enough.
She went on to state, ``Today's settlement levels the
playing field and opens the door for competition.''
Unfortunately, it is unlikely to do either.
It is telling that in describing the harm caused to competition
and innovation by Microsoft's practices, the attorney general used
the past tense. The particular practices the settlement addressed
were unquestionably key factors in Microsoft's rise to dominance in
the 1980s.
Among other things, Microsoft required PC manufacturers to pay a
license fee for its MS-DOS and Windows operating system software on
every PC shipped with an Intel microprocessor under long-term
agreements--whether or not those PCs actually contained that
software--and unreasonably restricted independent software
companies from working with Microsoft compatitors. In so doing,
Microsoft managed to insinuate its technology into the heart and
soul of 85 percent of the world's PCs.
By 1985, these practices had already had their intended effect:
making Microsoft's operating system the de fact?? PC standard. The
present source of Microsoft's domination in the PC world derives
from the status as the standard-holders, not the practices the
Justice Department condemned and which will now be prohibited under
the settlement.
Microsoft, understands this perfectly well, which, of course, is
why Bill Gates let the settlement happen. Nothing in the proposed
settlement is likely to have anything other than the most marginal
effect on Microsoft's future.
Inherent in the nature of software technology is the concept of
dependence.

[[Page 29282]]

Operating systems are useless without application programs and vice
versa. Neither has discrete, stand-alone value.
But of the two, operating systems software must come first and
clearly provides the most potential for leverage. To its credit,
Microsoft understood this earlier than everyone else and exploited
its insight relentlessly. So technically dependent in the PC
industry on Microsoft operating system software, that Microsoft
could afford the luxury of a five-year period in which to perfect.
Windows after its initial introduction in 1986.
When Apple introduced the Macintosh ``graphical user
interface,'' which replaced ob??cure and hard-to-remember
keyboard commands with easy-to-learn and easy-to-use screen icons
and a mouse, it marked a watershed in the development of consumer-
friendly computing. In response, Microsoft introduced Windows, which
was supposed to provide Macintosh-like ease-of-use.
[See VIEWPOINT, C-6]
??VIEWPOINT from C-1
Too little, too late
But the first several versions of Windows were so poorly
designed that very few people wanted them, preferring even the
archaic DOS with its incredibly difficult keyboard commands. It
wasn't until 1990, five years after its introduction, that Microsoft
finally produced a version of Windows that was ready for prime.
Now, one would think that if genuine competition existed in PC
operating systems, this five-year gap would have been exploited by
one or more competitors of Microsoft. Indeed, it's hard to conceive
that any company could have taken as long as Microsoft did to get a
basic technology right and still survive.
Yet, Microsoft not only survived during this period, it
prospered. The reason is that it was virtually impossible to shake
free of MS-DOS, even when clearly better alternatives were
available. The consumer investment in application programs that
could only run on the Microsoft system was too large and the cost of
switching to an alternative technology--even a clearly better
one--too great.
While this was obvious to everyone by 1985 or 1986, Bill Gates
understood it in 1980.
Almost 10 years later, PC manufacturers, consumers and software
developers are even more tightly bound to Microsoft operatins system
technologies. The ties that bind are not contractual, they are
technical, which is why the Jus- rice Department settlement will be
ineffective.
And while controlling this standard, Microsoft is free to
compete on applications based on the standard. Companies that
develop competing spreadsheet, word processing and other such
programs have complained for years that Microsoft programmers have
the unfair advantage of knowing changes to the operating system
specifications well before anyone else.
The fact is Microsoft owns-- and closely guards--the
de facto standard for desktop computers, a critical part of our
information infrastructure. And at least three steps could be taken
to ensure fair competition. Microsoft could be required to:
??Publicly disclose its operating system interface
specifications so that designers of competing operating systems
could have assurance that application programs written for MS-DOS or
Windows would run efficiently with their operating systems.
Microsoft should update its specifications in a periodic and timely
manner.
San Jose Mercury News
Se??ng Northern Cal?? Since 1851
?? NOVEMBER 13, 1994
Microsoft, Intel set to define technology
?? Duopoly: Apple, IBM, Motorola mounting last-ditch attempt to
make PC alternative. BY RORY J. O'CONNOR
Merr??ry News Staff Writer
Tomorrow, when Silicon Valley's brain trust arrives in Las Vegas
as part of a 200,000-strong crowd at the computer industry's largest
trade snow, conversation will almost certainly center on one topic:
Can anything stop Microsoft and Intel from controlling everything?
Some fear that as the digital future of the information
superhighway emerges, an unchallenged Microsoft and Intel will wind
up in total, undisputed control of the technology upon which the
country's citizens and economy will depend. And few believe that a
recently announced alliance between Apple and IBM will prove an
effective roadblock.
Who will control COMPUTING'S FUTURE? First in an occasional
series.
Today, Microsoft Corp. makes the world's most popular software
for personal computers, operating systems that control 85 percent of
the machines in use. Intel Corp.'s microprocessor chips are the
brains in 75 percent of all the computers made.
But the personal computer is rapidly becoming a home appliance,
and the PC is poised to expand from word processing and spreadsheets
to controlling a myriad of other jobs in our everyday work and
personal lives. The companies that control personal- computer
technology are in a position soon to dominate much, much more.
From video telephones to intelligent fax machines, from office
to home, from providing digital information and entertainment to
managing credit-card and other financial transactions, Microsoft and
Intel are already extending their reach far beyond traditional
personal computing.
Both companies have deep pockets to back the technology--
and their unofficial partnership is an effective duopoly that could
let the companies dictate the price of technology, minimize consumer
choices and slow the pace of technical progress.
In short, many believe, little stands between the two comparoes
and technical control of the future.
``Increasingly, I'm believing it's all over, and we're
going to be locked into Microsoft and Intel forever.'' said
Dataquest analyst Kimball Brown.
In the 13 years since IBM transformed the PC from hobbyist toy
to business tool. control of the industry has shifted from IBM and
Apple to their once-tiny corn- petitors. Now, Apple and IBM, despite
their combined annual revenues of nearly $75 billion, are the
underdogs.
Except for Apple, whose research and development spending
remains large despite a $100 million cutback in the past year, few
PC companies invest significant sums in new technology research. The
bulk of such money is spent by Intel to develop chips and Microsoft
to further its lead in software.
Many people in the industry decry this state of affairs, but
lack the money, the marketing or the technology to force meaningful
competition. Even the federal government has declined to step in,
punishing Microsoft with a slap on the wrist after a four-year
investigation into what Attorney General Janet Reno called
``illegal, monopolistic'' practices.
Perhaps the only force large enough to change anything is an
infant agreement announced last week by Apple, IBM and Motorola to
build a new kind of personal computer, one that would neither use
Intel microprocessors nor fea ture Microsoft operating systems.
The timing of their agreement, one week before the largest
annual gathering of technology power brokers in the world, is no
accident.
Even though the alliance will not produce a product until 1996,
IBM and Apple need every ounce of momentum they can muster for what
is probably the last- ditch attempt to topple Intel and Microsoft t
or even to hope to play a role in defining the technical future.
But most analysts insist that Apple and IBM are waging the wrong
war. ``The desktop operating system war is over,'' said
venture capitalist Ann Winblad, whose Emeryville firm specializes in
software companies. ``Microsoft has won.''
Instead. Apple and IBM should be looking to the information
superhighway for opportunities to sell new technology, expand their
business and regain the power to force technical competition, said
Richard Shaffer, publisher of the Technologic Letter in New York.
That's because there is a potentially more lucrative market in
the future, one that uses both the personal computer and its
technology.
It goes by the catch-all term of information superhighway, but
it encompasses a host of major changes in the role of personal
computers at work and at home.
Some of the latest home, computers are already touted as being
able to replace nearly everything in a small office except the
coffee pot.
Phones and fax machines are becoming smarter, thanks to more-
powerful computer brains. And when people are away from their home
or office in the future, they may well carry portable devices that
combine today's cellular phone with ready information access,
offering yet another umbilical cord to the PC.
Over the next decade, even television is poised to become
interactive, offering far more choices, two-way video and fountains
of information on demand--activities that require heavy use of
computer chips and sophisticated operating systems and other
software. Computers will manage nearly all financial transactions,
and will even be a citizen's primary conduit to the government.
Some experts envision a single intelligent box in the home, one
that would use the

[[Page 29283]]

functions of a personal computer to connect the home to information
and communications lines through phone-company wires or cable-
television hookups.
There's little doubt that each of these areas will be the site
of intense competition. In almost every case, Microsoft and Intel's
dominance of the PC business would give them a crucial advantage.
If they succeed in controlling key technology in any or all of
these areas, they will be able to determine much of how the devices
work, and could even control how people receive information or make
purchases. And the closer the digital world moves to merging control
into just one or two boxes connected to monolithic networks, the
better the chance Intel and Microsoft have to dominate them as they
have PCs.
But Apple insists it is not blind to the digital future, despite
initial failure in one new market-- that for personal digital
assisrants--and a very slow start for its E-World on-line
service.
``Clearly, there's a feeling at Apple that these other
technologies are very exciting areas,'' said Rick LeFaivre, the
head of the company's Advanced Technology-Group. ``But at the
same time we're making sure not to take our eye off the PC and say
it's dead. ... The PC side of our business will be by far the
dominant side for a long, long time.''
At the same time, Apple's partner is struggling to regain power
it has lost in nearly every area of its business. Internally, it is
replacing top managers, revamping its structure, changing key
technology, laying off workers and trying to figure out how it fits
into a world it once controlled. Externally, critics say they can't
fathom the company's strategy, especially in personal computing,
where it is unclear what software and hardware technology IBM
considers strategic--and, there, fore, safe for customers to
buy.
The problem for Apple and IBM, according to analysts, is that
they probably have little'' hope of competing effectively in
the digital future unless they can quickly establish their new
computer as a viable mauve.
But to become a PC alternative. the companies must overcome a
host of difficulties, from wrenching changes in their corporate
cultures to damaged balance sheets to the improbability of the
partnership they began with Motorola more than three years ago.
``The whole plan in 1991 was daring, kind of like
chemotherapy,'' stud Shaffer. ``The therapy might kill the
patient, but the al ternative is certain death.''
Few believe that Apple, IBM and Motorola can thrive against the
Microsoft-Intel duopoly short of a move even more unlikely than the
original IBM-Apple partnership.
``Without the merger of Apple and IBM into one corporate
entity, they are executing separate strategies, no matter what they
say,'' Winblad said. ``So while some people have called
this the David and Goliath story, with Microsoft as Goliath, there
is no David--perhaps a Tom, Dick and Harry.''
Not everyone believes that a world where two companies control
most of the technology is a cause for alarm, however. ``What's
wrong with there being just one operating system? It's supposed to
be transparent to the user,'' said analyst Doug Kass of the
Viewpoint Group in Aptos. ``I don't think that will lead to
huge increases in price. It's not competition among vendors, but
what the market will bear in terms of price. Consumers look for what
works, not the cutting edge. If some new (software) is priced beyond
the glass ceiling of what consumers are comfortable paying, it won't
sell.''
Not surprisingly, Microsoft officials share that view
``Things are very competitive now,'' said Brad Chase,
general manager of Microsoft's personal operating systems division.
``Apple is certainly not an uncompetirive company. IBM is a
very aggressive company. And the thing about technology is you can't
rest on your laurels. If you don't keep aggressive, your leadership
will melt like butter.''
Tomorrow in Business Monday: How far can Microsoft go?
San ``Jose Mercury News,'' Wednesday, December 21,
1994
MICROSOFT'S DOMINATION
Microsoft's revenues in the world market for personal computer
business grew more in 1994 than revenues in the market as a whole,
according to preliminary estimates by Dataquest Inc. Total revenues
grow by more than $550 million, while Microsoft's related revenue
grew by more than$650 million. ``Lotus 1-2-3.
WordPerfect. dBase.
Paradox and Harvard
Graphics once dominated their respective categories.'' said
Dataquest analyst Karl Wong. ``Today, Microsoft products have
replaced each of these one-time product category leaders.''
(Figures are in millions.)
``94 1993 ``93-'94 ``94 ``94 market
Rink Company Revenue % chg. ,Revenue, share (%)
1 Microsoft $2.221 +29.4 $2.873
34.7 District of Columbia.
2 Lotus 986--1.8 968 1
I. 7 3 Novell 698 -11.6 617
7.5 4 Adobe 197 +28.1 253 3.1 5 Symantec 207 +15.2 238
2,9 6 Clans 160 -t-9.3 175
2.1 7 Borland 360 -52.8 170
2.1 8 Intuit 104 +56.9 163
2.0 9 Corel 105 +41.6 148
t.8 I0 Delrina 65 +43.1 94
1.1 Others 2.617 -1.7 2.573
31.0 Total 7.720 +7.2 8.272
100.0
Source: Dataquest Inc
MERCURY NEWS
Positive Feedbacks
in the Economy
A new economic theory elucidates mechanisms whereby small chance
events early in the history of an industry or technology can tilt
the competitive balance Conventional economic theory is built on the
assumption of diminishing returns. Economic actions engender a
negative feedback that leads to a predictable equilibrium for prices
and market shares. Such feedback tends to stabilize the economy
because any major changes will be offset by the very reactions they
generate. The high oil prices of the 1970's encouraged energy
conservation and Increased off exploration, precipitating a
predictable drop In prices by the early 1980's. According to
conventional theory, the equilibrium marks the ``best''
outcome possible under the circumstances: the most efficient use and
allocation of resources.
Such an agreeable picture often does violence to reality. In
many parts of the economy, stabilizing forces appear not to operate.
Instead postitive feedback magnifies the effects of small economic
shifts; the economic models that describe such effects differ vastly
from the conventional ones. Diminishing returns imply a single
equilibrium point for the economy, but positive
feedback--increasing rerums--makes for many possible
equilibrium points. There is no guarantee that the particular
economic outcome selected from among the many alter-
W. BRIAN ARTHUR is Morrison Professor of Population Studies and
Economics at Stanford University. He obtained his Ph.D ??om the
University of California. Berkeley, In 1973 and holds graduate
degrees In operations research, economics and mathematics. Until
recently Arthur was on leave at the Santa Fe Institute, a research
institute dedicated to the study of complex systems. There he
directed a team of economists, physicists, biologists and others
investigating behavior of the economy as an evolving. complex
system. by W. Brian Arthur natives will be the ``best''
one. Furthermore. once random economic events select a particular
path, the choice may become locked-in regardless of the advantages
of the alternatives. If one product or nation in a competitive
marketplace gets ahead by ``chance.'' it tends to stay
ahead and even Increase its lead. Predictable, shared markets are no
longer guaranteed.
During the past few years I and other economic theorists at
Stanford University, the Santa Fe Institute in New Mexico and
elsewhere have been developing a view of the economy based on
positive feedback. Increasing-returns economics has roots that go
back 70 years or mote. but Its application to the economy as a whole
is largely new. The theory has strong parallels with modem nonlinear
physics (instead of the pre-20th-century physical models that
underlie conventional economics), it recluses new and challenging
mathematical techniques and it appears to be the appropriate theory
for understanding modem high-technology economies.
The history of the videocassette recorder furnishes a simple
example of positive feedback. The VCR market started out with two
competing formats selling at about the same price: VHS and Beta.
Each format could realize Increasing returns as its market share
Increased: large numbers of VHS recorders would encourage video
outlets to stock more prerecorded tapes in VIIS format, thereby
enhancing the value of owning a VIIS recorder and leading mote
people to buy one. (The same would, of course, be true for Beta-
format players.) In this way, a small gain in market share would
Improve the competitive position of one system and help it further
increase its lead.
Such a market is initially unstable. Both systems were
introduced at about the same time and so began with roughly equal
market

[[Page 29284]]

shares; those shares fluctuated early on because of external
circumstance, ``luck'' and corporate maneuvering.
Increasing returns on early gains eventually tilted the competition
toward VHS: It accumulated enough of an advantage to take virtually
the entire VCR market.
Yet it would have been Impossible at the outset of the
competition to say which system would win, which of the two
possible'' equilibria would be seleered. Furthermore, if the
claim that Beta was technically superior Is true, then the market's
choice did not represent the best economic outcome.
Conventional economic theory offers a different view of
competition between two technologies or products performing the same
function. An example is the competition between water and coal to
generate electricity. As hydroelectric plants take more of the
market, engineers must exploit more costly dam sites, thereby
increasing the chance that a coal-fired plant will be cheaper. As
coal plants take more of the market. they bid up the price of coal
(or trigger the imposition of costly pollution controls) and so tip
the balance toward hydropower. The two technologies end up sharing
the market in a predictable proportion that best exploits the
potentials of each, in contrast to what happened to the two video-
recorder systems.
The evolution of the VCR market would not have surprised the
great Victorian economist Alfred Marshall one of the founders of
today's conventional economics. In his 1890 Principles of Economics,
he noted that if firms'' production costs fall as their market
shares Increase, a rum that simply by good fortune gained a high
proportion of the market early on would be able to best its rivals;
``what. ever firm first gets a good start'' would corner
the market. Marshall did not follow up this observation, however.
and theore??cal economics has until recently largely ignored it.
Marshall did not believe that mcreasing returns applied
everywhere; agriculture and mining--the marestays of the
economies of his time-- were subject to diminishing returns
caused by limited amounts of fer??le land or high-quality ore
deposits. Manufacturing, on the other hand. enjoyed increasing
returns because large plants allowed improved organization. Modern
economists do not see economies of scale as a reliable source of
increasing returns. Sometimes large plants have proved more
economical; often they have not.
I would update Marshall's insight by observing that the parts of
the economy that are resource-based (agriculture. bulk-goods
production, mining) are still for the most part subject to
diminishing returns. Here conventional economics rightly holds sway.
The parts of the economy that are knowledge-based, on the other
hand. are largely subject to increasing returns. Products such as
computers, pharmaceuticals, missiles, aircraft, automobiles.
software, telecommunications equipment or fiber optics are
complicated to design and to manufacture. They require large initial
investments in research, development and tooling. but once sales
begin, incremental production is relatively cheap. A new airframe or
aircraft engine, for example. typically costs between $2 and $3
billion to design, develop, certify and put into production. Each
copy thereafter costs perhaps $50 to $100 million. As more units are
built, unit costs continue to fall and profits increase.
Increased production brings additional benefits: producing more
units means gaming more experience in the manufacturing process and
achieving greater understanding of how to produce additional units
even more cheaply. Moreover. expenence gained with one product or
technology can make it easier to produce new products incorporating
similar or related technologies. Japan-for example, leveraged an
initial investment in building precision instruments into a capacity
for building consumer electronics products and then the integrated
circuits that went into them.
Not only do the costs of producing high-technology products fall
as a company makes more of them. but the benefits of using them
increase.
Many items such as computers or telecommunications equipment
work m networks that require compatibility; when one brand gains a
significant market share, people have a strong incentive to buy more
of the same product so as to be able to exchange information with
those using it already.
If increasing returns are important. why were they largely
Ignored until recently? Some would say that complicated products-
high technology-for which increasing returns are so Important. are
themselves a recent phenomenon. This is true but is only part of the
answer. After all, in the 1940's and 1950's. economists such as
Gunnar K. Myrdal and Nicholas Kaldor identified positive-feedback
mechanisms that did not revolve technology. Orthodox economists
avoided increasing returns for deeper reasons.
Some economists found the existence of more than one solution to
the same problem distasteful--unscientific. ``Multiple
equilibria.'' wrote Joseph A. 5chumpeter in 1954, ``are
not necessarily useless, but from the standpoint of any exact
science the existence of a uniquely determined equilibrium is, of
course, of the utmost Importance. even if proof has to be purchased
at the price of very restrictive assumptions; without any
possibility of proving the existence or [a] uniquely determined
equilibrium-- or at all events, of a small number of possible
equilibria--at however high a level of abstraction, a field of
phenomena is really a chaos that is not under analytical
control.''
Other economists could see that
ALL A
RANDOM WALK on a convex surface illustrates increasing-returns
competition between two technologies. Chance determines early
patterns of adoption and so influences how fast each competitor
improves. As one technology gains mort adherents (corresponding to
motion downhill toward either edge of the surface), further adoption
is increasingly likely.
SCIENTIFIC AMERICAN February 1990 93
FLORENCE CATHEDRAL CLOCK has hands that move
``counterclockwise'' around its 24.hour dial When Paolo
Uccello designed the clock in 1443. a convention for clockfaces had
not emerged. Competing designs were subject to increasing returns:
the more clockfaces of one kind were built, the more people became
used to reading them. Hence. it was more likely that future
clockfaces would be of the same kind. After 1530.
``clockwise'' designs displaying only 12 hours had crowded
out other designs. The author argues that chance events coupled with
positive feedback. rather than technological superiority, will often
determine economic developments. theories incorporating increasing
returns would destroy theft familiar world of unique, predictable
equilibria and the notion that the market's choice was always best.
Moreover, if one or a few firms came to dominate a market, the
assumption that no firm is large enough to affect market prices on
its own (which makes economic problems easy to analyze) would also
collapse. When John R. Hicks surveyed these possibilities in 1939 he
drew back in alarm. ``The threatened wreck. age.'' he
wrote. ``is that of the greater part of economic theory.''
Economists restricted themselves to diminishing returns, which
presented no anomalies and could be analyzed completely.
Still others were perplexed by the question of how a market
could select one among several possible solutions. In Marshall's
example, the firm that is the largest at the outset has the lowest
production costs and must inevitably win in the market. In that
case. why would smaller firms compete at all? On the other hand. if
by some chance a market started with several identical firms, their
market shares would remain poised in an unstable equilibrium
forever.
Studying such problems in 1979, I believed I could see a way out
of many of these difficulties. In the real world, if several
similar-size firms entered a market at the same time, small
fortuitous events--unexpected orders, chance meetings with
buyers, managerial whims--would help determine which ones
achieved early sales and, over time, which firm dominated. Economic
activity is quantized by individual transactions that are too small
to observe, and these small ``random'' events can
accumulate and become magnified by positive feedbacks so as to
determine the eventual outcome. These facts suggested that
situations dominated by increasing returns should be modeled not as
static, deterministic problems 94 SCIENTIFIC AMERICAN February 1990
as dynamic processes based on ??ndom events and natural positive
feedbacks, or nonlinear??ties
With this strategy an increasingreturns market could be re-
created in a theoretical model and watched as its corresponding
process unfolded again and again. Sometimes one solution would
emerge, sometimes (under identical conditions) another. It would be
impossible to know in advance which of the many solutions would
emerge in any given run. Still, it would be possible to record the
particular set of random events leading to each solution and to
study the probability that a particular solution would emerge under
a certain set of initial conditions. The idea was simple, and it may

[[Page 29285]]

well have occurred to economists m the past. But making it work
called for nonlinear random-process theory that cud not exist in
their day.
Every increasing-returns problem need not be studied in
isolation; many rum out to fit a general nonlinear probability
schema. It can be pictured by imagining a table to which balls are
added one at a time; they can be of several possible
colors--white, red, green or blue. The color of the ball to be
added next is unknown, but the probability of a given color depends
on the current proportions of colors on the table. If an increasing
proportion of balls of a given color increases the probability of
adding another ball of the same color, the system can demonstrate
positive feedback. The question is. Given the function that maps
current proportions to probabilities, what will be the proportions
of each color on the table after many balls have been added?
In 1931 the mathematician George Polya solved a very particular
version of this problem in which the probability of adding a color
always equaled its current proportion. Three U.S. probability
theorists, Bruce M. Hill of the University of Michigan at Ann Arbor
and David A. Lane and William D. Sudderth of the University of
Minnesota at Minneapolis. solved a more general nonlinear version in
]980. In 1983 two Soviet probability theorists. Yuri M. Ermoliev and
Yuri M. Kaniovski, both of the Glushkov Institute of Cybernetics in
Kiev, and I found the solution to a very general version. As balls
continue to be added, we proved, the proportions of each color must
settle down to a ``fixed point'' of the probability
function--a set of values where the probability of adding each
color is equal to the proportion of that color on the table.
Increasing returns allow several such sets of fixed points.
This means that we can detern the possible patterns or soluti??
of an increasing-returns problem by solving the much easier
challenge of finding the sets of fixed points of its probability
function. With such tools economists can now define increasing-
returns problems precisely, identify their possible solutions and
study the process by which a solution is reached. Increasing returns
are no longer ``a chaos that is not under arialytical
control.''
In the real world, the balls might be represented by companies
and their colors by the regions where they decide to settle. Suppose
that firms enter an industry one by one and choose their locations
so as to maximize profit. The geographic preference of each firm
(the intrinsic benefits it gains from being in a particular region)
vanes; chance determines the preference of the next firm to enter
the industry. Also suppose, however, that firms'' profits
increase if they are near other firms (their suppliers or
customers). The first firm to enter the industry picks a location
based purely on geographic preference. The second firm decides based
on preference modified by the benefits gained by locating near the
first firm. The third firm is influenced by the positions of the
first two firms, and so on. If some location by good fortune
attracts more firms than the others in the early stages of this
evolution, the probability that it will attract more firms
increases. Industrial concentration becomes self-reinforcing.
The random historical sequence of firms entering the industry
deter. mines which pattern of regional setdement results, but the
theory shows that not all patterns ate possible. If the
attractiveness exerted by the presence of other firms always rises
as more firms are added, some region will always dominate and shut
out all others.
If the attractiveness levels off, other solutions, in which
regions share the industry, become possible. Out new tools tell us
which types of solutions can occur under which conditions.
Do some regions in fact amass a large proportion of an industry
because of historical chance rather than geographic superiority?
Santa Clara County in California (Silicon Valley) is a Likely
example. In the 1940's and early 1950's certain key people m the
U.S. electronics industry--the Varian brothers, William Hewlett
and David Packard. William Shockley-set up shop near Stanford
University; the local availability of engineers, supplies and
components that these early firms helped to create made Santa Clara
County extremely attractive to the or so firms that followed. If
these eatly entrepreneurs had preferred other places, the densest
concentration of electronics in the country might well be somewhere
else.
On a grander scale, if small events m history had been
different, would the location of cities themselves be different? 1
believe the answer ]s yes. To the degree that certain locations are
natural harbors or junction points on rivers or lakes, the pattern
of Ones today reflects not chance but geography. To the degree that
industry and people are attracted to places where such resources are
already gathered, small early chance concentrations may have been
the seeds of today's configuration of urban centers. ``Chance
and necessity,'' to use Jacques Monod's phras??ract. Both have
played crucial ro?? the development of urban centers ?? the U.S. and
elsewhere.
Self-reinforcing mechanisms other than these regional ones work
in international high-tech man. ufacturing and trade. Countries that
gain high volume and experience in a high-technology industry can
reap advantages of lower cost and higher quality that may make it
possible for them to shut out other countries.
For example, in the early 1970's, Japanese automobile makers
began to sell significant numbers of small cats in the U.S. As Japan
gained market vol. tune without much opposition from Detroit. its
engineers and production workers gained experience, its costs fell
and its products improved. These factors, together with improved
sales networks, allowed Japan to increase FERROMAGNETS AND REGIONAL
RAIL GAUGES become ordered in much the way. As a disordered magnetic
material is cooled (left% the atomic dipoles inside it exert tortes
on one another, causing neighboring dipoles to align. Eventually all
the dipoles in a sample line up, but the direction they all take (up
or down) cannot be predicted beforehand. Similarly, as Douglas
Puffert of Swarthmore College has shown, neighboring private
railroads (right) ha the past century adopted the same gauge to
extend their range mole easily. Eventually all (or most) railroads
used the Same gauge. Similar equations describe the behavior of
these two system.5. SCIENTIFIC AMERICAN February 1990 95 its share
of the U.S. ?? as a resuit. workers gained ?? expert. ence, costs
fell further and quality improved again. Before Detroit responded
seriously, this posture-feedback loop had helped Japanese companies
to make serious inroads into the U.S. market for small cars. Similar
sequences of events have taken place m the markets for television
sets. rotegrated circuits and other products.
How should countries respond to a world economy where such rules
apply? Conventional recommendations for trade policy based on
constant or diminishing returns tend toward lowprofile approaches.
They rely on the open market, discourage monopolies and leave issues
such as R&D spending to companies. Their underlying assumption
is that there is a fixed world price at which producers load goods
onto the market, and so interference with local costs and prices
means of subsidies or tariffs is unp?? ductive. These policies are
appropriate for the diminishing-returns parts of the economy, not
for the technology-based parts where increasing rerums dominate.
Policies that are appropriate to success in high-tech production
and international trade would encourage industries to be aggressive
in seeking out product and process improvements. They would
strengthen the national research base on which hightech advantages
are built. They would encourage rums in a single industry to pool
their resources in joint ventures that share up-front costs,
marketing networks, technical knowledge and standards. They might
even foster strategic affiances, enabling comparues in several
countries to enter a complex industry that none could NONLINEAR
PROBABILITY THEORY can predict the behavior of systems subject to
Increasing terns, in this model, balls of different colors are added
to a table; the probability that the next ball will have specific
color depends on the current proportions of colors (top). increasing
term-ns occur in A (the graph shows the two-color case: arrows
indicate likely directions of motion): a red ball is mote likely to
be added when there is already a hi fib proportion of'' red
halls. This case has two equilibrium points: one at which almost all
balls are red; the other at which very few are red. Diminishing
terns occur in B: a higher proportion or red balls lowers the
probability of adding another. There is a single equilibrium point.
A corn. bination of increasing and diminishing returns (C) yields
many equilibrium points. 98 SCIENTIFIC AMERICAN February 1990 tackle
alone. Increasing-returns theory also points to the Importance of
timing when undertaking research initiatives in new industries.
There is little sense in entering a market that is already close to
being locked-in or that otherwise offers Little chance of success.
Such policies are slowly being advocated and adopted in the U.S.
The value of other policies, such as subsidizing and protecting
new industries--bioengineering, for example-to capture

[[Page 29286]]

foreign markets, Is debatable. Dubious feedback benefits have
sometimes been cited to justify goveminent-sponsored white
elephants. Furthermore, as Paul R. Krugman of the Massachusetts
Institute of Technology and several other economists have pointed
out, if one country pursues such policies, others will retaliate by
subsidizing their own high-technology industries. Nobody gains The
question of optimal industrial and trade policy based on increasing
returns is currently being studied In. tensely. The policies
countries choose will determine not only the shape of the global
economy in the 1990's but also its winners and its losers.
Increasing-returns mechanisms do not merely tilt competitive
balances among nations; they can also cause economies-even such
successful ones as those of the U.S. and Japan-to become locked into
inferior paths of development. A technology that improves slowly at
first but has enormous long-term potential could easily be shut out,
locking an economy if. to a path that is both inferior and difficult
to escape.
Technologies typically improve u more people adopt them and
firms gain experience that guides further development. This link is
a positive-feedback loop: the more people adopt a technology, the
mote it improves and the more attractive it is for further adoption.
When two or more technologies (like two or more products) compete,
positive feedbacks make the market for them unstable. If one pulls
ahead in the market, perhaps by chance, its development may
accelerate enough for it to comet the market. A technology that
improves more rap. idly as mote people adopt it stands a better
chance of surviving--it has a ``selectional
advantage.'' Early superiority, however, is no guarantee of
long-term fitness.
In 1956, for example., when the U.S. embarked on its nuclear,
power program. a number of designs were proposed: reactors cooled by
gas, light water, heavy water, even liquid sodi
COMPANIES CHOOSE LOCATIONS to maximize profits, which are
determined by intrinsic geographic preference (shown by color) and
by the presence of other companies, in this computer-generated
example, mast of the first few companies set. de in the green
region, and so all new companies eventually settle there. Such
clustering might appear to imply that the green region is somehow
superior. In other runs of the program, however, the red and blue
regions dominate instead. urn. Robin Cowan of New York University
has shown that a series of trivial circumstances locked virtually
the en??e U.S. nuclear industry into light water. Light-water
reactors were originally adapted from highly compact units designed
to propel nuclear submarines. The role of the U.S. Navy in early
reactor-construction contracts, efforts by the National Security
Council to get a reactor--any reactor-- working on land in
the wake of the 1957 Sputnik launch as well as the predilections of
some key officials all acted to favor the early development of
Light-water reactors. Construction experience led to improved light-
water designs and. by the mid-1960's, fixed the industry's path.
Whether other designs would, in fact, have been superior in the long
run is open to question, but much of the engineering literature
suggests that high-temperature, gas-cooled reactors would have been
better.
Technological conventions or standards, as well as particular
technologies, tend to become locked-in by positive feedback, as my
colleague Paul A. David of Stanford has documented in several
historical instances. Although a standard itself may not improve
with time, widespread adoption makes it advantageous for newcomers
to a field--who must exchange information or products with
those already working there--to fall in with the standard, be
it the English language, a high-definition television system, a
screw thread or a typewriter keyboard. Standards that ate
established early (such as the 1950's-vintage computer language
FORTRAN) can be hard for later ones to dislodge, no matter how
superior would-be successors may be.
Until recently conventional economics texts have tended to
portray the economy as someflung akin to a large Newtonian system,
with a unique equilibrium solution preordained by patterns of amoral
resources, geography, population, consumer tastes and technological
possibilities. In this view, perturbations or temporary
shifts--such as the oil shock of 1973 or the stock, market
crash of 1987--are quickly negated by the opposing forces they
elicit. Given future technological possibilities, one should in
theory be able to forecast accurately the path of the economy as a
smoothly shifting solution to the analytical equations governing
prices and quantities of goods. History, in this view, is not
terribly important; It merely delivers the economy to its inevitable
equilibrium.
Positive-feedback economics, on the other hand, finds its
parallels in modern nonlinear physics. Ferromagnetic materials, spin
glasses, solid-state lasers and other physical systems that consist
of mutua?? elements show the same properties as the economic
examples I have given.
They #phase lock'' into one of many possible
configurations; small perturbations at critical tunes influence
which outcome is selected, and the chosen outcome may have higher
enerr/(that Is, be less favorable) than other possible end states.
This kind of economics also finds parallels in the evolutionary
theory of punctuated equilibrium. Small events the mutations of
history) are often averaged away, but once in a while they become
all-important in tilting parts of the economy into new structures
and patterns that ate then proserved and built on in a fresh layer
of development.
In this new view, initially identical economies with significant
increasing. returns sectors do not necessarily select the same
paths. Instead they eventually diverge. To the extent that small
events determining the overall path always remain beneath the
resolution of the economist's lens, accurate forecasting of an
economy's future may be theoretically, not just practically,
impossible. Steering an economy with positive feedbacks into the
best of its many possible equilibrium states requires good fortune
and good timing--a feel for the moments when beneficial change
from one pattern to an. other is most possible. Theory can help
identify these states and times, and it can guide policymakers in
applying the right amount of effort (not too little but not too
much) to dislodge locked-in structures.
The English philosopher of science Jacob Bronows??d once
remarked that economics has long suffered from a fatally simple
structure imposed on it in the 18th century. I find it exciting that
this is now changing. With the acceptance of positive feedbacks,
economists'' theories are beginning to portray the economy not
as simple but as complex, not as deterministic, predictable and
mechanistic but as process-dependent, organic and always evolving.
FURTHER READING MARKET S'TRUCTURE AND FOREIGN TRADE. Elhanan
Helpman and Paul Krugman. The MIT Press. 1985. PATH-DEPENDENT
PROCESSES AND EMERGENCE OF MACRO-STRUCTURE W. Brian Arthur, Yu M.
Ermoliev and Yu M. Kaniovs?? in European Journal of Operational
Research, Vol. 30. pales 294-303; 1987.
SELF-REINFORCING MECHANISMS IN ECDNOMICS. W. Brian Arthur in The
Economy as an Evolving Complex System. Edited by Philip W. Anderson.
Kenneth J. Arrow and David Pines. Addison. Wesley Publishing Co.,
1988.
PATH-DEPENDENCE: PUTTING THE PAST INTD THE FUTURE OF ECONOMICS.
Paul David. I.M.S.S.S. Tech Report No. 533, Stanford University;
November, 1988.
COMPETING TECHNOLGIES, INCREASING RETURNS, AND LOCK-IN BY
HISTORICAL ??TS. W. Brian Arthur in The Economic Journal. Vol. 99.
No. 394, pages 116131; March. 1989.
SCIENTIFIC AMERICAN February 1990 99
BYLINE: From LARRY CAMPBELL in Atlanta District of Colum??ia
BODY:
NETWORK computing industry leader Novell is bailing out of a
number of its existing markets and terminating several product
lines--including Novell DOS 7 to concentrate on new
``technology initiatives'' and usher in an era of
``pervasive computing''
Novell is initially pulling out of the personal computer
operating system business by stopping production of Novell DOS 7, a
product it acquired as part of its take-over of Digital Research.
``The battle for the office desktop is over and MS-DOS and
Windows have won,'' Novell chairman and chief executive Robert
Frankenberg said at last week's Networld+Interop ``94
conference in Atlanta, Georgia, in the United States.
``We will support Novell DOS, but we will not enhance
it.''
``Novell has as much DOS marketshare as Microsoft has
network marketshare,'' said Novell executive vice-president
John Edwards.
``we are focusing on strong areas.''
Novell used Networld+Interop ``94 to introduce these strong
areas, which are part of its vision of the future of computing.
Novell sees networking as it is today evolving to encompass a
much wider, global concept. It envisages everyone now owning a
computer will use networking technology--through the global
information superhighway, among other things.

[[Page 29287]]

It also expects a growing number of people using computers for
the first time in future will also need to connect to information
hubs to share and exchange information.
``Our goal is to take people one step at a time,'' Mr
Edwards said.
``The future is pervasive computing: connecting people to
allow them to work anytime they want--any way.''
The term ``Pervasive computing'' is one Novell has
chosen to define its
South China Morning Post, September 20, 195 vision for the
future. To usher it in, the company is turning its attention to a
range of new products--encompassing operating systems and user
interfaces--and services.
Top of the list is SuperNOS, a planned killer operating system
that will see the best of Novell's existing NetWare network
operating system being combined with the best of UnixWare--its
UNIX counterpart.
There are an estimated 40 million NetWare users on four million
local area networks (LANs) worldwide--more than double the
number of users of all other network operating systems combined.
In addition, there are about 30 million users of UNIX
applications around the world.
It is this formidable market that Novell aims to capture with
SuperNOS, according to Mr Frankenberg.
``The time has come for NetWare NOS to provide all the
services of an operating system,'' he said.
``This is why we are evolving a SuperNOS with NetWare and
UNIXWare on a common Novell microkernel.
``We have left the world of the mainframe. Organisations
have many servers. By ensuring that NetWare and UNIXWare work
perfectly together, we allow our customers to chose which technology
they need on which servers.''
Novell planned to make both products run on a single set of
hardware, or ``as a single system image on multiple hardware
sets'' on a network.
``You get the best of both and a progressive, evolutionary
path from today's specialised, robust backend,'' he said.
``All applications, trained programmers, tools,
interoperability, support, and network services continue on without
change. Perhaps best of all, we build on success, adding
functionality rather than simply re-writing the old.'' SuperNOS
is still a ``concept'', according to Mr Edwards.
``(It is a) codename for a technology initiative to bring
the best of UNIX and NetWare together in a common system''
When complete, the system would be open to licensing and would
be provided on a wide range of platforms, he said.
In addition to its focus on the network operating system market,
Novell is also looking at the client side of the business.
Last week Mr Frankenberg unveiled plans for an ``advanced
Novell client interface that will make it compelling to be connected
networks''
Featuring a graphical three-dimensional user interface with a
``world metaphor'', the system would make network
navigation simple for the first time, he said.
South China Morning Post, September 20, 199.,
However, it would not be a new operating system in its own
right, Mr Edwards said.
Instead, it would be built on existing systems such as Windows
95.
``We will see over four to six months of demonstrating and
customer testing of this system (before it is brought to
market),'' he said.
``It will browse the Internet, NetWare and NCS networks and
live in MS Windows, Chicago, UNIXWare and other desktop operating
systems,'' Mr Frankenberg said.
``It will bring not only these end user services, but also
compelling consistent NAPIs (network application programmer
interfaces) for Windows, UNIX and other developers to unlock the
power of the network from client applications.''
These new areas of focus do not just see Novell pulling out of
the desktop operating system market--which was itself a move
the market ``welcomed'', Mr Frankenberg said.
In addition, Novell is pulling out of the database business, up
to a point. Having sold off Btrieve, its database product, the
company is now only working with partners in the database area.
It will steer clear of creating vertical applications and, while
working with information service providers as part of its networking
technology initiative, it will not become an information service
provider itself, or attempt to provide communications
infrastructure.
``This frees up a considerable number of people who are now
making the network fulfil our vision,'' Mr Frankenberg said.
Hardware would also be an area that Novell would abstain from
dabbling in, he said.
``(Former Novell chairman and chief executive) Ray Noorda
got us off hardware in the ``80s. I will keep us on the wagon
in the ``90s,'' he said.
Microsoft's Operating System and Application Strategy for
Servers Fine-Tuning Microsoft's Server Strategy
Microsoft competitors have taken great pleasure in the slow
acceptance of the much hyped NT. Some of this gloating is certainly
well deserved. After a long period of anticipatory eulogies for
competitive operating systems, NT barely shipped 400,000 units in
its first full year of availability. This is one-sixth the number of
OS/2 shipments and only marginally higher than Solaris'' 1993
shipments (see Figure I). Moreover. the majority of NT shipments axe
either free copies or axe being used for development or evaluation.
2.400
4OO*
Figure 1: Licenses of 32-Bit Operating Systems Shipped in 1993
(thousands of units)
Are competitive OS vendors beginning their celebrations too
soon? After all, consider how much solace Apple took in the slow
acceptance of Windows. Just as importantly, many NT cynics are
finding their evidence in the wrong places. They axe looking at the
small number of total NT units, the minimal acceptance on the
desktop and the technical deficiencies of the operating system. Many
of those competitors who view NT as a server operating system are
focusing on comparisons with and difficulties in displacing NetWare.
Those who want to objectively assess the prospects for NT should
instead examine the positioning, capabilities and increasingly high-
profile endorsements of NT as an application server operating
system. While NT's acceptance as a desktop and file server OS has
been slow. a growing number of large, leading-edge corporate
customers see tremendous potential for NT as an application server
in department-sized environments. More importantly, Microsoft has
optimized NT's server capabilities by segmenting its development
focus between desktop and server versions and by introducing a broad
range of complementary server offerings.
Meanwhile, most of the leading server application
vendors--including those introducing client/server versions of
applications that had been available only on minicomputers and
mainframes-- have selected NT as one of their first server
operating systems, and the one offering the largest market
potential. What are NT's real prospects as a server operating
system': How can one capitalize on its potential without making
their company's future too dependent on Microsoft and ,'OT?
Windows NT: The Rumors of its
Death Are Premature
Make no mistake. Windows NT and its successors are Microsoft's
strategic operating system. As Microsoft Executive Vice President
Mike Maples states. ``by the end of the 1990s there will be one
Microsoft operating system--NT-- but there will be three
of them: NT Advanced Server, NT Advanced Workstation and Windows
NT'' (see Figure 2) Microsoft views the slow initial acceptance
of NT as only a relatively minor delay in its quest for global
software domination.
As discussed in Summit Strategies'' report, Profiting from
the Transition from Personal Desktops to Enterprise Desktops,
initial NT desktop acceptance will Figure 2: Future of the Windows
Architecture be limited primarily to engineering, publishing,
software development, trader workstation and a few other specialized
applications with particular performance. security and reliability
requirements. This will begin to change as developers write
applications to Win32. Most of these applications will be optimized
for Chicago. but they also will provide hative performance on NT
Workstation and. then, on Cairo.
The story is very different for NT Server. NT Server is
Microsoft's future. It is THE FOUNDATION of all of Microsoft's
target growth markets-- workgroup, department, enterprise.
advanced consumer and information highway. Microsoft, however, has
little or no experience or credibility in those markets. It must
develop them essentially from scratch.
Microsoft recognizes these limitations and is dedicating
extensive commitment and resources to its efforts to establish NT as
a standard server operating system. It has carefully studied the
factors that made other enterprise and server operating systems
successful and has developed a strategy that combines some of the
most important of

[[Page 29288]]

these factors with Microsoft's own unique twists. NT Server as the
Foundation for Microsoft's Solution Platform Microsoft's most
obvious work on NT is in the form of Daytona. which will be more
formally known as Windows ,NT 3.5. Daytona will deliver higher
performance with smaller memory and will provide better reliability,
robustness. SMP support and connectivity than version 3.1.
It will be divided into two optimized versions--one for
advanced desktop users (NT Workstation) and one for servers (NT
Server). This division will mark the beginning of separate, but
still binary compatible code bases that are targeted at separate
markets. Daytona also will provide a migration path to Cairo I NT
version 4.0), the scaleable, object-oriented OS that Microsoft plans
to release by the end of 1995.
However. as important as all of these operating system
enhancements may be. they are only the foundation of a much broader
Microsoft server strategy. This strategy is based on a broad range
of server applications that Microsoft is developing to run on top of
NT Server and which will tailor the OS for use in specific
functions.
Microsoft plans to ship five server applications that will run
on top of NT Server. some of which are already shipping: SQL Server.
SNA Server. Systems Management Server. Exchange Server. and
``Tiger'' Video Server.
These server applications will likely be joined by others,
including a search and navigation engine, server versions of many of
its client-based Microsoft Ofrice applications and. possibly, some
``diagonal'' server-based business applications, such as
accounting, human resources management and sale, automation.
Microsoft is also developing an online service t code-named Marve??
that ,a ill generally compete with Prodigy and America Online.
Although these server applications are very different from each
other, all share at least two important factors: They are designed
as general, extensible frameworks on which partners are encouraged
to write their own specialized applications: and each is available
on and optimized for use with NT Server and is designed to work
seamlessly with all other Solution Platform tools and applications.
The combination of these factors will make NT Server a unique,
very formidable server operating environment. Creating a Consistent,
Universal Server Environment Each Microsoft server applications
competes with some third-party offerings. SQL Server. for example.
competes with Oracle7 and Sybase System 10. Exchange competes with
Lotus Notes and Novell GroupWise.
,Microsoft. however, is positioning each of these applications
as generalized. extensible platforms on top of which smaller, more
specialized and verticallyfocused applications can be when.
Like Oracle and Sybase. Microsoft is attracting third-party
developers to write specialized applications on top of its own
generalized platforms. Unlike Oracle and Sybase, however, Microsoft
will not develop these applications itself. It will leave this add-
on market exclusively to third-party partners and has developed a
number of large, well-funded cooperative technical, marketing,
distribution and consulting programs to help these partners enter
and expand their markets. .Microsoft has already attracted more than
600 partners to write applications on top of SQL Server. more than
25 to write for Systems Management Server. and 70 partnets to write
for Exchange Server. SQL Server applications, for example, range
from diagonal accounting and document management through vertical
applications for insurance and health care.
This base of third-party applications will help make the
generalized Microsoft Server Platform a viable foundation for a
broad range of highly specialized applications. In and of itself,
however, this is not different from what is provided by competitive
OSs (i.e., NetWare and Unix), databases (such as Oracle7) and
groupware environments (such as Notes). Microsoft. though, takes a
giant step beyond these competitive environments by:
Optimizing its applications for, and integrating them closely
into its OS to provide fast performance, permit the application to
take full advantage of all operating system capabilities (without
duplicating them) and provide the basis for integrating important
application capabilities directly into future versions of the OS.
Providing a common set of development tools and integration
protocols that allow third-party applications to be easily
integrated into and take full advantage of the operating system and
all Microsoft server applications and to integrate closely with
Microsoft desktop OSs and applications.
This integration is critical to Microsoft's entire server
strategy, it provides developers with a single set or'' APIs
and communications protocols with which they can develop to all
Microsoft desktop and server OSs and integrate with all compliant
Microsoft and third-party applications. It provides customers with a
modular, comprehensive. ``easy-toown'' server environment.
Microsoft also is laying out a road map under to make this
integration closer and deeper. As a result, data semantics and query
technology will be common across both desktop and server components
and communications will be facilitated between them.
More importantly, the OLE object model--already supported
by all Microsoft and a small, but growing number of third-party
applications--will form the foundation of Microsoft's next-
generaLion Cairo operating system. In addition. many new Microsoft
products are based on a technology that will be used in Cairo. This
will simplify the upgrade path to Cairo and will allow the new OS to
take over many of the capabilities of previously distinct
applications.
Since Cairo will be a pure object-based OS, it will be highly
modular. Components will be easily added, deleted or replaced,
making it relatively easy for resellers or customers to customize
the operating system and incorporate traditionally distinct
functions into it. In fact, since all Microsoft server applications
will fit into a single, integrated Cairo model, it will be almost
impossible to distinguish between the operating system and the
applications.
Redefining Server Industry Rules to Match Microsoft Strengths
Microsoft's approach promises to make NT Server a much more
comprehensive. integrated server environment than is available from
any other client/server operating system, relational database or
messaging backbone vendor. In fact. NT Server will approach the
level of integration that previously had been available only in
proprietary mainframe and minicomputer environments.
In and of itself, this integration will be attractive to large
numbers of customers. application developers. OEMs and resellers,
but Microsoft plans to go even further. It will offer these
capabilities in a new way that no other competitor can directly
match. It will combine capabilities that had traditionally been
available only as high-priced, customdeveloped solutions on
expensive platforms with price levels and distribution channels that
were available only for basic PC-level solutions. In other words.
Microsoft plans to redefine the rules of competition in the server
operating system and applications market.
It will rewrite these rules in a way that builds on its existing
business model and makes it difficult, if not impossible, for other
vendors to follow.
Summit Strategies believes that Microsoft will execute this
strategy gradually and in a way that permits the incremental
extension of its traditional low-overhead, product-oriented, virtual
company business model.
It will establish this presence in a niche in which there is
very little entrenched competition--department-level, decision
support application servers (see Figure 4). It will position the NT
server environment as a more functional, scalable application
platform than NetWare and a less expensive, easier-to-own
alternative to Unix. While Microsoft plans to ultimately replace
NetWare and Unix, initially it will coexist with them by emphasizing
connectivity with Unix and its use as an application server within
existing NetWare file server environments.
Enterprise Application Server
??
File/Print
Workgroup Application Serve,
Unix Core Market
?? Server
Core Market Core Market
1993
2000
Source: Su?? Strategies, Inc.
Figure 4 : Microsoft's Trojan Horse Strategy
Microsoft will use this market as a beachhead from which to
expand gradually into complementary segments, such as department-
level and branch transaction servers, workgroup application servers.
file and print servers and eventually, into some division-level
environments. Summit Strategies expects this strategy to allow
Microsoft to grow NT's position in the network server operating
system market from about 2.5 percent in 1993, to almost 15 percent
by 1997. It will play much larger roles in the application server
market and. especially, in the lowend to midrange of that

[[Page 29289]]

market. Obstacles to Microsoft's Dominating the Application
Server--
Part I
Microsoft is certainly well-positioned to establish a strong
position in the application server market. Its success, however, is
far from assured. The company still faces a number of strong
competitors and must overcome a number of self-imposed obstacles.
These obstacles fall into two primary categories: some are product-
based while the others are a result of the company's business model.
Microsoft's productbased obstacles are:
The perceived unreliability of
Microsoft server solutions Everybody recognizes the limitations
inherent in the Windows desktop environment. Most customers are
willing to put up with these Limitations in return for the benefits
of low cost, application availability and standardization.
Customers, however, are much less willing to accept such limitations
in application server environments, particularly when they are using
the servers to run business-critical applications that had
previously been entrusted only to mainframes and mini-computers.
On one ban''. NT is relatively robust for a Version 1.0
operating system. However. it is still immature, unproven and lacks
many of the complementary tools that will be required for acceptance
in business-critical environments. Microsoft does promise more
robust upgrades to its operating system. RDBMS and communications
software, new versions of needed system management and messaging
software, and improved fault tolerance and recoverability. However.
its continual missed shipment deadlines do not instill great
confidence.
The limited openness and scalability of the Microsoft solution
Although Microsoft operating systems may be standards, they are
not open. This creates a risk, since customers who adopt them will
have a difficult time migrating to another operating systems, should
the need arise. This problem will be particularly acute for
customers who buy into Microsoft's server applications. since these
applications will be available exclusively on NT Server and will be
integrally linked to it.
This lock-in could be particularly dangerous for customers who
require that their applications be highly scaleable, up through
enterprise environments. Microsoft solutions currently support
symmetric multiprocessing and will support clustering and be
portable to all major processors. However, NT Server is currently
tuned for single and dual processing. Its next implementation is
only likely to scale to four processors. which is far below the 16-
to 30-CPU tuning of a number of versions of Unix. There are,
however, mitigating factors for each of these concerns. Consider
robustness. While Microsoft has missed deadlines in shipping
virtually all of its key products, once they do ship. they are
reasonably stable and deliver on most of the company's promises.
When push comes to shove, most customers would prefer to receive a
stable product late. than a buggy product on time. But. regardless
of when Microsoft ships. computing environments with overwhelming
needs for proven, reliable server environments are unlikely to
select Microsoft products, at least for the next several years.
As for openness and portability, it is largely a question of
target markets and tradeoffs. Generally speaking, large corporate
MIS departments are most likely to demand that their server
environments be open, flexible and scalable. Most of these MIS
groups have the capabilities or the resources required to configure.
develop for and administer these solutions. In contrast, many small
businesses and department-level customers will be willing to trade
off such benefits in return for solutions that are easier and less
expensive to buy, configure and manage, and for which off-the-shelf
applications are generally available.
The percentage of the market that will fall into each camp is
certainly debatable. While everyone says that they want open,
scalable and robust solutions, when it comes time to make a final
decision. Summit Strategies believes that many more customers will
choose easy, cheap and standard.
Obstacles to Microsoft's
Dominating the Application
Server--Part II
The other, and more difficult obstacles to Microsoft's success
in the server market are more dependent on the company's business
model and style of operation, than on its technology. Summit
Strategies sees three primary. obstacles in this category.
Microsoft's penchant for making enemies
Microsoft has always had a way of making enemies due to such
factors as its sheer market power, position as industry upstart,
cocki

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3AX02-140503. Public record. Not legal advice.