apple powerbook design quality & time to market 994023-pdf-eng

Do Not

Cop

y or

Pos

t

This document is authorized for use only by Mandeep Singh Hayer at Lovely Professional University until October 2012. Copying or posting is an infringement of copyright. [email protected] or 617.783.7860.

DESIGNMANAGEMENT

INSTITUTE

CASE STUDY

Apple PowerBook:Design Quality

andTime to Market

9-994-023

Do Not

Cop

y or

Pos

t


This case study came from the CaseStudy Research and DevelopmentProgram at the Design ManagementInstitute’s Center for Research. TheCenter conducts research and developseducational materials on the role of designand design management in businesssuccess. Case studies, the DesignManagement Journal, reprints from theJournal, and other educational materialsare available from the Design Manage-ment Institute Press.

Design Management Institute PressDesign Management Institute29 Temple Place, 2nd FloorBoston, MA 02111-1350 USAPhone: 617-338-6380Fax: 617-338-6570E-mail: [email protected] site: www.dmi.org

Harvard Business School Publishing is theexclusive distributor of this publication.To order copies or to request permission tophotocopy, please call 617-495-6117;in the U.S. call 800-545-7685; or write:Harvard Business School Publishing,Customer Service Dept.,60 Harvard Way, Boston, MA 02163

Acknowledgments: Many thanks to BobBrunner and the people at Apple Computer fortheir openness in discussing their experiences ofthe PowerBook project, and especially to JonKrakower whose long memory and extraordi-nary commitment to this project made it possibleto tell a coherent story; to Bill Evans of BridgeDesign, who conducted many of the interviewsand provided early orientation to the project; andto Earl Powell, President of DMI, whoconducted interviews and envisioned thepossibility of the larger program of which thiscase is one part.

Dr. Artemis March

© Copyright 1994Version 1.0, August, 1994The Design Management InstituteAll rights reserved. No part of thispublication may be reproduced withoutwritten permission.

Do Not

Cop

y or

Pos

t


Design Management Institute Case Study

Apple PowerBook:Design Quality and

Time to Market

3

■ This case was prepared by Dr.Artemis March of the DesignManagement Institute as a basis forclass discussion rather than toillustrate either effective or ineffectivehandling of an administrativesituation.

Apple Computer, Inc. launched the development of a portable com-

puter late in 1985. None of the portables in the market had been

successful, and Apple’s goal was to address its competitors’ shortcomings:

passive and underlit screen displays, very short battery life, and limited

performance. Apple did not compromise in its search for solutions.

Graphical user interfaces, which Apple had pioneered in desktop comput-

ers, required more speed, higher resolution, and more sophisticated

displays than could be provided by the passive matrix displays used by its

portable competitors. Unwilling to forego its leadership in front-of-

screen performance, Apple insisted on having an active matrix display in

its first portable. Finding a display supplier with a reliable manufacturing

process delayed introduction by twelve months. A full performance Mac

Portable was introduced in late September 1989 at a list price of $4,999.

The Portable successfully addressed its objectives, but the initial back-

log of orders collapsed before Christmas. Although the battery ran for

8-12 hours, the product weighed nearly 17 pounds while Compaq’s full-

function LTE, introduced just six weeks later, weighed seven. Apple soon

had a huge inventory of unsold Portables. Randy Battat, then Vice-Presi-

dent of Worldwide Product Marketing, commented, “We had been

concerned about competition but I think we felt we had a lot of inherent

competitive advantages that would outweigh weaknesses in other areas.

Do Not

Cop

y or

Pos

t


4

Apple PowerBook: Design Quality and Time to Market

We didn’t recognize how fast the marketwas moving.” Had the Portable come outtwo years earlier, the consensus was that itwould have been a smashing success. But,as one manager noted, “When you put aPortable next to an LTE, you had to say,‘Where were we?’”

The Mac Portable brought into sharperfocus a growing recognition within thecompany that major changes were neededin the way it was organized and how it op-erated. One manager summarized thecompany: “Apple is not ‘time to market’.It’s more ‘time to perfection’.” Time tomarket was described as being alien to aculture that “shipped no product before itstime.” Yet Apple was now challenged togain entry into the notebook market in 18rather than 48 months.

In October 1991, Apple introduced thePowerBook 140/170, a seven-pound note-book, and immediately had a runaway suc-cess. Originally envisaged as a temporaryplaceholder, the PowerBook sold 400,000units in its first year, becoming a billion-dollar business overnight.

THE PORTABLE COMPUTER MARKET

Portable computers were originally con-ceived as “second computers” that wouldbe particularly useful to salespeople, con-sultants, and others who worked exten-sively outside their offices. The “portables”of the middle 80s were in truth “lug-gables,” such as Compaq’s 28-pounder,introduced in 1983. Price-performancecontinued to improve, while technologyadvances made it possible to shrink thesize and reduce weight. When portablesalso became battery-operable, their rangeof use and customer base broadened con-siderably. Factory sales of portables were$2.3 billion in 1988, and jumped to $3.3billion the following year.

By the end of the decade, four seg-ments were emerging, based on weight.“Portables” weighed over 11 pounds;“laptops,” 7-11 pounds; “notebooks,” lessthan 7 pounds; and “handhelds,” undertwo pounds. Notebooks were projected tobecome the leading segment. Pricesranged from $1000 - $6000. The high endmodels contained more powerful micro-processors, and were full-performancecomputers.

APPLE COMPUTER, INC.

Product Strategy

Apple Computer, Inc. was founded in1977 by two young entrepreneurs with apassionate commitment to their physicalproduct. Apple dramatically changed thepersonal computer (PC) market in 1984when it brought out the Macintosh (Mac).The Mac introduced a graphical user inter-face (GUI) which allowed users to per-form numerous functions by pointing atobjects (icons) on the screen with a devicecommonly called a “mouse.” DespiteApple’s proprietary operating system andits incompatibility with IBM/MicrosoftDOS, the Mac was enormously successful.Its user-friendly interface made computeruser/owners out of nonusers, while itsgraphics capabilities soon gave Apple aleading position in the areas of desktoppublishing and presentation graphics.Apple’s share of the PC market was secondonly to IBM’s.

Apple’s basic strategy was to developuser-friendly, differentiated productsbased on its proprietary technologies.These products commanded a price pre-mium, which gave Apple the high grossmargins that allowed it to reinvest inR&D, and continue to develop very attrac-tive products. Successful implementationalso required growth. But growth and highmargins had become more difficult toachieve as the desktop computer industrymatured, the number of competitors pro-liferated, competition became more price-based, and differentiation harder toachieve.

Organization

In 1989, Apple’s activities were organizedas functions and departments. No re-sources were dedicated to particularprojects; each department supported allprojects. Product Development (i.e., engi-neering), Product Design, WorldwideProduct Marketing, and Manufacturing allreported to Jean Louis Gassee, president ofApple Products. “Engineering” at Applemeant electrical engineering. The central-ized engineering organization housed bothhardware and software. (See Exhibit 1.)

With this departmental organization,project managers could not, as Jon Sedmak

Originally

envisaged as a

temporary

placeholder, the

PowerBook sold

400,000 units in

its first year,

becoming a

billion-dollar

business

overnight

Do Not

Cop

y or

Pos

t


5


put it, “make decisions stick.” Formerlywith Texas Instruments, and more re-cently a project manager at Apple, Sedmakdescribed some of the frustrations. “Youhad to deal with other departments thathad other things to do on numerousprojects. You had to get consensus notonly with the product design engineersthat were working on your stuff, but alsotheir boss and their boss’s boss. Everythingwent up. When John Louis [Gassee] washere, he controlled industrial design; so IDwould kick things up to John Louis andthen they would come down again.”

Industrial Design at Apple

During the period in which the Portableand PowerBook were developed, mechani-cal engineers and industrial designers werepart of a single centralized departmentcalled product design. Headed by RichardJordan, its role was to create the productenclosures that housed the electronics andinterfaced with the user.

When Jordan first approached Robert(Bob) Brunner to take over managementof the design group, Apple had only fivedesigners, and did 90% of its design workthrough outside consultants. Brunner,who was then 31 and had worked withApple for two years as a consultant,turned down the offer. “It wasn’t right forme or for Apple. With such a skeletal

staff, there really could be no direction-setting.”

Subsequent discussions generated a dif-ferent offer—to create a strong internalgroup that would chart Apple’s strategicdirection in design. Brunner accepted theposition as manager of the IndustrialDesign Group (IDG) in November 1989;he reported to Jordan. “I came to Applebecause it has great products, and is drivenby product design. People here are very vi-sually oriented. They rally around objects,and the company centers around physicalobjects. Design is embedded in the cul-ture. I sometimes say I have a staff of10,000.” Brunner noted as well the archi-tectural environment that Steven Jobs, oneof the company’s founders, had created forApple. “We are surrounded by a very highquality physical environment.”

PORTABLES

Product Divisions

In the months following the Portable’s in-troduction, senior managers became com-mitted to the idea of reorganizing thecompany by product divisions and busi-ness units. Early discussions favored sub-dividing the company by high-end andlow-end products, but by the spring of1990, this was rejected in favor of a moresustainable organizational difference

CEOJohn Sculley

President, Apple ProductsJean Louis Gassée

Product DesignDirector

Richard Jordan

World-Wide MarketingVice PresidentRandy Battat

ManufacturingVice PresidentFred Forsyth

Product DevelopmentVice PresidentEric Harslem

CPU EngineeringDirector

John Medica

•Other Hardware•Software•Testing

MechanicalEngineering

Industrial DesignGroup

Bob Brunner

Exhibit 1Apple’s Organization,November 1989

Do Not

Cop

y or

Pos

t


6


between desktops and portables. This deci-sion was made at a time when Apple hadvirtually no portable business; MacPortable had been written off, and nothingelse had yet been developed. A portableprogram was underway, however. (SeeExhibit 2.)

Twister

Senior industrial designer Gavin Ivesterhad been asked to undertake some con-

cept studies in June of 1989. Ivester was“volunteered” for this project because hehad pointed out some of the Portable’sdesign problems, and was subsequentlytagged as “Mr. Portable.” Working withtwo outside design groups, Ivester wasguided by the concept of the “mobileMac.” Aiming at a 3.5 pound notebookthat “you carry with you all the time,” hisslender models incorporated a level ofminiaturization that “didn’t quite exist. I

(Cancel)

1989 1990 1991

J J A S O N D J F M A M J J A S O N D J F M A M J AJ S OPortable Program

Twister

Portable enhancement

Companion

Portable introduced

Compaq LTE Introduced

TIM (PowerBook 140/170

PowerBook 100 (with Sony)

TIM

JK paper model

JK working model

ID configuration studies

Trackball integration

Internal/External issue

Sculley approves concept

Project formed

Design test iterations

“Millimeter Madness”

First builds

First articles

150 items

Prototype builds/tests; ramp

140/170 Introduced

Organizational Changes

Brunner heads ID

Gassée retires

Engineering decentralized;product divisions

TIM Project formalized

Product design co-locates

Bentley joins as product designmanager

Infusion of resources for TIM

Exhibit 2PowerBook Design Timeline

Do Not

Cop

y or

Pos

t


7


went to see all the technology experts atApple to see what was coming.” Ivester’sseries of five studies, called “Twister,”drew on these next-generation technolo-gies such as a 2.5 inch hard drive, midastrackpad, and touch-screen. (See Exhibit3.) The 3.5 pound target was establishedin conjunction with a marketer who madeup some wooden blocks in a range ofweights, and through testing, found thebreak point for “carry with you all thetime” to be 3.5 pounds.

Companion

In August 1989, Apple began to formalizeits next step in portables by launching athree-year program called “Companion.”The program gave some of Apple’s bestand most experienced engineers and de-signers the opportunity to completely re-think portables from scratch. Their overallproduct objective was to create an ultra-light, ultra-small notebook. The productwould be risky because it would incorpo-rate several new technologies to achievethe needed degree of miniaturization.

Enhancing the Portable

Despite the Portable’s size and weight,many hoped it would hold Apple’s place inthe portable market for a couple of yearsuntil the much more miniaturized Com-panion could be developed. Their hopeswere pinned on high demand from Apple’sinstalled base. A second portable projectgot underway in August to enhance the15.5" x 14" x 4" Portable. It would maintainthe same weight and look as the Portable,but add backlighting, upgrade the power ofits 68000 microprocessor to a 68030, andmodify the main logic board; the latter en-tailed a considerable electrical engineeringeffort.

DEFINING THE NOTEBOOK CONCEPT

Some, including John Sculley, chairmanand CEO of Apple, were not satisfied withthis portable program. Concern mountedas the Portable’s failure in the market be-came evident. Late in 1989, engineeringwas asked if it could pull in Companionwithin a year. It could not. Could the Por-table be shrunk to notebook size? Simplyentering the market required a productthat weighed no more than seven pounds,

in a package no more than 9" x 11" x 2.5",and that used an 030 chip. It would have touse existing technologies to get to marketin 18 months. Jon Sedmak, who was man-aging the Portable enhancement project,was asked to oversee the exploration of thisnotebook concept as well. During a subse-quent meeting, people kept writing TTM[meaning “time to market”] all over theboard. A latecomer misread the board andasked, “What is this ‘TIM’?” The namestuck; people immediately began callingthe project TIM.

ID Models

In January, Bob Brunner and Gavin Ivesterled a small group of Apple designers in anexploration of component layouts forTIM. Their configuration studies movedcomponent blocks around, producing atleast eight primary models of internal con-figurations. They used two kinds of 3Dmodels—foam to show scale and form,and crude hard models to give a bettersense of size and weight. All of these de-signs were “clamshells” (i.e., top and bot-tom halves that opened and closed like abriefcase), and used a trackball of indeter-minate size to control the cursor. A front-and-center trackball arrangement pre-dominated. Some models had an internalfloppy disk drive, and some had an exter-nal floppy module. The designers exploredhow a modular floppy would be attached,

Exhibit 3Twister Model, summer, 1989

Do Not

Cop

y or

Pos

t


8


Exhibit 4ID Models for TIM, January-

February 1990

e.g., by snapping into the side or dockingon the bottom. (See Exhibit 4.)

While the clamshells were more con-ventional than Twister, both conceptsdrew on Apple’s existing design language,called “Snow White.” Brunner describedthis language as “rectilinear, light, ordered,and strong.” This “language of the desk-top” had been applied across all Appleproduct lines, and the designers were feel-ing its constraints. Brunner described thesituation: “Snow White was predictable,and making it harder to innovate. It lackedextensibility.”

Trackball

Portables required a device to control thecursor; the standard approach was to dothis through the keyboard. Because of itsGUI, Apple was the first company to befaced with solving the ergonomic/space/weight problems of building a pointing de-vice into a notebook. The Portable hadused a trackball—a freely rotating, rounddevice that had previously been used invideo game arcades as well as high-endworkstations. By rolling the trackball, theuser controlled the position of the cursor.By pressing on the trackball buttons, theuser made a selection.

For the Portable, Apple had designed a33 mm trackball which could be config-ured by the dealer for either side of thekeyboard. The size of a trackball had manyimplications. A large diameter enhancedfine motor control of the cursor, but ittook up more space and was heavier, andwas thus more difficult to integrate. Its di-ameter affected its height, a factor whichhad to be considered in relation to theheight of the keys and the depth of the dis-play, for the display closed over both thekeyboard and trackball when the notebookwas closed.

Krakower’s Concept

Another person who had been dissatisfiedwith the timetable of the portable programwas Jon Krakower, an electrical engineerwho was passionately committed to mak-ing Apple a major competitor in the note-book market. He had begun his graduatework in architecture, and had a deep andabiding interest in design. Although thenarrowest definition of his role had to dowith board design and layout, he was bet-ter known as a system integrator—some-one who made sure that all the boards,electricals, mechanicals, and softwareworked together, and could be manufac-tured easily.

Krakower felt that Apple was alreadytwo years behind in portables, and if itdidn’t get a competitive notebook out in18 months, it would miss the window ofopportunity and might never get anotherchance. He also felt that Apple’s productwould have to be clearly differentiatedfrom the competition’s, and that ergonom-ics should be a priority in its distinctive-ness. On his own initiative and time, hebegan thinking about a concept for a small,light package whose pointing device wasambidextrous. His first notion of what thatmight be took the form of a paper modelmade from marketing materials for thePortable at its New York introduction tothe press in September. Its central featurewas a large, centered trackball. (See Ex-hibit 5.)

A centered trackball located in front ofthe keyboard addressed the needs of left-handers as well as making it possible forusers to control the cursor with theirthumbs without moving the hands awayfrom the keyboard’s home-row position.But what about the space on either side ofthe trackball? Krakower began to envisage

Do Not

Cop

y or

Pos

t


9


a “palm rest” that would support the user’shands. By pushing the keyboard furtherout from the user as well as providingsupport in its own right, palm rest re-duced the user’s lateral wrist angle whenworking in a cramped space, and moreclosely approximated the normal distanceand angles between a user and a desktopkeyboard.

Early in 1990, Krakower put together aworking model of his design. (See Exhibit6.) He cut up a Portable, devised a palmrest, put the trackball in the middle, andstarted showing it informally, beginningwith peers and first-level management,and later, second and third-level manage-ment. Many within senior managementthought it looked strange, and that it mightgive Apple a second black eye in the por-table market. In addition, many productdesigners seriously doubted that it waspossible to be competitive on size andweight with a large trackball, full-size palmrests, and a built-in floppy disk drive.

To validate the concept and ease man-agement concerns, Krakower set up a tem-porary user test facility where he videotaped45 subjects using a modified version of hisoriginal model that allowed him to vary thesize of trackball and of palm rest. In addi-tion to this test data, he also produced doc-umentation showing how the size/weightgoals could be attained through usingproper module placement, a multitieredPCB/cable partitioning scheme, and the lat-est miniaturization techniques in compo-nent packaging technology.

Integrating the Trackball

Brunner recalled, “The biggest challengethat we faced initially was the integrationof the pointing device. How could we in-tegrate it without growing the box, or do-ing some ugly appendage? This is wherewe kept hitting the wall.” One day aroundthe middle of February, Krakower stoppedby and suggested a solution. Brunner re-called, “Then Jon said, ‘Wait a minute, wecan just push the keyboard back, pull thebattery forward into the left front cornerand the hard drive forward into the rightfront corner, and there you go. You candrop the trackball right in between the bat-teries and the hard disk.” This layout hadthe further advantage of locating two of

the heaviest components in front; this cre-ated a forward center-of-gravity, a criticalconsideration for wheelchair users and anattraction for mobile users when balancingthe unit on their lap.1

Floppy Drives

The issue that next came into focus waswhether or not to have an internal floppydrive. The LTE and most other competi-tive products used an internal floppy. Aninternal floppy added weight to the prod-uct, and competed with other componentsfor interior space. If Apple chose an inter-nal floppy, it could match the size/weight

Exhibit 6Krakower’s working model,January-March 1990

Exhibit 5Krakower’s paper model,September, 1989

1. Gavin Ivester, “Shrinkingthe Mac Wrap,” Innovation,Fall, 1992

Do Not

Cop

y or

Pos

t


10


norms for notebooks; but if it chose an ex-ternal floppy, it could possibly gain anedge for lightness. Many people at Applefelt—and some quite strongly—that an in-ternal floppy was not necessary; thosepeople included Gavin Ivester; John

Medica, director of CPU engineering;Rodger Mohme, project leader for Com-panion; and Bruce Gee who later becamethe marketing manager for PowerBook.

Krakower believed it was critical thatApple’s notebook have an internal floppy

Modular Floppy Integrated Floppy

Advantages Disadvantages Advantages Disadvantages

1. Feature Set Provides Customer option to Customer perception of less Full Waimea feature set, minus NONEremove floppy drive. than full-function machine; internal PDS card capability.

does not keep pace withcompetition.

2. ID Volume Fits with 9.5" x 11" form NONE Fits within 9.5" x 11" form NONEDimensions factor; max height--52mm. factor; max height->52 mm;

same size as floppyless model.

3. Weight Unit is about 0.75 lb. lighter NONE Will weigh-in at 7 lbs., +/- NONE:-> with the machinethan other model (all other 0.25 lb.; very competitive for being transported in the carryingfactors being equal) outside a full-function ‘030/386 lap- case, the overall package weighscase. top in ‘91. less.

4. PC Board(s) One continuous 10.5" x 3.8" Only 79 sq. in. (top & Total of 138 sq. in. of rout- Requires two PC boards; 4.0" xrectangle; good for routing bottom). able area (top & bottom); 4.75" two-sided SMT boardand production handling. 75% more area. stacked on top of motherboard.

5. ASIC NONE Requires massive integration, Requires not further ASIC NONEIntegration including TAB on bottom integration; can use WalmeaRequirements side. 84-pin PLCC chip set as is.

6. Ergonomics TBD Unproven 1.5" palmrest feat- Tested 3.85" palmrest feat- Needs further ergonomic testing.ure; unproven mini-trackball ure; with sample of 45 sub-design. jects; proven trackball design.

7. Product TBD Thicker reinforced walls re- Internal plastic support walls NONEDesign quired to support HDA, in- minimized, resulting in red-

verter, modem above PCB; uced design and tooling leadschedule risk due to modular times and less plasticfloppy design issues. weight.

8. Battery NONE 4.4" x 10" flat battery is not Standard 2.4 Ah mono-block NONEuser friendly nor very rugged; style cartridge SLA battery iscosts more and is heavier rugged and easily inserted bythan SLA block. customer.

9. Floppy Can be dismounted at will. Increases susceptibility for One less thing to worry about; NONEmechanical damage to mach- frees up room in carrying caseine; bad aesthetics on right side for spare battery, power adapt-regardless of whether floppy is er, etc.attached.

10. Trackball TBD Schedule risk due to new de- A slightly reduced version of NONEsign; requires further ergo- current Portable trackball cannomic testing; difficult to use be used; test subjects preferin lap position. larger trackball.

11. RAM/ROM NONE Only one slot available for Two identical slots available; NONEExpansion RAM expansion; customer allows for ROM expansion as

must take unit to dealer for well as 8 MB of RAM; custom-installation. er installable cards.

12. Modem NONE Customer must take unit to Easily installed by customer; NONEdealer for installation; difficult RJ11 connectors readily ac-to design modem to fit mech- cessible on left side of mach-anical outline. ine.

13. Carrying NONE Overall weight and size of Size and weight is less than NONECase case (with machine and floppy other model's case because

inside) will be greater than of modular floppy's metal/other model. plastic housing and extra pad-

ding/pocket requirements.

Exhibit 7Comparing internal and external

floppies

Do Not

Cop

y or

Pos

t


11


the internal floppy and full-size palm rest.He favored the large trackball because thetrackpad was about the same size, andTIM would be able to accept the trackpadwhen it was ready. Brunner noted that,“After the Sculley meeting, we didn’t ex-plore other configurations. There was notime. We were looking for the finaldesign.”

Gee later noted, “Going with an inter-nal floppy was the smartest decision wemade. A machine was not perceived asfull-functional unless it had a built-infloppy drive.” Brunner retrospectivelycommented on the early commitment tothe product concept: “When we began ourusability tests in April, we were alreadycommitted to what we were going to do;all we were doing was refining it. Such anearly commitment was exclusive, I think,to this project. We knew we were behind,and we felt we had nothing to lose. There-fore, we made a commitment to that con-cept without a lot of data backing it up.”

TIM: FORMING A PROJECT

In the spring of 1990, several things hap-pened that had an impact on TIM. Gassee,who had taken a strong and active interestin ID and been its de facto head, an-nounced his retirement. Apple decentral-ized into product divisions in April. Bothengineering and product marketing weredecentralized while product design wasnot. John Medica was asked to form, lead,and build the new Portable engineeringgroup. His role would be to drive the suc-cessful product development of all por-table programs. He was described by oneof his colleagues as “the brightest young

Exhibit 8Brunners’s winning design, lateFebruary, 1990

drive because, “in an immature market,people thought they couldn’t get by with-out it.” Brunner and Sedmak also favoredan internal floppy for TIM. Some of ID’smodels had floppies built in, and somedidn’t. Brunner created most of theformer, and Ivester most of the latter.Building on Krakower’s partitioningscheme, Brunner found a way to incorpo-rate the floppy so that it took up very littleadditional space. The partitioning schemedivided the main PC board into twoboards and stacked them, leaving enoughspace under the keyboard for the drive.This proposal, which was considered dar-ing, was not ideal for manufacturing, butthe its space benefit could be argued tooutweigh the additional costs.

Design Review

Toward the end of February, an earlymorning meeting was held to review theID designs for TIM. Brunner, Ivester,Medica, Sedmak, Krakower, and Gee werepresent. The biggest decision on the tablewas whether or not to have an internalfloppy. The ID designs were comparedwith one another. Krakower also presentedhis model to show that the amount of ad-ditional space consumed by an internalfloppy and a large trackball could be de-signed to be very minimal, and summa-rized the advantages and disadvantages ofeach choice on a handout. (See Exhibit 7)While some of the others were skeptical,particularly about the “wasted space” ofpalm rests, Brunner thought the largetrackball and full-size palm rest was viable,and that ID could make it look right. Herecalled, “We in ID tend to like radical so-lutions, but most of the group still had afairly conventional image in their mind ofwhat the product should look like, whichwas Compaq’s LTE.” It required a stronglobbying effort on the part of bothBrunner and Krakower before people’sperceptions began to change.

Brunner quickly designed a new modelwith full-size palm rests. (See Exhibit 8)On March 6, Sculley was shown this andother ID models, as well as some of theTwister designs. From a pure designstandpoint, he liked some of the soft,sculptural forms of the latter, but for TIM,he preferred Brunner’s latest model using

Do Not

Cop

y or

Pos

t


12


rising hardware manager” in the company.Medica selected Jon Sedmak as the projectmanager for TIM. Sedmak also continuedoverseeing the Portable enhancementproject until it was cancelled in September.

For the duration of the PowerBookproject, both ID and the mechanical de-signers continued to report to Jordan.Product design remained centralized, withthe understanding that it would co-locatethe mechanical designers in the divisions,and that industrial designers would spendabout three days per week with theirprojects. Co-location was considered espe-cially critical for portables because, asMedica put it, “everything is so integrallyintertwined. You have to get portablepeople tightly coupled, all breathing thesame air, talking the same talk.” Strong re-sistance to co-location was expressed dur-ing the April-May time frame, but beganto settle down in the following months.Because of the tight coupling needed forminiaturized products, product design re-

ported to the portable division, yet servedother product divisions as well. (See Ex-hibit 9.)

Project Objectives

Sedmak declared that TIM was to be thesmallest, lightest, and quickest portableApple could do; everything else was nego-tiable. One manager recalled Sedmak’s di-rectives to him when he joined the team:“10,000 a month, presence in the market-place, place holder till the real product getsout there.” Brunner described Sedmak’sfocus: “He had his priorities straight in hismind: to get the product to market, makesure it was competitive in size and weight,and everything else was last.”

Acquiring Resources

In March, the team consisted of Sedmak,Krakower, Ivester, Gee, a senior EE, a co-ordinator, and a product design manager.Sedmak’s first priority was to try and getengineering resources. TIM was at priority

Macintosh Systems DivisionFred Forsyth

PortablesRandy Battat,

Senior Vice President

Imaging ProductsJohn Moon

DesktopEric Harslem

Corporate IndustrialDesign

Bob Brunner,Manager

Product DevelopmentJohn Medica,

Director

Product DesignRichard Jordan,

Director

Product MarketingNeil Selvin,

Director

Product Design Teams

High-end Products Design Team

Low-end Products Design Team

Imaging Design Team

Labs and Shops

Exhibit 9April 1990 Decentralization

Do Not

Cop

y or

Pos

t


13


level six (out of seven levels, where onewas the highest level). Other desktop pro-grams and Companion had higher priori-ties, and Companion had already takenwhat several people described as the “A-team” talent. TIM was able to secure eitherjunior people with little experience, orpeople from desktop who had no experi-ence in portables. As one manager put it,“You might have product design experi-ence, but it is very different when it comesto doing low weight and getting a verysmall package.” The mechanical designersbegan joining TIM during the May-Juneperiod. They included two product de-signers, a draftsperson, and a fourth whodid both. Part-time representation was se-cured from other departments.

Decision-Making

The major decision-making forum was theteam leaders’ weekly meeting withSedmak. Their first major decision was tofind a balance between battery life and theweight of the product. Every hour of bat-tery life increased weight by half a pound.A three-hour, seven-pound target wasagreed upon.

DESIGNING THE PRODUCT

During the next several months, IDworked with Krakower on board size andlayout, and with Apple’s user testing lab tofurther validate the ergonomic principlesand refine the details of the keyboard/trackball/palm rest design.

Product Core

During the January-June time period,Krakower put together a package of draw-ings that included outlines of the six PCboards, their connector/cable partitioning,module placements, and top level assem-bly sequence. He noted, “even though wedidn’t have the product design resourceson board, we couldn’t let the time slip by.”His work also reflected his broad interpre-tation of what it meant to be a technicallead or system integrator. The role wasambiguously defined as handling the logicboards and their interfaces. Typically,projects partitioned things very cleanly; thesystems integrator worked with the archi-tect to make sure the board designs wereimplemented while another group handled

the interfaces. By contrast, Krakower tooka broader approach, working on the inter-relationships among all the componentsinterfacing with the boards. He later com-mented on this approach: “This wasApple’s first real attempt at shrinking thedevelopment schedule to less than half thatof previous first-generation products. Theneed for the technical lead to take a strong,multi-disciplined, proactive approach toengineering was certainly crucial then andis becoming more and more importantwith shorter design cycles and limited re-sources. The technical lead has to makesure there isn’t a cross-functional discon-nect somewhere.”

When the product design managerjoined the team in late April, he felt thathis mechanical engineers should have theopportunity to design from scratch.Schooled in the Apple tradition of design-ing rugged, snap-together plastics, hebrought this same approach to designingportables. After a month’s effort, his prod-uct design team concluded that Krakower’sapproach was indeed the only way to do itwithout increasing the size and weight ofthe product.

Design and Test

Design alternatives began to be tested inApril, and continued through August in aconcentrated series of seven usability tests-design cycles. The process was iterative; asIvester noted, “It was designing to test andtesting to design.” Learning from oneround of tests was incorporated into designchanges which were tested in the nextround. Gee recalled, “Brunner’s organiza-tion was coming up with these differentdesigns. We would sit and brainstorm withthem about what they had learned lasttime and what they wanted to do nexttime.”

Ireng Wong of the user test lab ranmost of the test research. She encouragedthe designers and other team members tosit behind one-way mirrors, to watchvideotapes of the tests, and to attend userfeedback sessions. Tests were conductedwith Apple employees whose hand sizeranged from very large to very small, andwhose height ranged from very tall to veryshort. Wong wrote up reports after eachround of testing.

“This was

Apple’s first real

attempt at

shrinking the

development

schedule to less

than half that of

previous first-

generation

products”

Do Not

Cop

y or

Pos

t


14


The test results clearly supported thecore concept of centering a large trackball.Wong, Ivester, Krakower, and their col-leagues worked on many design details,such as: size of trackball; number oftrackball buttons; placement of trackballbuttons; depth and width of the keyboard;width of the palm rest; whether the palmrest should be flat or slanted, and, if thelatter, at what angle. Discussions were ex-tensive and sometimes heated. Ivesterpointed out that:

The biggest challenge was to designtrackball buttons which would be easy tounderstand, physically comfortable, andefficient. The ongoing user testing provedvery helpful in finding the right balance,since these goals conflicted to some ex-tent. For comfort and efficiency, buttonsboth above and below the ball seemed towork well. Users could move the trackballwith one thumb and click the top buttonwith the other, leaving the hands posi-tioned over the home row of the key-board. For fine pointing work, it was bet-ter to drop the hand down, moving theball with the index finger and clicking thebottom button with the thumb of thesame hand. But having two mouse but-tons was not the Macintosh way, and itcaused some confusion. To counter theconfusion, we needed to communicatethat the buttons have the same function.

Several designs were tried, and multipleinserts were built that could be placed into

the test models. (See Exhibit 10.) Peoplefound the three-button designs to be espe-cially confusing. Ivester explained the finaldesign. “The two buttons were designed tooccupy the same amount of space, with thelargest button surfaces placed where fin-gers naturally reach when using the ball.The bottom button’s lower edge is curvedto help it clear the palms when typing. Thetop button’s top edge has a similar curve tohelp communicate that the buttons havethe same function.” The buttons them-selves were designed through ID models,and sanded to get them to feel just right.

When the machine was flat on a desk,people felt as if they were typing downhill.This sensation was accentuated when thepalm rest was tilted. The designers revisedthe palm rest to a flat position, while alsoshortening the keyboard to reduce thereach. In addition, they incorporatedwheel-like feet or lifters in the rear cornersof the machine. (See Exhibit 11.) Whenthe notebook was on a person’s lap, thelifters could be left folded up, but theycould be rotated down to raise the key-board to a more comfortable position fortyping on a flat surface.

Portability and Object Value

These months of iterative testing anddesign led to an evolutionary shift in themeaning and execution of the productconcept. Brunner looked back upon theemergence of the themes of portability andobject value:

Exhibit 10Trackball inserts for user testing

Do Not

Cop

y or

Pos

t


15


To be competitive, we had certain sizeand weight specs we had to meet. Thiswas driving the TIM program. We wereassuming that it was just a catch-up ma-neuver, not a way of moving ahead.Where we felt we could win and takeleadership was to by integrating hard-ware, software, and industrial design, be-cause those are the things that we have ortake more seriously than anybody else inthe industry.

But we were still looking at TIM as asmall computer. We made models of asmall computer, but we weren’t thinkingusability scenarios or emotional content.On the Companion project, we had be-gun thinking about usability, and whatpeople really want in a portable product.As we built and tested more models forTIM, we started thinking more about itsenvironmental aspects. As we got furtherinto TIM, we could begin to see real ad-vantages to what we were coming upwith, and we began to develop this ideaof true portability versus the mobiledesktop. This is where we began to buildupon the idea of integration as beingmore than hardware and software; itmeant viewing the product from a port-ability point of view.

Portability referred to usability in a varietyof mobile environments, which mightrange from a dining table, a lap, or an air-line tray to a hotel bed. The design had tohave the flexibility to deliver consistentperformance across environments. Thedesigners also began to recognize the us-ability advantages of a full-size palm restand its implications for portability.

Portability was very intertwined with asecond emergent theme, that of objectvalue. Notebook users didn’t just sit down,flip a switch, and start working. They car-ried their notebook around with them,used it on their laps, they opened andclosed its case, and took batteries in andout. A notebook was not a piece of busi-ness equipment that someone operated; itwas a personal object with which theyformed a relationship. Bruce Gee recalled:“When we first started on the program, itwas, ‘Build a small notebook computer’. Itwasn’t, ‘Build the most ergonomic note-book computer’. Through the use of

testing, we began to push ergonomics as akey differentiation. So rather than being asmall Macintosh, it became a great note-book that has an easy-to-use operating en-vironment, and by the way it is aMacintosh.”

Shrinking the Design

While the designers and researchers wereiteratively evolving the ergonomics ofportability, Sedmak, Krakower, and themechanical designers were shrinking thedesign, millimeter by millimeter. Theyzealously rooted out every extra millimetertaken up by components and plastics. Thisapproach came to be known as “millimetermadness.” While everyone agreed thatTIM’s footprint and size had to shrink,Battat felt that this millimeter approach“missed the whole emotional connection.”Brunner cautioned against taking it too far.In a memo to the team in May, he ac-knowledged the need to reduce TIM’s sizebut pointed out that, “In the end, few buy-ers will care about a few millimeters offootprint, but rather the perception of sizeand mass in their eyes … ID [has the abil-ity to alter] perceived size and mass … [so]ID may come back [to you] with some de-signs that increase slightly the X or Y over-all dimensions, but will decrease the percep-tion of size to the human eye.” [emphasisadded] In his view, the customer was lessconcerned with millimeters than with thequestion, “Does it fit in my briefcase?”

Led by Krakower’s PCB componentminiaturization efforts, the product de-signers took the width down from about10 inches to 9.3 inches, and reduced theheight by about .25 inch. The result, inSedmak’s view, was that by mid-August,“We finally figured out the right size andwhat we needed to do.”

Exhibit 11Rear lifters to tilt keyboard

Do Not

Cop

y or

Pos

t


16


The approach taken by the first productdesign manager was to get “first article”parts off of tools as quickly as possible, andthen refine them. Brunner observed,“Apple is normally a front-ended loadedprocess. We usually do several rounds ofengineering models and refinements be-fore going to tooling. In this case where wedid it the other way, we missed some bigthings, and had to jump back into the pro-cess in a big way after tooling.” Few earlyindications of the scope of these issueswere given by “first builds,” or engineeringprove-outs, in October 1990. First buildswere expensive prototypes that were ma-chined to look and feel like the finalproduct.

New Product Design Manager

Medica replaced the product design man-ager with Tom Bentley in November.Bentley not only knew manufacturing, butalso had an appreciation for design, for thelook and feel of a product, and was experi-enced with leading edge design methodol-ogy tools. Upon his arrival, Bentleythought TIM looked like the “project fromhell.” “Management was putting their en-ergy into other programs. The directorswere micro-managing design and tellingthe design team how to implement. Mostof the designers had never done a plasticpart before. Everything was line-to-line fit.I got there two months before first articles,and first articles came back with all theseissues appearing out of the woodwork.”

Prototype Builds

By line-to-line fit, Bentley meant that theinexperienced designers had not takenmanufacturing tolerances into account.Therefore, some of the plastic parts did notfit precisely or line up as they should havebecause of tolerance stack-up. But it was notuntil parts were made from tooling, ratherthan machined, that it became evident thatmany refinements were still needed beforeTIM met Apple’s design standards. First ar-ticles were the first of some fifteen proto-type builds over a seven-month period; eachbuild revealed new issues.

150 Items

By late February 1991, Bentley compiled alist of nearly 150 items that were problems

Low-Cost TIM

Apple managers originally anticipated thatTIM would have a low-cost (LC) version,differentiated by its microprocessor; TIMwould use the 68030 chip, and TIM LCthe 68020. After exploring new LCD tech-nologies in the May-June period, the teamdecided to delay the schedule by fourweeks so that it could offer both the oldand new technologies, with a price differ-ential of around $1000. TIM would havean active matrix display and EL backlight-ing, and TIM LC would use a passive dis-play with either no backlighting or a lessexpensive lighting technology. In themeantime, Gee noted, “We learned thatpeople did not want the 020, just the 030.”Gee discussed the implications of cus-tomer-driven design changes. “If you arebuilding a line, you want something in themiddle that is less than the top. We startedwith 020 in the middle and 030 at the top.But when everyone wanted the 030, howcould we differentiate two 030 machines?By processor speed and displays. We hadto build a faster 030 for the high-end ma-chine, and at the time we didn’t know if itwas feasible.” The team decided that theactive matrix 170 model would be usedwith a very powerful 25 Mhz 68030 chip.Low-cost TIM would have a passive dis-play, and use a less powerful 16 Mhz68030; it became the model 140.

DESIGN QUALITY AND TIME TO MARKET

Because of the intense pressure to minia-turize the notebook and get to market assoon as possible, it became obvious to ev-eryone that compromises would have to bemade in its design. The pressure was par-ticularly intense on mechanical engineer-ing, which was continually on the criticalpath. To this was added Gassee’s very pub-lic view that mechanical engineers in theU.S. lacked the ability to miniaturize torequisite levels of size and quality. Thus,mechanical engineering’s needs tended todominate TIM’s development. Its re-sponse to other group’s needs was, asBrunner put it, “We’ll do it if we havetime.” For example, the quick, low-riskway to meet shielding requirements meantthat parting lines were not what ID wouldhave liked.

By late February

1991, Bentley

compiled a list

of nearly 150

items that were

problems or less

than satisfactory

Do Not

Cop

y or

Pos

t


17


or less than satisfactory. Many of the itemswere fit and feel issues with the productand enclosure parts: doors and latches thatwere too difficult to open; hinges thatcreaked or latches that rattled; a sharp edgeon the I/O door; a caps lock key that didnot click down or have a light. Some wereaesthetic: the power plug hole was toolarge and ugly, the gap around the trackballbuttons too large. Some were functionalissues such as connector interferences, orEMI issues that required shielding,grounding, gaskets, or some other fix.Some were significant for product integ-rity, such as the addition of a shroudaround the display to seal up the unitwhen closed, and the addition of ribs tostrengthen the thin-wall structure in cer-tain key areas.

Dissatisfied with the quality but mind-ful that tooling was already being cut,Bentley asked Sedmak and his team lead-ers which five or ten items needed tochange in order to get the product out thedoor. People had difficulty deciding.Bentley insisted that they prioritize theproblems, but in every meeting, the list ofnumber-one-priority items seemed togrow rather than to shrink. In the end, theteam decided to change all 150 items.

Resources

The scope of what had to be done in sucha limited time was way beyond what fiveproduct designers could do. TIM wouldrequire a great infusion of resources.Medica explained what this meant:

It was not until well after thePowerBooks had been introduced thatProduct Design for portables actually re-ported directly into our group. Duringdevelopment of the 170, ProductDesign—which includes the actual me-chanical product designers, the toolingengineers and the CAD designers—re-ported to a functional engineering man-ager. He managed all of Apple’s productdesign across imaging, desktops and por-tables. I had to influence him to steal re-sources from other groups, and havethem come back and work for us. It wasapparent to him that he had no choicebut to take resources from Companionand move them over to PowerBooks.

For six-to-eight weeks in the March-April1991 period, TIM commandeered the re-sources of the entire portable design orga-nization—22 engineers. Bentley describedit rather colorfully from his perspective:

I stopped the prima donna group [i.e.,Companion] that had been told theywere the center of the universe. I said,“Stop. Sorry, Rodger, but I am takingyour entire team. Sorry, John [themanager of the PowerBook 100, a jointventure with Sony that began in August1990], I am taking your entire team. I amgutting your department and taking ev-eryone over until we get this done. Weare going to learn what it takes to shipthe product no matter what the costs.”

An incremental million dollars in toolingcosts was incurred to make hundreds oftooling changes. Bentley believed thatthrough this experience, “People gained areal good sense of portables. If you haven’tdesigned a portable before, you are notused to the things that happen. You can’tdesign with normal clearances. All themoving parts on PowerBook were rede-signed at least four times.”

NOTEBOOK DESIGN AND MANUFACTURING

The team’s broadest design challenge waslearning and understanding what it meantto design a portable as compared to a desk-top product. The PowerBook’s develop-ment was an extended lesson in learningwhat it meant to design a seven-poundnotebook from the ground up.

Apple DFM Philosophy

Apple normally designed thick wall plasticparts that snapped together easily; itsdesign-for-manufacturability (DFM)policy was to use no screws. It designedzero-draft parts, which meant it routinelyinvested in expensive tooling that yieldedbeautifully molded products. These designpractices also made manufacturing’s lifeeasier, but added weight and size to prod-ucts. Because the Portable’s designers hadabided by these rules, their product be-came much bigger and heavier than themarket would accept.

Because of its need to meet muchmore stringent size/weight requirements,the PowerBook was the first Apple

“All the moving

parts on

PowerBook were

redesigned at least

four times”

Do Not

Cop

y or

Pos

t


18


product that was allowed to violate theserules. It could not be designed with desk-top clearances. Heavy plastic structuresand thick walls were out of the question.Thin walls were necessary, as were posi-tive fasteners. TIM had 50-75 parts, andover 40 screws. Routing complexity of PCboards was much greater than for a desk-top; boards were more closely stacked andrequired header connections. TIM hadmore subassemblies, and much morelabor content.

Manufacturing Expectations

Sedmak described the “established rhythmat Apple” as one in which “it took three-to-four years to do a product; the designteam waited nine months, and did another.A three-month slip was not that big a deal.It was important, but it wasn’t, ‘boy, we’vegot to make it.’ Now we are having tomove at a speed that is straining the entirecorporation. Manufacturing sites are hav-ing to move at a speed that they are notcomfortable working at because theirinfrastructure was put there by a differentdesign cycle.’’

The Fremont plant was designated tobuild the PowerBook. The people therehad never built a portable. When Bentleytook a prototype over there, they said,“You’re crazy. We can’t build this thing. Ithas 47 screws in it.” Bentley recalled, “Itwas exactly the same thing with the manu-facturing organization as it was with thedesign organization. People didn’t knowwhat the rules were, what the standardswere. Apple is 100 percent snap-togetherparts, which means we can use 80 percenttemps on the line.”

One way that Sedmak dealt withmanufacturing’s discomfort with portableswas to bring competitive products to teamleader meetings, and proceed to take themapart. They all had thin walls and positivefasteners. Compaq had even resorted togoing to a Japanese company to assembleits first high-volume model. When theLTE was taken apart, Fremont manufac-turing took it as a kind of challenge. Resis-tance gradually lowered, and there wasacknowledgment that, in Sedmak’s view,‘’We had no choice. If we wanted to be inthis business, we had to do this.’’

Initial sales

forecasts were

low—perhaps

100,000 units

the first year,

and, at most,

150,000

POWERBOOK INTRODUCTION

By the end of 1990, marketing was work-ing on TIM positioning and pricing. Earlyin 1991, it began thinking about namingTIM, and about treating it differently froma Macintosh. Based on formal research,marketing began calling it PowerBook.

Market Projections

Initial sales forecasts were low—perhaps100,000 units the first year, and, at most,150,000. That changed very dramaticallyand very quickly during the spring andsummer. Even with first plastics, Gee re-called, “the level of excitement was incred-ible. People were just amazed that wecould get something that small. Once theygot their hands on it, we had to pry theirhands loose at the end of the presenta-tion.” His boss, Neil Selvin, elaborated onwhat was coming back from usabilitytesting, focus group presentations, andcustomers of the early prototypes:

What hit people was the ergonomics.The DOS clones all looked alike. Theyhad squared corners and looked clunky.They had clip-on mice or trackballs thatyou had to use with a long stringy cable.Control over the screen from one appli-cation to another was very weak. Therewas an immediate attraction toPowerBook. The center-mount trackball,the palm rest, the keyboard’s beingmoved back, the slope and tilt of thescreen, the slider bars for adjusting key-board tilt—all these little things thatmade it comfortable.

Ramp Challenge

As a result, over a period of several weeksbeginning in March 1991, marketing esti-mates for quarterly sales went from 50,000to 75,000, then 100,000, and finally175,000. Even the 50,000 estimate caughtSedmak’s attention because, compared tothe Portable, this meant a huge ramp.Sedmak and Bentley began holdingweekly “ramp challenge” meetings, prima-rily with the factory and Apple’s criticalsuppliers. Working through all the prob-lems took nearly a year, and for manymonths Bentley had one of his productdesigners with a beeper on the line at alltimes on both shifts. He would beep

Do Not

Cop

y or

Pos

t


19


Bentley at home or wherever he might be;if they couldn’t get the line up in half anhour, they called in the manufacturingmanager.

Verification Testing

Some of the most important problemswere revealed through Production Verifi-cation Tests (PVT) at the plants duringJune and July. Problems showed up withvirtually every subassembly Sedmak de-scribed why PVT was so important:

It is fairly typical of the electronics busi-ness that when you go through yourbuilds, you are only building a couplehundred units, and usually the subas-semblies are being put together by engi-neers. They often build them on proto-type lines and use premium componentsbecause you haven’t turned your vendorloose yet so that you can see where theprocess really goes. So when you finallydo turn them loose, and build them on aregular line with regular productionpeople, you get surprised.

By June 1991, however, TIM was a first-level priority, had enormous visibility inthe corporation, and it got whatever help itneeded. Sedmak described the pressure as“ferocious.” “You find problems and youhave no options and you are up against thewall. Our active matrix display went intoproduction in May so that we would notbe short. But if we found a problem inJune or during PVT test, we already had30,000-40,000 displays in inventory. If youwant to change something then, you aretalking big time scrap heaps.”

Product Launch

The PowerBook 140/170 was announcedOctober 20, 1991. Within two weeks,Apple was backlogged by 140,000 units.During the first 12 months in the market,over 400,000 were sold, generating overone billion dollars in revenue for Apple.(See Exhibit 12.)

RETROSPECTIVES AND LEARNING

Apple had been concerned that in doing anotebook, it would have to make a flimsy,non-Apple quality product to get to thesize and weight required by the market.When its engineers looked at competitors’

portables, their response had been, “Wecan’t do a product like that. That is notApple quality.” Brunner felt that TIM hadbeen “a real learning experience for Applein coming to grips with what traditionalApple quality means in this time to marketenvironment. TIM helped us to do thiseven though it was occasionally painful.”He elaborated,

During development, time was king.When we got some of the first units back,we realized that there were some areasthat had been cut. We went back and did alot of refinement that we normallywouldn’t have had to do in order to meetApple quality. I think initially we let thetiming push us and then we came backlater and said, “wait a minute, this is toofar on the other side of the line. We’ve gotto come back and fix some of these thingsbecause it just won’t be good enough.” Itwas a valuable lesson for every portableprogram I have been involved in since.We definitely learned our lessons aboutwhat is good acceptable quality and whatare the compromises that we have tomake to get something out.”

Design refinement and correction of the150 items was the difference between, asMedica put it, “Something that I would beproud of versus a piece of junk.” Retro-spectively, he mused, “If we hadn’t donethem, I believe the PowerBook would not

The PowerBook

140/170 was

announced

October 20,

1991. Within

two weeks,

Apple was back-

logged by

140,000 units.

Exhibit 12Production model, PowerBook140/170

Do Not

Cop

y or

Pos

t


20


have been nearly as successful. I think thePowerBook is about the whole customerexperience—the ease of use, the ergonom-ics, the elegant design, the fit and finish,the performance.” He also believed thatthe long-term effects of TIM’s pulling ma-jor resources during the second calenderquarter of 1991 had been beneficial. Heexplained,

At the time, people on the Companionproject felt they were being significantlyshortchanged. It wasn’t a popular deci-sion, but we had no choice because wehad to deliver the 100, 140, and 170 inOctober.2 In hindsight, it actually madethe Duo [the Companion team’s firstproduct] a better product because mostof the Duo team had never done a por-table before. They had only done aneight-to-ten piece part design for desk-tops. On TIM, they got involved in afifty-part piece design with all the sec-ondary operations, vendor-related issues,and factory-related issues when every-thing had to come together. Theylearned a hell of a lot. It was a great train-ing ground, and subsequently, the Duodidn’t go through nearly the pain that wedid on the original 140 and 170.

Bentley believed that design—what he de-scribed as the look and feel of the prod-uct—was responsible for 95 percent of the“people essence” of the PowerBook. Dur-ing its first year in the market, he went to atrade show where he took apart and playedwith 72 portables; he found PowerBook tobe unquestionably the best, 20 percentabove the rest.

Battat shared his view of the importanceof design to PowerBook’s success. “I thinkcustomers have an emotional attachmentto PowerBooks, and that has nothing to dowith size and weight and meeting theirperformance expectations. In some cases, itmay be bigger than it needs to be becausethere are curves in it. But those curves arewhat helps generate emotions. Unfortu-nately, you can’t prove it.”

Organizational Maturation

In Bentley’s view, the responsibility of theproduct design manager in a program wasto look at problems and escalate those thatpeople would care about. But on TIM,

We didn’t have the experience level toflag and escalate such issues. Peoplecouldn’t look at a hinge and say, “We aredoing to have problems with that hinge,”or, “That latch doesn’t look like it’s goingto work.” Since then, we have built thesecapabilities. The marketing guy nowwould take things out and test them, andcome back and say, “These 20 of the 150need to change, and these 70 peopledon’t care about.”

The idea of empowering and supportingproject managers was increasingly in theair during and after the 1990 reorganiza-tion. It took time, however, before it wasunderstood well enough to become trans-lated into practice. Used to the “championmodel,” many senior middle managers in-terpreted “supporting” as meaning thatproblems which could not be resolved bythe project manager should be brought up-stairs to them. This had been formalizedunder Gassee; if the senior managers couldnot agree, then the matter had to be re-ferred to Gassee. One manager com-mented on his frustrations with this pro-cess, “You could never go to John Louiswith a problem; you always had to go tohim with a solution. You had to get con-sensus before you went in there, or hewould rip you up. Then you had to get onhis calendar.” After Gassee’s departurelater that year, directors still resolved a lotof issues, but did so more quickly, whileproject managers began assuming moreauthority.

PowerBook’s Impact on Design

In Brunner’s view, “The two most power-ful things about the PowerBook design areits design for portability so it can be usedin a variety of environments, and its objectappeal.” He recalled that, “IndustrialDesign was driving the notion of ‘objectvalue’, and got definite support from prod-uct marketing. Most manufacturers weretreating their portables as extensions oftheir desktop environment, as businessequipment, rather than as personal objects.A lot of our focus on the product detailingcame from our sense of its object appeal.”This experience with portability had alsoaffected how Apple thought about designlanguage. Brunner continued,

“I think the

PowerBook is

about the

whole customer

experience—

the ease of use,

the ergonomics,

the elegant

design, the fit

and finish, the

performance”

2. The PowerBook 100 was ajoint venture with Sony. It wasa miniaturized version of thePortable, using a 6800 chipand weighing 5.5 pounds.Apple lacked the engineeringresources to do this low-endnotebook concurrently withTIM, TIM LC, andCompanion.

Do Not

Cop

y or

Pos

t


21


Prior to the PowerBook, we looked atgood design as being policy. It was nottailored to the products or the individual,but an aspect of Apple philosophy. Theresult was homogenous design. So whenwe did the Portable, we took the sameApple rules and applied them. But as wegot into the PowerBook and startedthinking about portability, we were usingdesign to move the product out in a cer-tain direction toward certain people. Thishas changed our thinking about designlanguage. We have more freedom now interms of form, color, detail, and layout todefine, to move towards a certain cus-tomer group.

PowerBook was designed primarily in thelanguage of Snow White. Brunner ob-served that it “looked like a module for thedesktop,” yet its curves and dark colorbegan the transition toward “Espresso,”Apple’s new design language. Brunner de-scribed “Espresso”:

Its global themes are curvature, centralsymmetry—or an axis of symmetry, butit’s usually a central axis—complex sur-faces that catch the light and are comfort-able to hold (like the new mouse), surfacetearing and bulging, overt expression offunction, and general detail consistency. Atthis global level, the design language itselfserves as a unifying theme to define Appleas a whole. Espresso is also flexible andscaleable. That means that the attributesand goals of the product itself define thefinal implementation of the language.

Brunner also added that, “Randy [Battat]asked me not too long ago whether wecould do another PowerBook and I saidno. It would be much harder because atthat point we had nothing to lose. Nowthat we are successful, we have to say,‘Wait a minute, that is radical. We better goout and test that.’ It’s healthy to find a bal-ance, because we could have fallen on ourface if the concept had been too radical.”l

Do Not

Cop

y or

Pos

t


22

than hardware and software; it meantviewing the whole product from a port-ability point of view.”

This experience with portability has af-fected how Apple thinks about design lan-guage. “Prior to the PowerBook, welooked at good design as being policy. Itwas not tailored to the products or the in-dividual, but an aspect of Apple philoso-phy. The result was homogenous design.So when we did the Portable, we took thesame Apple rules and applied them. But aswe got into the PowerBook and startedthinking about portability, we were using

design to move the productout in a certain direction to-ward certain people. This haschanged our thinking aboutdesign language. We havemore freedom now in termsof form, color, detail, and lay-out to define, to move to-wards a certain customergroup.”

This freedom is particularlypronounced in the “outerring” of Apple’s design lan-

guage. “We used to be very retentive aboutdesign language, but now I look at it likerings of a tree. We have core systems,monitors, and displays that need to be veryconsistent. Then we move out into the sec-ond ring to things that are related to theMacintosh brand, such as printers and por-tables. In the third ring are consumer prod-ucts that are not directly related to comput-ers; these we should design for the specificapplication and our target customers.”

One of Brunner’s major challenges is toaddress Apple’s triple messages of innova-tion, cost, and time to market. To meetcost pressures, “we have had to learn toscale our process.” By that he means thatID must leverage its time/resources wisely;it cannot overinvest in incremental prod-ucts, or those having limited market po-tential. To insure innovation, “you have tokeep the pressure on, do design studies, doearly explorations. And if you want thingsfast, then you have to do more work in ad-vance. A high percentage of that will windup in the dumpster, but that is the price ofinnovation. And you need to allow forquality think time —‘grave time’ — on theproject itself.” l

Appendix: Bob Brunner, Director, Industrial Design

When Robert (Bob) Brunner wasfirst approached to take over managementof Apple’s ID group, Apple had only fivedesigners, and did 90% of its design workthrough outside consultants. Brunner,who was then 31 and consulting to Apple,didn’t find this appealing, nor did he thinkit right for Apple unless the charter be-came one of forming a strong internalgroup to chart Apple’s strategic directionin design. When this charter was accepted,he joined the company in November,1989. “I came to Apple because it has greatproducts, and is driven by product design.People here are very visuallyoriented. They rally aroundobjects, and the companycenters around physical prod-ucts. Design is embedded inthe culture. I sometimes say Ihave a staff of 10,000.”

In fact, Brunner now has astaff of 21 who do about 60%of Apple’s design work. InsideID people are used for direc-tion-setting products. A feware assigned to a specificproduct line, and have a strategic responsi-bility to take that line somewhere. Theothers are part of the model shop, or of thedesign studio, a captive, creative pool oftalent that crosses all areas. By thus lever-aging his design resources, Brunner aimsto keep headcount down and creativity up.Apple has long-term relationships withnumerous external designers who are usedfor follow-on products, as a source of freshideas, and as a problem-solving resource.

The PowerBook, for which ID beganmaking sketches and models a few weeksafter Brunner’s arrival, was a learning ex-perience for everyone associated with it.“We had certain size and weight specs wehad to meet, and that was driving the pro-gram. We were assuming that it was just acatch-up maneuver, not a way of movingahead. But as we got into it, we began todevelop this idea of true portability versusthe mobile desktop. We felt we could winand take leadership through integratinghardware, software, and industrial design,because those are the core things that wehave and take more seriously than anybodyelse in the industry. We began to buildupon the idea of integration as being more

Bob BrunnerDirector, Industrial Design

apple powerbook design quality & time to market 994023-pdf-eng

Documents