A P U B L I C AT I O N O F T H E C O M P U T E R H I S T O R Y M U S E U MW W W. C O M P U T E R H I S T O R Y. O R G

S E P T E M B E R 2 0 0 3

CORE 4.1

ANNOUNCING OURALPHA PHASE!

JOHN C TOOLE

EXECUTIVE DIRECTOR & CEO

C O M P U T E R H I S T O R Y M U S E U M

With the extraordinary efforts andsupport of so many people this year, wehave reached another major milestone.The Museum initiated its Alpha Phaseon June 2 at 1401 N Shoreline Blvd!This marks the official opening of a newVisible Storage display area, thededication of the Hahn Auditorium, andthe beginning of our public presence inour new home. Yet there’s much moreahead to realize our dream—plans arealready underway for the Museum’sBeta Phase, and Releases 1.0 and 2.0over the next several years.

It’s exciting to see the opportunities ournew home has given us. In the newVisible Storage, for example, you willstill find many favorite artifacts from thedisplay at Moffett Field. However, wenow have an entirely new look, labelsfor all of the items, and about doublethe number of artifacts in a largergallery for a more completerepresentation of computing history. Agreat online virtual visible storage canbe found at www.computerhistory.org. Iknow you will be pleased to see how weare bringing that story to the world.

We also opened our new HahnAuditorium and multi-purpose room,named in honor of the Hahn family, ourlargest donor to date. Eric Hahn is alsoa trustee who has helped us growdramatically and who served as the firstchair of the Development Committee inCalifornia. This space will become acommunity gathering place forcomputing history enthusiasts, hosteverything from history lectures tomajor events, and allow us to recordimportant events for posterity.

Our new building has been drawing theinterest of corporations, attracting manynew volunteers, and allowing trustees,staff, and volunteers to productivelywork together in one environment. SinceJune, we have moved most of ourcollection from offsite storage intounused portions of the Museumbuilding, and we may soon be offeringspecial member tours to explore thedeep recesses of our great collection.

Although accomplishments have beengreat, the economic climate has beenextremely challenging. We need yourhelp in finding people and organizationsto support us as we grow, and arelooking for volunteers to help in manyareas, including development. We alsoreorganized early this year inanticipation of next year’s economicsituation, and have consolidated somefunctions. The staff has beentremendous during this difficult time ofchange, and you will see some of theirtitles have changed.

With our new presence in MountainView, you can tangibly see how we areable to grow into a great museum. Yetour programs take support anddedicated people. Please considerincreasing your annual campaignsupport and making a capital campaigngift. In addition, we have just kicked offa great corporate membership programand have created recognition walls forall to see. For information on themember programs and namingopportunities please contact KarenTucker.

Be sure to read the report on Museumactivities. While we’ve been preparing toopen our great building, our publicprograms have been spectacular. Theawesome lectures and events we planand support have truly brought to lifethe meaning of preserving the stories ofthe information age.

The launch of our Alpha Phase is a giantstep forward for the Museum. We nowhave an unparalleled public presencethat will be built in phases over timealong with great programs and accessto the largest collection in the world.Help us to grow and let us know howyou enjoy your Museum!

24

CORE 4.1September 2003A publication of the Computer History Museum

VISIONT O E X P L O R E T H E C O M P U T I N G R E V O L U T I O N A N D I T SI M PA C T O N T H E H U M A N E X P E R I E N C E

MISSIONT O P R E S E R V E A N D P R E S E N T F O R P O S T E R I T Y T H EA R T I F A C T S A N D S T O R I E S O F T H E I N F O R M AT I O N A G E

EXECUTIVE STAFF

John C TooleE X E C U T I V E D I R E C T O R & C E O

Michael FalarskiV I C E P R E S I D E N T O F O P E R AT I O N S & F A C I L I T I E S

Kirsten TashevD I R E C T O R O F C O L L E C T I O N S &E X H I B I T I O N S

BOARD OF TRUSTEES

Leonard J Shustek, ChairmanV E N C R A F T L L C

Sally M AbelF E N W I C K & W E S T L L P

David L AndersonS E N D M A I L

C Gordon Bell M I C R O S O F T C O R P O R AT I O N

Peggy Burke1 1 8 5 D E S I G N

Lori CrawfordI N F I N I T Y C A P I TA L L L C

Andrea Cunningham C I T I G AT E C U N N I N G H A M

William Davidow M O H R , D AV I D O W V E N T U R E S

Donna Dubinsky H A N D S P R I N G

David EmersonC L A R E N T C O R P O R AT I O N

Eric HahnI N V E N T U R E S G R O U P

Gardner C HendrieS I G M A PA R T N E R S

BOARD OF ADVISORS

Gene Amdahl

William AsprayI N D I A N A U N I V E R S I T Y

Robert BrodersonU N I V E R S I T Y O F C A L I F O R N I AB E R K E L E Y

Paul CeruzziN AT I O N A L A I R & S PA C E M U S E U MS M I T H S O N I A N I N S T I T U T I O N

Federico FagginS Y N A P T I C S

Samuel Fuller A N A L O G D E V I C E S

James GrayM I C R O S O F T C O R P O R AT I O N

PUBLICATION STAFF

Karyn WolfeE D I T O R

Karen (Mathews) TuckerV I C E P R E S I D E N T O F D E V E L O P M E N T,M A R K E T I N G , & P R

Michael R WilliamsH E A D C U R AT O R

Peter HirshbergG L O S S . C O M

Charles H (Chuck) HouseI N T E L C O N V E R G E DC O M M U N I C AT I O N S G R O U PD I A L O G I C D I V I S I O N

Dave House

Christine HughesA C H I E V E M E N T P L U S

John Mashey C O N S U LTA N T

Ike R NassiF I R E T I D E

Suhas PatilT U F A N

Bernard L Peuto C O N C O R D C O N S U LT I N G

F Grant SaviersP R I VAT E I N V E S T O R

John Shoch A L L O Y V E N T U R E S

Stephen L SquiresH E W L E T T - PA C K A R D C O M PA N Y

Pierluigi Zappacosta

Burge JamiesonS I G M A PA R T N E R S

Randy KatzU N I V E R S I T Y O F C A L I F O R N I AB E R K E L E Y

Steve KirschP R O P E L S O F T WA R E C O R P O R AT I O N

David PattersonU N I V E R S I T Y O F C A L I F O R N I AB E R K E L E Y

James N PorterD I S K / T R E N D

Eric SchmidtG O O G L E

Copyright ©2003, Computer History Museum. All rights reserved. The Museum is an independent501(c)(3) organization, FID #77-0507525. Printed in the USA.

Computer History Museum1401 N Shoreline BlvdMountain View, CA 94043-1311, USA+1 650 810 1010+1 650 810 1055 (fax)

WWW.COMPUTERHISTORY.ORG

Submission guidelines: www.computerhistory.org/coreor write core@computerhistory.org.

Cover: Celebrating the opening of the Alpha Phase of the Computer Histor yMuseum at 1401 N. Shoreline Boulevard in Mountain View, California! Seear ticle on page 2.

I N S I D E F R O N T C O V E RANNOUNCING OUR ALPHA PHASE!John C Toole

2COMPUTER HISTORY MUSEUM: ALPHA PHASEKirsten Tashev

I N S E R TGUIDE TO VISIBLE STORAGE

6THE MCM/70 MICROCOMPUTERZbigniew Stachniak

1 3RECENT ADDITIONS TO THE COLLECTION

1 5PRESERVING HISTORY: THE SDS SIGMA 5FINDS A NEW HOMELee Courtney

2 4REPORT ON MUSEUM ACTIVITIESKaren Tucker

2 8MEMBERS & DONORS

2 9UPCOMING EVENTSCONTACT INFORMATION

O N T H E B A C K C O V E RMYSTERY ITEMS FROM THE COLLECTION

IN THISISSUE

BACK

C O R E 4 . 1

15

6

2

The Museum reached a key milestonewith the opening of its Alpha Phase onJune 2, 2003. At an open housecelebration held at the new building, theMuseum unveiled a 400-person HahnAuditorium and meeting space, donoracknowledgment walls, and a 9,000-square-foot Visible Storage exhibit area.The opening was a great success andwas attended by about 600 people,including city officials, Museummembers, trustees, staff, contractors,and guests. The celebration began witha ribbon cutting ceremony andpresentations by Executive Director &CEO John C. Toole and Museumtrustees, and was followed by tours ofthe new Visible Storage exhibit area anda reception.

In October 2002, the Museumpurchased its landmark building at1401 N. Shoreline Boulevard inMountain View, Calif. Built in 1994, thebuilding was state-of-the-art for its timeand its open concept design lends itselfwell to the Museum’s future plans. But,several key upgrades were necessary totransform the building from office tomuseum in compliance with publicassembly building codes. Renovationswere begun shortly after purchasing thebuilding and included such requirementsas fire walls, mechanical and safetyupgrades to accommodate a largeroccupancy capacity, and structuralupgrades to support both the increasedoccupancy of potential visitors as wellas the significant weight of some of theMuseum’s most historic artifacts.

THE HAHN AUDITORIUM

The renovations also allowed theMuseum to significantly improve itsfacilities for ongoing public programs.The Museum’s popular speaker serieswill now be held in the new 400-seatauditorium. The Hahn Auditorium,named after major benefactors ElaineHahn and Museum Trustee Eric Hahn, is equipped with a high quality soundand recording system that will allow

the Museum to produce well-engineeredevents and to support its archivalefforts. The Hahn Auditorium alsoserves as a multi-purpose space,allowing the Museum to hold anynumber of institutional and potentialrental events, including banquets,receptions, meetings, etc. This year theMuseum plans to hold the annual FellowAwards Celebration in the new facility onOctober 21, 2003.

Kirsten Tashev is Director of Collections andExhibitions at the Computer History Museum

PA G E 2

C O M P U T E R H I S T O R Y M U S E U M

After a ribbon cutting ceremony, over 600guests celebrated the opening of the Museum’sAlpha Phase.

The new Hahn Auditorium was named afterEric and Elaine Hahn, has a 400-seat capacity,and will serve as a multi-purpose space forbanquets, receptions, and other events.

The Museum’s entrance at 1401 N. Shoreline in Mountain View, Calif.

The open house included tours of VisibleStorage and a reception.

ALPHA PHASEBY KIRSTEN TASHEV

COMPUTER HISTORY MUSEUM:

Phot

o by

Ear

nest

J.

Ray

Phot

o by

Ear

nest

J.

Ray

Phot

o by

Ear

nest

J.

Ray

UNVEILING OF DONOR WALLS

Also at the opening, the Museumunveiled a series of donor walls,including plaques that recognize theearly Boston and Silicon Valley founders,as well as the current annual donors,corporate members, and capitalcampaign donors. The new donor wallsare prominently displayed in the lobbyreception area and will continue torecognize the financial support of themany individuals and companies whoenable the future growth of theMuseum.

A REVITALIZED VISIBLE STORAGE

Key to the Alpha Phase opening was thereincarnation of the Museum’s VisibleStorage exhibit area. Formerly housed ina World War II-era warehouse at MoffettField, the original Visible Storagesuffered from cramped quarters, poorlighting and climate control, and little tono interpretation in the form of labels orother self-guided explanatoryinformation. Although docent-led toursgreatly enhanced the Visible Storage atMoffett Field, docents were oftenrequired to double back and point overand between objects in order to provideadequate tours of the history ofcomputing.

As soon as it was clear that theMuseum would likely purchase thebuilding at 1401 N. Shoreline Blvd., theMuseum’s collections and exhibitionsstaff, along with the exhibit design firm,Van Sickle & Rolleri Ltd., with the inputof trustees and other subjectspecialists, began the work of designingthe new Visible Storage. From the

outset, the design embraced theconcept of a Visible Storage rather thana full-fledged museum exhibit. VisibleStorage—sometimes known in themuseum field as Open Storage—hasbecome quite popular in the last decadeas a legitimate display technique. Unlikea Museum exhibit that attempts toexplain more complex informationthrough extensive graphics, audio-visuals, and computer interactives, aVisible Storage relies mainly on artifactsor objects as the primary means ofcommunication along with a limitednumber of explanatory labels. Since themajority of museums have limited spaceand can only exhibit between 10 and 20percent of their collections at any onetime, the low-cost method of OpenStorage allows them to provide publicaccess to “back end” collections thatthey otherwise would not be able todisplay. In the case of the ComputerHistory Museum, this approach allowsus to make our collection available tothe public while we fundraise to buildmore content- and multimedia-richexhibitions.

Having worked for a little over a year onthe future 15,000 square-foot Timelineof Computing History exhibitionscheduled to open in two to four years,the team had a solid outline of themajor highlights of computing historyfresh in their minds. Although restrictedto using the artifacts themselves and tolimited explanatory techniques, theteam wanted to create a new andimproved Visible Storage that wouldattempt to touch on all aspects ofcomputing, including software, hardwareand underlying technology, graphicssystems, networking, Internet, andcomputing precursor systems.

In developing the new Visible Storage, atension between presenting informationchronologically versus thematicallypresented itself. The end resulted in acompromise, so that the artifacts arelaid out in a loose chronological order,yet the plan allows for diversion hereand there to show developments bytheme in a specific area such asstorage or peripheral devices, orsupercomputers, etc.

The experience is greatly enhanced byexplanatory labels for each artifact.Expanded labels explain more complexor less object-based information, suchas computing concepts or developmentsin the field of software. Although thenew Visible Storage is not a full-scaleexhibition, nor by any means acomprehensive presentation ofcomputing history, with approximately600 artifacts, 150 historic photos, anda computer restoration workshopfeaturing the IBM Model 1620, it offersmuch greater access to the Museum’srich collection.

FUTURE EXHIBITIONS

While the Alpha Phase opening is nodoubt a significant achievement for theMuseum, there still remains much to bedone in order to preserve and presentthe amazing story of the computingrevolution. The Museum is currentlyraising funds to expand its offeringsthrough world-class exhibitions in futurephases or releases. The Museum plansto create exhibitions that are rich incontextual media and interpretivecontent, covering all aspects ofcomputing history. These will include a

PA G E 3

C O R E 4 . 1

A series of new donor walls in the lobby receptionarea acknowledge current annual donors, corporatemembers, capital campaign donors as well as earlyBoston and Silicon Valley founders.

Visible Storage has been moved from awarehouse at Moffett Field (above) to the newbuilding (below). Many enhancements havebeen made, including improvements in lookand feel, expanded labels, and an organizedlayout displaying 50% more artifacts.

PA G E 4

C O M P U T E R H I S T O R Y M U S E U M

1) Enter our new Visible Storage exhibit area,displaying approximately 600 artifacts, 150historic photos, and a computer restorationworkshop featuring the IBM 1620.

2) An organized layout, expanded labels, andprotective stanchions are helping to improvevisitor experience.

3) An office area of the building was convertedfor Visible Storage. The carpeting was removedand the cement flooring was re-finished, blindsin the floor-to-ceiling windows are kept shut toprotect the artifacts from light, and cubiclewalls were re-purposed as dividers betweensections.

4) Even though the new Visible Storage hasmore space and more artifacts, the displaysare still fairly dense with a lot of items locatedclosely together. To the interested, theexperience can be one of great depth.

5) Just one of the new item labels foundthroughout Visible Storage.

1

2

3

45

PA G E 5

C O R E 4 . 1

Timeline of Computing History that willfocus on headline stories in achronological format, five Theme Roomgalleries that will explore specific topicsin more detail and show developmentsin sub-fields of computing side by side, and a large gallery for changing topicalexhibits, the possibilities of which areendless.

Plans also include a rich onlineCyberMuseum experience to includeaccess to the Museum’s collections,and a variety of interpreted CyberExhibits. In addition, the Museum plansto offer a reference library, gift shop,café, and multipurpose event andclassroom spaces. These amenities will

enable us to expand our community andwill support the Museum’s educationalprograms, including a speaker series,seminars, workshops, artifactrestorations, and other volunteer-ledprojects.

As I hope you can see, the Museum hasa solid base upon which to achieve itsobjectives, with its deep collection,enthusiastic supporters, promisingfacility, and public mission. To reach ourgoals, however, we need your help as asupporter, donor, and volunteer. Pleasestop by, see our Alpha Phase, and helpus achieve our future plans through yoursupport and participation.

Tours of the new Visible Storage arenow given on a regular basis, and aVirtual Visible Storage is also nowavailable. For more information, pleasevisit the Museum’s Web site atwww.computerhistory.org.

The Museum is developing a Timeline Exhibit that will be media-rich and highlyinterpretive. These artists renderings explore ideas for how this Timeline Exhibitcould be configured.

The Museum is planning additional buildingrenovations for two upcoming phases, Release 1.0and Release 2.0. In addition to the Timeline Exhibitand Theme Galleries, plans call for a referencelibrary, gift shop, café, and multipurpose event andclassroom spaces.

INTRODUCTION

In early 1972, a small group ofcomputing professionals came togetherin Kingston, Canada, to design a novelcomputer system based on emergingmicroprocessor technology. The result oftheir work at Micro Computer MachinesInc. (MCM) was the MCM/70 personalcomputer. The following article detailsthe early stages in the development ofthe MCM/70 microcomputer. The articleis based primarily on development notesauthored by Mers Kutt, the firstpresident of MCM. These notes, mostlikely written between February and Mayof 1972, are among the oldest recordschronicling the coming of the personalmicrocomputer. Quotations by Kutt,Gordon Ramer, José Laraya, andMorgan Smyth were obtained throughinterviews by the author betweenMarch, 2001, and December, 2002.Kutt’s notes and recordings of theinterviews currently reside at the YorkUniversity Computer Museum in Toronto,Canada.

MCM/70 UNVEILED

The MCM/70 computer, designed byMCM between 1972 and 1973, ispossibly the earliest example of amicrocomputer manufacturedspecifically for personal use. From thehardware and software engineeringpoints of view it does not have much incommon with early hobby computers,such as the MITS Altair 8800 or Apple I,except that these computers weremicroprocessor-based. By the time theAltair 8800 kit was offered to hobbyistsin early 1975, with its 256 bytes ofRAM memory and no high-levelprogramming language capability, theMCM microcomputers were providing in-house APL (A Programming Language)support for applications ranging fromengineering design, modeling andsimulation, to investment analysis andeducation. By the time the Apple I boardwas offered for sale in 1976, the MCMmachines were being used by ChevronOil Research Company, Firestone,Toronto Hospital for Sick Children,

Mutual Life Insurance Company of NewYork, Ontario Hydro-Electric PowerCommission, NASA Goddard SpaceFlight Center, and the U.S. Army, toname just a few MCM customers.

The official announcement of theMCM/70 came on September 25,1973, in Toronto. Two days later, it wasunveiled in New York and the followingday in Boston. An early prototype hadbeen demonstrated to the APLcommunity in May of 1973 during theFifth International APL Users’Conference in Toronto. Anotherprototype was touring Europe in Augustand September of that year and wasshowcased by the MCM team inHolland, Germany, Switzerland, France,Italy, and the U.K. Other prototypes ofthe machine included an earlyrefinement of the Intel SIM8-01development board, a rack-based wire-wrapped system, a desktop bare-bonessystem, and even a cardboard mockup.

THE MCM/70 MICROCOMPUTER

BY ZBIGNIEW STACHNIAK

PA G E 6

C O M P U T E R H I S T O R Y M U S E U M

Announcement of the MCM/70 during the press conference at the Royal York Hotel in Toronto, September25, 1973. From left: Mers Kutt, Gordon Ramer, Ted Edwards, and Reg Rea. Source: Canadian Datasystems,October 1973, p. 49.

The MCM/70 desktop bare-bones system. 3Dmodel created by André Arpin.

KUTT SYSTEMS INC.

In the fall of 1971, just after he left hisfirst company, Consolidated ComputerInc., Canadian inventor andentrepreneur Mers Kutt decided todevelop a small desktop personalcomputer that could be programmed inAPL. Kutt followed technologicaldevelopments and market trends in thesemiconductor industry closely. Hepersonally knew Bob Noyce, then theCEO of Intel Corporation, and wasmeeting with Intel marketing staff andparticipating in Intel promotionalseminars. He had a good knowledge ofthe technical specifications and of thedevelopmental progress of Intel’s first8-bit microprocessor—the 1201, laterrenamed the 8008. For Kutt, the nearcompletion of the 8008 chip in late1971 was a technological trigger pointurging him to move ahead with hispersonal microcomputer project.

In the beginning, there were just two:Mers Kutt and Gord Ramer, whom Kuttrecruited to work on the softwareaspect of the project. Before joiningforces with Kutt, Ramer was the directorof the Computing Center at St.Lawrence College in Kingston. Beforethat Ramer had worked at YorkUniversity, then on the outskirts ofToronto, and developed the York APL

dialect of Iverson’s APL language. Hisexperience with space-efficient York APLwas critical for the writing of theMCM/APL interpreter, which Ramerinitiated even before the 8008 chip wasavailable in quantity from Intel.

The small company was incorporated onDecember 28, 1971, under the nameKutt Systems Inc. On the same day,Hank Smith, who was in charge ofIntel’s Micro Computer Systems Group,signed a shipment invoice for a SIM4-01 development system, an MCS-4 chipset, and an MP7-01 EPROMprogrammer, together valued at $1,231,to be delivered to Kutt at no charge.The second shipment from Intel, on May 23, 1972, delivered a SIM8-01development board and an MP7-02programmer. By that time, the companyhad hired, among other people, ahardware engineer by the name of JoséLaraya, APL programmers Don Gennerand Morgan Smyth, and softwareengineer André Arpin, whose main jobwould be to develop the virtual memorysystem for the MCM/70 computer.

PA G E 7

C O R E 4 . 1

Mers Kutt speaking at York University, Toronto, October 2001

MCM was aimingat a small,microprocessor-based desktopcomputer thatwould beaffordable, aseasy to use as ahand-heldcalculator, andfunctionally aspowerful as amainframecomputer runningAPL. Nothingsimilar had everbeen built before.At the time, inDecember 1971,the news of aCPU on a singlechip was onlyabout one monthold. Furthermore,APL interpreterswere not evenavailable forminicomputers.

Phot

ogra

ph b

y Z.

Sta

chni

ak

PA G E 8

C O M P U T E R H I S T O R Y M U S E U M

The Key-Cassette drawing. Source: M. Kutt’s notes

A fragment of the Key-Cassette’s keyboard. Source: M. Kutt’s notes

Page two of the MCM/70 User’s Guide introduces the keyboard layout.

THE KEY-CASSETTE CONCEPT

The company was aiming at a small,microprocessor-based desktop computerthat would be affordable, as easy touse as a hand-held calculator, andfunctionally as powerful as a mainframecomputer running APL. Nothing similarhad ever been built before. At the time,in December 1971, the news of a CPUon a single chip was only about onemonth old. Furthermore, APLinterpreters were not even available for minicomputers.

The size, price, and usability targets setby Kutt Systems for its microcomputerfocused the attention of the companyon the calculator market. “The worldwas full of calculators,” recollects Kutt.“They made a real Big Bang.” In hisnotes, Kutt entered, “Try and useexisting calculator cover, display, modifypower supply, and replace keyboard.”Indeed, off-the-shelf calculatorcomponents could save the companymoney. For instance, to package thecomputer into a case that would matchthe design elegance of a calculatorcover, the case would have to bemanufactured using the injectionmolding technique. But that wasexpensive: a good quality mold withsharp corners would cost around$25,000.

Kutt’s notes provide an early glimpse ofthe “computer of the future.” Hisdrawing, entitled Key-Cassette, isamong the oldest preserved sketches ofa microcomputer to be manufactured forthe consumer market. The name “Key-Cassette” most likely derives from “Key-Edit,” the name of the data entrysystem manufactured by Kutt’s formercompany, Consolidated Computer Inc.The drawing depicts a case in the styleof a typical desktop calculator of thattime. The lower part of the front panelhosts a built-in keyboard and the toppart depicts a single cassette drive onthe right and either an acoustic coupleror the second cassette drive on the left.A small display and some switches areplaced in the middle of the panel.

The annotated drawing provides enoughinformation to grasp the basicoperations of the Key-Cassette. Thesmall 32-key keyboard of the Key-Cassette would allow the user to enter

all the alphanumeric characters as well as the APL and special functionsymbols. To achieve such a degree ofcompactness, each key was designed to enter up to 5 symbols (using acombination of key strokes). Thesymbols on the keys would be colorcoded to distinguish between thesymbols that can be entered directly(red symbols in the center of the keys)and those that could be entered via acombination of key strokes (blacksymbols placed in the corners of the keys).

The one-line display of the Key-Cassettewould allow the user to view a singleline of APL code, a computer output, oran error message. Using the rotate keys“�” and “�”, the displayed informationcould be scrolled left and right to fullyreveal its contents. Using the roll keys“�” and “�”, one would scroll throughthe lines of APL code. The sketch of theKey-Cassette is augmented with twodrawings of possible segmented displayelements: one comprised of 13 displaysegments and the other of 15segments. Finally, the tape cassettedrives were to provide external storage.

The production model of the MCM/70would be equipped with a more “user-friendly” APL keyboard (layout modeledafter the keyboard of the IBM 2741terminal), with a one-line plasma displayand up to two digital cassette drivesproviding over 100KB of storage each.Only the sides of the case would beinjection molded, while the rest of thecase would be made of cheaperaluminum.

FROM THE KEY-CASSETTE TO THE

M/C PROTOTYPE

Kutt’s notes contain a detailed analysisof Intel’s MCS-4 chip set, the 8008processor, the SIM8-01 developmentsystem, and the MP7-02 programmer.Kutt looks at the technicalspecifications, pricing, and secondsourcing for electronic components. He looks at Intel itself, its marketingactivities.

In April of 1972, Kutt paid Intel a visitand learned from Bob Noyce and HankSmith about the status of the 8008chip, the availability of the SIM8-01development board, and its supporting

software. He inquired about thepossibility of Intel manufacturing customCPU boards for the MCM computer. Inhis notes, Kutt entered that standard8008 prototyping boards, ones thatcould be used to prototype and test anMCS-8 based system without buildinghis own board, would have a“tremendous impact.”

A month later, Kutt received the SIM8-01 board from Intel and gave it to JoséLaraya for the evaluation and theestimation of its potential for growing an APL machine out of it. Laraya recalls:“Mers brought it [the SIM8-01] in andsaid, ‘Here, see what it does.’ It wasreally computing, it really did things, one little chip.” The experimentationwith the SIM8 board concentrated oninterfacing with various devices (suchas the teletype) and on the use of theMP7-02 programmer for the purpose ofburning Ramer’s APL interpreter into the EPROM chips.

But this early attempt at building amicrocomputer, now called the M/Cprototype in the notes, was adisappointment. Kutt wrote that themachine “is useless as is,” and has tobe “drawn up, rewired, and debugged.”In the end, Laraya decided to abandonthe SIM8 approach and, instead, wasdetermined to build his own hardwarefrom the ground up. He remembersthinking, “OK, this [SIM8-01] is fine,great, interesting, works withteletype…But now, let’s build somethingserious.’’ Laraya adds, “Mers got thechips and on the basis of that Ideveloped the rack version….It was veryfast from the time we had the [SIM8-01]development board.”

Laraya modularized the design of theM/C prototype. One card included the8008-based CPU as well as the displayand the keyboard interfaces. Anothercard contained memory. There was aspecially designed APL keyboard, withthe soft character generator, and asmall plasma display (Burroughs Self-Scan 32-character display). Theproduction model would have one moreboard with the cassette controller andthe Omniport interface on it (to connecta variety of peripherals via the Omniportconnector at the back of the machine).

PA G E 9

C O R E 4 . 1

The rack prototype of the microcomputerwas good enough for Ramer and Gennerto start porting their APL interpreter intoit. On November 11, 1972, theprototype was demonstrated toshareholders during the Special GeneralMeeting of the Shareholders of KuttSystems in Kingston, Ontario. Duringthat meeting a motion was passed tochange the name of the company toMicro Computer Machines.

THE MCM/APL

In the early 1970s, APL was onlyavailable on mainframe computers. Thedevelopment of an APL interpreter andthe memory management system for an8008-platform characterized by lowspeed, restricted instruction set, andsmall memory addressing space wasthe most challenging aspect of thepersonal computer project at MCM.

The team that developed the interpreterhad worked together before. In late1960s, Gord Ramer designed a dialectof the APL/360 language that he namedYork APL. He implemented the languagewith the assistance of Don Genner whileboth Ramer and Genner held positionsin the Computing Center at YorkUniversity. J. Morgan Smyth was amongthe first users of the York APL and hewas frequently commuting between hiswork place—Ryerson PolytechnicInstitute in Toronto—and York Universityto discuss the implementation issues ofYork APL with Ramer and Genner. AtMCM, the trio would develop one of thefirst high-level language interpreters fora microprocessor: Ramer would designthe language, Genner would implementit, and Smyth would document it in anexcellent MCM/70 User’s Guidepublished by MCM in 1974.

The work on the interpreter started inearly 1972 even before MCM built anyhardware that could be used by thesoftware engineering group. Having onlythe specifications of the 8008 chip,Ramer and Genner used the IBMSystem/360 assembler to emulate the8008. “The 360 assembler was writtenin such a way that you could usemacros to generate code for anyhardware,” says Ramer. “Thus [we]generated macros for each 8008instruction and voila!” A similaremulator, the INTERP/8 (written inFortran IV), was later available fromIntel. It provided a software emulation ofthe 8008 chip along with someexecution monitoring commands.

When the rack prototype of themicrocomputer was finally working atMCM’s manufacturing facility inKingston, the development of the APLinterpreter could be done directly on the8008-based hardware and theinterpreter’s code could be burned intoEPROMs using the Intel MP7-02programmer. Programming “was reallyslow,” says Laraya, “and you had toprogram it by hand using switches….Wehad to put the code and set theswitches and the addresses and hit‘program’ [the EPROM]. Every time weprogrammed, Don [Genner] used tosmoke one cigarette and say, ‘That’show long it takes to program a chip.’ Hesmoked one cigarette, and when hefinished, [the chip] was programmed.”

In his notes, Kutt sketches thedirections for the development of theAPL interpreter for the microcomputer.First, the basic, stripped-down version ofAPL/360 would be implemented. Thedescription of an APL fragment thatcomprises single dimension vectors,some defined and some systemfunctions, spans two pages in thenotes. Then the interpreter would beextended in two directions to supportthe scientific and business utilization ofAPL. “When we came up with the APL[interpreter] for our PC,” says Kutt, “ourprime target was to make it simple touse…so [the user] wouldn’t have tobecome embroiled in the little nitty-grittythings you have to look after in APL.”

The full description of the MCM/APLinterpreter for the MCM/70 computerappeared in the MCM/70 User’s Guide.

SELLING THE FUTURE

The notes disclose some urgency toprepare a viable demonstration of theM/C computer. Early demonstrationswere vital to attract venture capital andfinance the operations of the youngcompany. Kutt sketches a “short-cut todemo” in his notes and estimates itscompletion at the early June of 1972.

The M/C demonstrator was to consistof a single CPU and memory board anda power supply packed into a desktopcalculator-like case that featured a built-in keyboard and a small display. It wasto be basic hardware with just enough

PA G E 1 0

C O M P U T E R H I S T O R Y M U S E U M

Morgan Smyth (left), Don Genner, and Gord Ramer(right) at York University, Toronto, October 2001

Programming“was really slowand….every timewe programmed,Don [Genner]used to smokeone cigarette andsay, ‘That’s howlong it takes toprogram a chip.’He smoked onecigarette, andwhen he finished,[the chip] wasprogrammed.”

—José Laraya,MCM Engineer

Phot

ogra

ph b

y Z.

Sta

chni

ak

software stored in ROM to demonstratethe way the 8008 could handle a subsetof APL. It is unclear whether such amachine was ever constructed.However, a more refined portableprototype of the MCM/70 was built and shown during the Fifth InternationalAPL Users’ Conference in Toronto, inMay of 1973. “I remember we had thefiberglass model there,” says Laraya. “It was heating up.” Ramer, too,remembers the event vividly: “The demo had to be interspersed with shorttalks to allow José [Laraya] to exchangethe heat-sensitive parts and thenrestart the system for the next segmentof the demo.”

With limited RAM and no externalstorage, that prototype was nothing butan advanced APL-based scientificcalculator. However rudimentary, it didattract attention of the APL communityand made it evident that in the nearfuture high-level programminglanguages, such as APL, would bereadily available on small desk-topmachines. Other prototypes of theMCM/70, including one mounted in anattaché case and powered by batteries,were showcased throughout Europe andNorth America in the second half of1973, attracting the attention of dailyand technical press.

One of the most successfuldemonstrators that the company puttogether had, in fact, no hardware at all.“We had a cardboard mockup of thecomputer,” says Smyth. “It…was asmall, slick little box...it was justcardboard. And we went around to a lawfirm in downtown Toronto and met withthe bunch of senior lawyersthere….Mers was gonna try to get someventure capital.…These guys were quiteold and, at one point [during Kutt’spresentation], actually the secretary

came in with a can of candies….to perk‘em up. And I thought, ‘man, we arewasting our time here.’”

Smyth continues, “And near the tail ofour presentation, which went on for overtwo hours, …one of these guys said,‘Now, just a minute. This contradictswhat you said at the very beginning.’I’m thinking, ‘What?’ They were payingattention and I was very impressed…We walked out of there with half amillion dollars….It was…justcardboard….He [Kutt] is waving thisaround: ‘This is what it’s going to looklike!’ You are talking about selling thefuture!”

GOOD LUCK, AND WELCOME TO THE

COMPUTER AGE!

The production model of the MCM/70shared many technical features with theKey-Cassette and the M/C concepts.From its prototypes, the MCM/70inherited a desktop design with built-inAPL keyboard, a one-line plasmadisplay, and cassette drives mounted onthe front panel. The computer waspowered by the Intel 8008

microprocessor and its ROM chipscontained the MCM/APL interpreter.

But other features, not discussed inKutt’s notes, also make the MCM/70 atruly unique piece of microcomputerengineering. The MCM/70’s 14KB ofROM contained not only the MCM/APLinterpreter but also the cassette andvirtual memory operating systems(called EASY and AVS, respectively).AVS, designed by André Arpin, used oneof the cassette drives to provide virtualmemory by swapping programs and databetween the cassette and RAM. Withvirtual memory, the MCM/70 offered anexcess of 100KB of memory. A powerfailure protection system built into thepower supply of the computer allowedcontinuous operation by battery in theevent of power failure. For extendedpower loss, the computer initiated anorderly shutdown: it automaticallyprovided system back-up by copying thecontent of RAM to a cassette beforeshutdown. The system wasautomatically reinstated when powerwas restored and batteries wererecharged.

Kutt made some market analysis noteslooking primarily at the IBMSystem/360 users who might benefitfrom a smaller and much less expensivesystem. He calculated APL universityprospects at around 15 in Canada and75 in the U.S. And indeed, following theannouncement of the MCM/70, manyacademics expressed interest in theMCM hardware: “APL is currently usedin all of our introductory courses so that

PA G E 1 1

C O R E 4 . 1

The Short-cut to demo drawing. Source: M. Kutt’s notes

The MCM/70. Courtesy of Harley Courtney

Phot

ogra

ph b

y Z.

Sta

chni

ak

the potential for systems like yours atYale is very high,” wrote Martin H.Schultz, Professor of Computer Scienceat Yale University, to MCM in November1973. By 1976, an estimated 27.5% ofthe MCM systems sold in North Americawent to educational institutions.

The notes, however, do not make anyreference to “personal computing” norto possible marketing strategies aimedat promoting the personal utilization ofMCM’s microcomputers. This is hardly asurprise as the notes were made in thevery early stages of the development ofthe MCM/70. That situation wouldchange with the publication of the firstpromotional documents by MCM in1973. “It has been a combination ofthe complexity of the large computermachines and the complexity of thespecial computer languages,” reads theMCM/70 Introductory Manual publishedby MCM in 1973, “that has till nowprevented the general public from usingcomputers directly themselves. But thesimplicity of the MCM/70 and itsassociated computer language (knownas APL) make personal computer useand ownership a reality….Enjoy theprivilege of having your own personalcomputer—It’s a privilege no computeruser has ever had before theMCM/70….Good luck, and welcome to the computer age!”

It is difficult to explain unequivocallywhy the MCM/70 was not thecommercial breakthrough to launch the personal computing industry. It isalso difficult to estimate the number of MCM/70 computers sold worldwideor the scope of impact it had on theAPL community and on the rise ofpersonal computing. Even so, it wasMCM’s historical role to show that withthe advent of microprocessortechnology, affordable personalcomputing was at our fingertips. It wasnot too farfetched to imagine that, “inthe coming years the computer field isgoing to be made of millions of smallcomputers and a limited number oflarge computers” (Mers Kutt, Boston,September 28, 1973).

Zbigniew Stachniak is an associate professor ofcomputer science at York University in Toronto,Canada. His research concentrates on formalmethods in artificial intelligence (automatedreasoning, knowledge representation), on symboliclogic in computer science, history of computing,and history of logic.

The author extends his gratitude to the NationalScience and Engineering Research Council ofCanada for supporting his research on MCM.

FURTHER READING

Chevreau, J. “The Third Coming of Mers Kutt.”Report on Business Magazine, November 1985, pp. 110-115.

Stachniak, Z. “The Making of the MCM/70Microcomputer.” IEEE Annals of the History ofComputing, May/June, 2003.

The MCM Collection at York University Computer Museum:http://www.cs.yorku.ca/~zbigniew/MCM_col.html

PA G E 1 2

C O M P U T E R H I S T O R Y M U S E U M

“The complexityof the largecomputermachines and thecomplexity of the specialcomputerlanguages…hastill now preventedthe general publicfrom usingcomputersdirectlythemselves. Butthe simplicity ofthe MCM/70 andits associatedcomputerlanguage…makepersonalcomputer use and ownership areality….Enjoythe privilege ofhaving your ownpersonalcomputer.”

—The MCM/70 Introductory

Manual, 1973

PA G E 1 3

RECENT ADDITIONSTO THE COMPUTER HISTORY MUSEUM COLLECTION

Two (2) Core Memory Plane Assemblies (1973),X2523.2003, gift of Richard Walters

Two (2) microchip portfolio sets (1976 – 1989),X2497.2003, gift of Hewlett-Packard Company

Three (3) publications relating to early computers(c. 1950-1955), X2529.2003, gift of Gordon Uber

Four (4) original photographs of Steve Wozniak andSteve Jobs (1976), X2554.2003, gift of Joe Melena

Amdahl 470/V6 MCC (Multi Chip Carrier) (c. 1985),X2576.2003, gift of Mr Naoya Ukai

Ampro Computers, Inc., Series 100 “Bookshelf”CP/M Computer, operating software, anddocumentation, (c. 1982), X2535.2003, gift ofSteve Brugler

Anderson Jacobson Model ADC 260 AcousticCoupler (c. 1975), X2523.2003, gift of RichardWalters

Anderson Jacobson Model ADC 300 AcousticCoupler (c. 1975), X2523.2003, gift of RichardWalters

Apollo Computer, Inc., “Network Outlet” connection(1983), X2573.2003, gift of Jonathan Gross

Apple Macintosh (1984), X2523.2003, gift ofRichard Walters

Apple Macintosh Laptop with Duo Dock (c. 1994),X2523.2003, gift of Richard Walters

Bell Laboratories Picturephone (1964),X2560.2003, gift of Les Earnest

“The Binary Slide Rule” (c. 1940), X2551.2003,gift of Wolfgang Schaechter

Bowmar Instrument Company, MX55 PersonalCalculator (“Bowmar Brain”) (1970), X2510.2003,gift of Michael Percy

Bowmar Model 901B Calculator (c. 1973),X2577.2003, gift of Mary M Mourkas

Burroughs Adding Machines (two) (c. 1935 and c. 1945), X2525.2003, gift of Chuck Kaekel

Burroughs B1900 Mainframe Computer System(including peripherals) (c. 1985), X2550.2003, giftof the Pennyroyal Center

Canon Cat V777 Work Processor, associatedsoftware and documentation (1987), X2538.2003,gift of Paul Cubbage

Canon Cat180 Daisy Wheel Printer with CanonCat40 Cut Sheet Feeder Option (c. 1987),X2538.2003, gift of Paul Cubbage

Check Point Software Technologies, Inc., FireWall-1Ver. 2.0 Media Pack (1994), X2513.2003, gift ofCheck Point Software Technologies, Inc.

Check Point Software Technologies, Inc., SofaWareS-box Internet Security Appliance (2002),X2513.2003, gift of Check Point SoftwareTechnologies, Inc.

Check Point Software Technologies, Inc., VPN-1 &FireWall-1 Media Pack (2002), X2513.2003, gift ofCheck Point Software Technologies, Inc.

Collection of early IBM ephemera, software, anddocumentation (c. 1950-1970), X2517.2003, gift ofBob Brubaker

Collection of Apple Developer Group CD Seriescompact discs (c. 1988-1993), X2531.2003, gift ofLars Borresen

Collection of Apple marketing materials on compactdisc (1990-1993), X2546.2003, gift of Terry LKristensen

Collection of artifacts, documents and mediarelated to the WEIZAC and GOLEM computers(various dates), X2556.2003, gift of Gerald Estrin

Collection of computer industry business cards(1982-2003), X2561.2003, gift of Tom Halfhill

Collection of Cray software documentation (variousdates), X2514.2003, gift of Warren Yogi

Collection of Digital Equipment Corporationephemera (various dates), X2524.2003, gift ofJudith Burgess

Collection of documents, photographs and slidesrelated to the history of supercomputing, massstorage systems and networking at the NationalCenter for Atmospheric Research (c. 1979-2000),X2548.2003, gift of Basil L Irwin

Collection of early timesharing manuals byTymshare, Inc., (1974-1984), X2547.2003, gift ofJoe Smith

Collection of early Xerox Corporation computermanuals, newsletters and reports (various dates),X2542.2003, gift of Mike Rutenberg

Collection of ephemera related to the developmentof computer memory (various dates), X2537.2003,gift of William F Jordan

Collection of ephemera, documents and slidesrelated to the history of microelectronics (variousdates), X2549.2003, gift of Olive Thompson

Collection of fourteen (14) advertisements forHoneywell computers (c. 1965), X2539.2003, giftof Mark Barnett

Collection of IBM ephemera, documents and media(c. 1955-1967), X2558.2003, gift of Neil Lewis

Collection of machine and program manuals andbrochures (c. 1950-1969), X2564.2003, gift ofChuck Baker

Collection of photographs and ephemera related tothe IBM Model 1360 “Cypress” Photo-DigitalStorage Systems (c. 1967), X2509.2003, gift ofJack Harker

Collection of photographs of the UNIVACIncremental Computer (1956), X2529.2003, gift ofGordon Uber

Collection of reference manuals, flowchartingtemplate, pocket guides, and programmer referencecards (1964-1985), X2553.2003, gift of Ken North

Collection of selected materials from the TandemArchival Collection (various dates), X2528.2003,gift of Hewlett-Packard Company

Collection of seven (7) boxes of assorted softwareand related documentation (1980-1990),X2502.2003, gift of Arel Lucas

Collection of software and documentation related topersonal computing (c. 1980-1995), X2557.2003,gift of George Glaser

Collection of the first one hundred SunMicrosystems Laboratories technical reports (1991-2002), X2544.2003, gift of Sun Microsystems, Inc.

Collection of UNIVAC materials (various dates),X2562.2003, gift of Carol Canzano-Zito

Collection of various materials relating to theindustrial design of the Xerox Alto (1973),X2536.2002, gift of Terry West

Commodore “SuperPET” SP9000 personalcomputers (two), dual disk drive, hard drive,software and documentation (c. 1981),X2494.2003, gift of Vladimir Steffel

“Computer Music from the University of Illinois”record album (c. 1963), X2552.2003, gift ofRichard Ellis

Computers (Boy Scouts of America Merit BadgeSeries) (1968), X2493.2003, gift of Dag Spicer

Computran Model AN 7 Computer Trainer (c. 1965),X2514.2003, gift of Warren Yogi

Control Data Corporation Removable Disk Pack (c. 1970), X2523.2003, gift of Richard Walters

Core Memory Plane Assembly (1960),X2523.2003, gift of Richard Walters

“CRAM-80” homebrew computer (c. 1975),X2566.2003, gift of Steven E Young

DEC software and manual collection (variousdates), X2571.2003, gift of Kenneth L Voss

Designing with FPGAs & CPLDs (2002) and VerilogDesigner’s Library (1999), X2506.2003, gift of BobZeidman

Digital Equipment Corporation “Computer Lab”Digital Logic Trainer (c. 1962), X2518.2003, gift ofRob Keeney

DOS 3.30 for the Dynabyte 5200 Computer Unit(1982), X2535.2003, gift of Steve Brugler

Dynabyte Business Computers, Technical Manualfor the Dynabyte 5200 Computer Unit (c.1982),X2535.2003, gift of Steve Brugler

DYSEAC components and documents (1954),X2489.2003, gift of David E Hartsig

E & L Instruments Mini-Micro Designer (MMD) 18080 trainer board (c. 1976), X2534.2003, gift ofPhil Keller

Electronics Australia EDUC-8 microcomputer (1975),X2520.2003, gift of John Whitehouse

Franklin ACE 1000 with documentation andsoftware (c. 1983), X2523.2003, gift of RichardWalters

Friden Flexowriter (c. 1961), X2515.2003, gift ofRichard Leamer

Fujitsu Stylistic ST4100 (2003), X2575.2003, giftof Mr Toshio Morohoshi

Gavilan Mobile Computer, software, anddocumentation (1984), X2505.2003, gift ofAngelina M Jimenez

Gear and arm from Science Museum, London,Babbage Engine construction (2003), X2563.2003,gift Dr Thomas Bergin

C O R E 4 . 1

PA G E 1 4

C O M P U T E R H I S T O R Y M U S E U M

Handspring, Inc., Treo 180 Communicator (2002),X2503.2003, gift of Donna Dubinsky

Hewlett-Packard HP-85 Personal Computer (c. 1983), X2523.2003, gift of Richard Walters

Hewlett-Packard HP-97 Programmable Calculator (c. 1979), X2523.2003, gift of Richard Walters

Hewlett-Packard Model 200C Oscillator (c. 1940),X2526.2003, gift of George Durfey

Hewlett-Packard Model 200C Oscillator (c. 1940),X2565.2003, gift of SRI International

Hewlett-Packard production prototype DDS-1 tapedrive and data cartridge (c. 1987), X2512.2003,gift of Dominic McCarthy

IBM “Reflexione” ("THINK") Sign (c. 1970),X2545.2003, gift of Tom Reif

IBM 360/30 CCROS card (c. 1965), X2578.2004,gift of Brian Knittel

IBM 5110 minicomputer system, with original CPU,monitor, disk drive, tape unit, printer,documentation, and software library (c. 1978),X2511.2003, gift of Jan Engel

IBM 700-series pluggable unit (1952),X2491.2003, gift of Gwen Bell

IBM AN/FSQ-7 (SAGE) Theory of ProgrammingManual (1957), X2527.2003, gift of Robert F Martina

IBM core plane (c. 1960), X2499.2003, gift of ArtSiegel

IBM Hard Drive Assembly (c. 1970), X2523.2003,gift of Richard Walters

IBM Hexadecimal Adder (1957), X2545.2003, giftof Tom Reif

IBM Manual of Instruction Customer Engineering(1946), X2568.2003, gift of Warren Yogi

IBM Model 10 Card Punch (c. 1940), X2523.2003,gift of Richard Walters

IBM Model 5151 Personal Computer Display (c. 1981), X2523.2003, gift of Richard Walters

IBM Time Clock (c. 1913), X2569.2003, gift of LenShustek

IMSAI 8080 Microcomputers with documentationlibraries (two) (c. 1976), X2523.2003, gift ofRichard Walters

IMSAI Dual 8” Floppy Disk Drive (c. 1976),X2523.2003, gift of Richard Walters

International Correspondence Schools ComputerCode Translator Slide-Chart (1983), X2519.2003,gift of Bill Kochanczyk

Kaypro 2000 Personal Computer and Docking Port(c. 1987), X2523.2003, gift of Richard Walters

Keuffel & Esser Company, Beginner Slide Rule (c. 1954), X2553.2003, gift of Ken North

Let ERMA Do It (1956), X2507.2003, gift of GeorgeDurfey

Livermore Data Systems Model B Acoustic Coupler(c. 1965), X2523.2003, gift of Richard Walters

M & R Enterprises Pennywhistle 103 modem,X2559.2003, gift of Bill Hill

Mechanical Analog Computer (c. 1965),X2523.2003, gift of Richard Walters

Monroe Epic 2000 Electronic Printing Calculator (c. 1955), X2530.2003, gift of Dorothy Burkhart

Netronics Research and Development, Ltd.,COSMAC ELF microcomputer (1976), X2532.2003,gift of Bill Buzbee

Okimate 10 Personal Color Printer (c. 1984),X2523.2003, gift of Richard Walters

Olympia magnetic dictation machine (c. 1970),X2521.2003, gift of Bob Feretich

Original Homebrew Computer Club T-shirt (c. 1986),X2579.2004, gift of Carrie Karnos

“The Orm” robotic arm (1965), X2574.2003, gift ofStanford University

Packard Bell PB 250 minicomputer and collection ofassociated software and documentation (1961),X2515.2003, gift of Richard Leamer

PCD Maltron Ergonomic Keyboard (c. 1990),X2523.2003, gift of Richard Walters

Philips Nino 300 Personal Data Assistant (c. 1998)and Nino T-Shirt, X2555.2003, gift of Kevin Turner

Processor Technology Corporation SOL TerminalComputers (two) (c. 1978), X2523.2003, gift ofRichard Walters

Programming Systems & Languages (1967), A SNOBOL4 Primer (1973), and Computers andSociety (1972), X2567.2003, gift of Jim Gross

Punch card carrying case (c. 1960), X2504.2003,gift of Herman Griffin

Quantum Computer Services, Inc., America OnlineVer. 1.0 (1989), X2508.2003, gift of Adam Gross

Radio Shack (TRS-80) 64K Color Computer 2 (c. 1985), X2523.2003, gift of Richard Walters

Radio Shack (TRS-80) Model 4 Micro Computer (c. 1985), X2523.2003, gift of Richard Walters

“Rancho Arm” robotic arm (1963), X2574.2003,gift of Stanford University

Remington Rand Corporation magnetic tape (c. 1966), X2573.2003, gift of Jonathan Gross

Rockwell International R6500 Advanced InteractiveMicrocomputer (c. 1979), X2522.2003, gift of BobBynum

Russian “Microcalculator Electronica B3-36”calculator (1983), X2514.2003, gift of Warren Yogi

Russian “Olimpik-C” personal computer (c. 1993),X2514.2003, gift of Warren Yogi

Russian abacus (stchoty) (c. 1963), X2514.2003,gift of Warren Yogi

Sama & Etami, Inc., “The Concise ConversionTables and Circular Slide Rule” (c. 1960),X2551.2003, gift of Wolfgang Schaechter

Seagate ST-225 hard disk drive (1984),X2572.2003, gift of Henry Plummer and RobertLewis

Sharp Corporation Model OZ-7000 “Wizard”Electronic Organizer and Interface Software (dateunknown), X2543.2003, gift of Eugene Miya

Signed promotional poster: “Intel DeliversSolutions” (c. 1982), X2533.2003, gift of StephenCasner

Silicon wafer collection (c. 1965-1995),X2495.2003, gift of Mark Noreng

Smithsonian Institution Annual Report 1874,X2563.2003, gift of Dr Thomas Bergin

Tadpole Technology SPARC-book 2 (1993),X2500.2003, gift of Bill McKie

Tandy 1400 FD Personal Computer with associatedcables, manuals and software (c. 1989),X2540.2003, gift of Mark Gilkey

Tandy Acoustic Coupler 2 (c. 1989), X2540.2003,gift of Mark Gilkey

Tandy Corporation TRS-80 III (1981), X2501.2003,gift of Bob Zeidman

Tandy Radio Shack Model 200 Portable Computer(1985), X2523.2003, gift of Richard Walters

Technical Design Labs Xitan microcomputer (c. 1977), X2516.2003, gift of Cappy Jack

TeleSensory Systems, Inc., Speech+ Calculator(English language model) (1975), X2535.2003, gift of Steve Brugler

TeleSensory Systems, Inc., Speech+ Calculator(German language model) (1976), X2535.2003, gift of Steve Brugler

Unisys historic videotape collection (various dates),X2492.2003, gift of Unisys Corporation

UNIVAC Products Handbook (copy) (1959),X2570.2003, gift of Unisys Corporation

Vacuum Tube Flip Flop Module (c. 1975),X2523.2003, gift of

VISICALC 1.0 software (1982), X2490.2003, gift ofMary Cooper

Wozniak “Blue Box” (c. 1972), X2541.2003, gift ofAllen Baum

Wyse Technology, Inc., Series 7000i Model 760 MPcomputer (c. 1994), X2498.2003, gift of BarbaraGasman

Xerox 860 information processing system (1980),X2496.2003, gift of Ken Lehmann

(Dates represent dates of introduction and notnecessarily dates of manufacture.)

RECENT ADDITIONS, CONT’D

One never knows where one will findhidden treasure. In this case, a photo posted on Usenet contained a serendipitous glimpse of hiddentreasure, and this is the story of how an operational 35-year-old mainframecomputer system was discovered,donated, and moved to the ComputerHistory Museum.

FINDING A PIECE OF MAINFRAME

HISTORY

In August 2000, while perusing thePDP-8 newsgroup, I ran across a queryabout an old computer (seewww.computerhistory.org/projects/pdp8_restoration/). Pittsburgh graduatestudent Raymond Jensen was askingfor information on an old system—thePDP-8 made by the now-defunct DigitalEquipment Corporation—he haddiscovered in a basement laboratoryalong with several other large unknowncomputers. He subsequently provided aWeb page with photos of theequipment, which indeed depicted aPDP-8 minicomputer. However, onepicture also showed a portion of a largetape drive, which I immediatelyrecognized as a Scientific Data Systems(SDS) reel-to-reel tape drive like the oneI had used on an SDS Sigma 7mainframe computer in the early1970s. Was it possible that a Sigma-series mainframe was attached to thattape drive? By chance had I stumbledacross what was perhaps a 1960s-eramainframe still persisting “in the wild?”

I immediately contacted Jensen andtold him the machine just outside thepicture could be an older SDSmainframe, and might be of interest tothe Computer History Museum. A fewdays later I received an email from himindicating that it was indeed a Sigma 5mainframe and that it was located atCarnegie-Mellon University (CMU) in thechemistry department’s NMR (NuclearMagnetic Resonance) Facility forBiomedical Studies. Jensen knewnothing more about the system, itsorigin, or even its manufacturer, exceptthat it was really big and was comprisedof several cabinets. He provided theCMU email addresses of Dr. AkselBothner-By and Dr. Joseph Dadok, bothof whom were retired from thechemistry department.

PRESERVING HISTORY:

THE SDS SIGMA 5FINDS A NEWHOMEBY LEE COURTNEY

PA G E 1 5

C O R E 4 . 1

CPU and tapedrive cabinets

My immediate thought was that thiswould make an excellent addition to thecollection at the Computer HistoryMuseum. I wrote both Bothner-By andDadok, along with the current ChemistryDepartment Chairman RichardMcCullough, asking for more informationon the Sigma 5, its current status, andtheir interest in donating the system tothe Computer History Museum.

Bothner-By and Dadok relayed that thesystem was indeed an SDS Sigma 5mainframe first installed at CMU around1968 and was still operational althoughno longer used. And the chemistrydepartment was interested in seeingthe system preserved at the ComputerHistory Museum. Since the system wasstill operational, it would be importantto make sure it arrived at the Museumin the same condition. This wouldrequire careful de-installation,documentation, and packing of thesystem. Unfortunately, CMU would notbe able to provide resources to perform these tasks.

DONATION RECONNAISSANCE AND

AGREEMENT

It turned out I was taking a businesstrip to the east coast in November 2001and was able to make a side trip tomeet with both Dadok and Bothner-By.The purpose of the trip was to examinethe system in person, determine arough inventory of what items would bepart of the donation, assess the scopeof the necessary shipping preparation,and learn more about the machine’shistory and possibilities it might havefor future use.

I anxiously arrived at CMU and metBothner-By in his office at the Mellon

Institute. The Sigma 5 resided in a sub-basement laboratory in part of thebuilding that has not been renovatedsince original construction in the1940s.

We went downstairs through thebasement to the sub-basement, withpipes overhead and long corridorsstretching into the distance. We passedmany offices and labs as we descendeddeeper and deeper into the building. I felt like I was walking into an episodeof the X-Files. Bothner-By explained thatthe Sigma 5 resided in a part of thebuilding that had housed a small-scaleprototype chemical plant during WorldWar II. At the end of one long corridor,we reached the NMR Lab with themagnet and control rooms on the rightside of the hallway and a computerroom with the Sigma 5 on the left.Dadok, who had emigrated fromCzechoslovakia in 1968, soon joined usafter an event honoring a visit by thenew president of that country.

Entering the Sigma 5 computer roomwas a step back in time to the dayswhen mammoth computers wereisolated in large rooms with raisedfloors for snaking cables betweencabinets and for hiding large powerconduits, while air-conditioning unitsconstantly blew cold air to cool thesystems. The Sigma 5 sat in the centerof the room, as it had for the last 33years. At opposite ends of the computerroom were a 1960s-era PDP-8 system,a 1970s-era Harris minicomputer, and alate-1980s VAX. Racks holding 1/2-inchmagnetic tapes used with the Sigma 5stood along one wall. Various storagecabinets and bookshelves with printoutsand systems’ documentation covered

the other walls. We had to talk loudlyover the low rumble of a large air-conditioning unit.

The atmosphere took me vividly back to the early 1970s when I used an SDSSigma 7 in high school and later workedas the console operator for myuniversity’s IBM mainframe. That onemoment standing in the computer room with the Sigma 5 made the tripworthwhile.

We talked briefly about the current state of the Sigma 5. One importanttask I wanted to accomplish on thatvisit was to capture Bothner-By andDadok’s experiences with the systemand stories of its use. I videotapedBothner-By discussing the work done in his lab and Dadok discussing hiscareful maintenance of the machine and associated instrumentation duringall the years of use. We spent severalhours videotaping the machine inoperation, with Dadok powering on the system and going through thedifferent components, how the machine operated, and quirks of theSigma 5. Unfortunately, a hardware error prevented us from booting up the system.

In addition to videotaping the Sigma 5,we also discussed the instrumentationattached to the Sigma 5 and its functionand contributions to science over theyears. The attached equipment includedthe original NMR spectrometer, morerecent NMR instrumentation, and acustom-built system for controlling theinstrument that interfaced to the Sigma 5. The control system wascomposed of two 19-inch rackscontaining various consoles and

PA G E 1 6

C O M P U T E R H I S T O R Y M U S E U M

Lee Courney inventories the Sigma 5 shipment after arrival at theMuseum.

Console teletype. Through the mid-1970s, most mainframe computersystems used hardcopy terminal as consoles to aid in resourceaccounting, operator tasks, and general system debugging. The Sigma 5was unusual in that it employed a TTY utilizing EBCDIC rather than themore common ASCII character encoding.

PA G E 1 7

C O R E 4 . 1

electronic equipment including an SDSA-D (Analog-to-Digital) converter, whichconverted signals from the instrumentand sensors to digital form that couldthen be stored and processed by theSigma 5.

Bothner-By and Dadok provided a wealthof information on how the Sigma 5 wasused at CMU. Originally the Sigma 5was purchased by the NIH (NationalInstitute of Health) and installed at theUniversity of Indiana. After a year itbecame available and Bothner-By wrotea proposal for its installation in his lab at CMU. In 1968 the system wasmoved to CMU and installed as part ofthe NIH-sponsored National NMR Facilityfor Biomedical Studies in the chemistrydepartment. This facility providedaccess to an extremely powerful nuclearmagnetic resonance spectrometer thatallowed biologists, biochemists, andother scientists to analyze the chemicalmakeup and structure of organiccompounds.

When introduced, the Sigma 5 wasmarketed as a real-time and process-control system, as well as a smallmainframe for business or scientificcomputing. True to its real-time nature,the Sigma 5 was the primary control,data collection, and analysis tool usedfor the spectrometer. Since the Lab wasa national facility sponsored by the NIH,users came from all over the UnitedStates as well as other countries. Formany years, the magnet used in theNMR spectrometer controlled by theSigma 5 was one of the most powerfulof its type.

When a sample was being analyzed, theSigma 5 was connected to various

sensors that collected data that wasrecorded on a reel-to-reel magnetic tapeor a high-speed fixed-head single-platterdisc called a RAD. Once collected, theSigma 5 could reduce, analyze, ordisplay the data. Originally the systemwas outfitted with a plotter and couldproduce graphical representations ofdata. A FORTRAN compiler was alsoavailable for creation of programs toperform analysis. However, manyscientists using the instrument wrotetheir data to magnetic tape for lateranalysis at their home institution.

An artifact of 1960s computing thatbenefited the lab was that completehardware documentation and systemsoftware source code were providedwith the Sigma 5. In addition to I/Odesigned to facilitate “custom”hardware and interfaces, the completedocumentation allowed Dadok to designand interface scientific instrumentsunanticipated by the original designersand to apply the system to newproblems.

Today we would refer to these attributesas open standards, design, and source.By studying the Sigma 5 and itscontemporaries, we can see that the concept of “open source” wasalready a well-established practice evenby the time the Sigma 5 came intobeing in the mid-1960s.

After capturing about four hours of videoof the system along with Bothner-By andDadok describing the lab, instruments,Sigma 5 system, and how all were usedin the scientific community, it was timeto head to the airport. I had taken thistrip to learn what I could about thesystem in order to prepare a proposal to

the Museum for acquiring it. It wasapparent that this could be a significantaddition to the Museum’s collection andcould help to document computinghistory, especially 1960s-era mainframetechnology.

APPROVING THE DONATION

Once back in California, it was time towork on the Museum end of things. Each week the Museum receivesmultiple inquiries about potentialdonations. Unfortunately, it cannotautomatically accept all of them.Currently, the collection occupies over35,000-square-feet of horizontal storagespace, so obvious practical constraintsaffect the acceptance of new items.Because the Museum must carefullyconsider the historical value of anartifact before accepting it, aCollections Committee—a group of staffand volunteers chaired by MuseumTrustee John Mashey—meets regularlyto consider, accept, and declinedonations.

Using collective experience as well asformal evaluative criteria, theCollections Committee looks fordonations that are relevant to themission of the Museum—to preserveand present for posterity the artifactsand stories of the information age—andthat add to our understanding ofcomputing history.

These items generally fall into one offive categories: hardware, software,documentation and printed matter,films/video/photos, and ephemera.Items currently in the collection rangefrom individual hardware componentssuch as vacuum tubes from earlycomputers to software such as Bill

Button detailThe card reader, the primary input device forprogramming the Sigma 5

The cables were hard-wired into the cabinets and had circuit boardsattached at various points.

The weave of cables from cabinet to cabinet to power source were like aGordian knot—almost impossible to untangle without a sword.

Gates and Paul Allen’s original BASICpaper tape, to complete mainframe/supercomputer systems such as theCray-1, and include films and videos ofimportant lectures given by pioneers inthe field of computing.

In preparing a proposal for theCollections Committee, I consideredhow the Sigma 5 would contribute to thecollection, its value in establishing anhistorical record for its era, and how itcould provide insight into the evolutionof computing. The donation met theMuseum’s desire to collect itemsgreater than 10 years old, having beeninstalled at CMU in 1968. In addition, Iconsidered the role SDS played as acompany in the mid-1960s, theMuseum’s need for a representativesample of 1960s-era mainframecomputing technology, and how a well-documented and operational mid-1960smainframe would contribute to theunderstanding of computing. While theSDS Sigma series was not the mostprevalent system of its time, it was anexcellent touchstone and example of1960s computing.

This donation would provide theMuseum with a rare and very desirableopportunity: to approach an artifactacquisition from a systems perspective.Often items, especially larger ones suchas the Sigma 5, arrive at the Museumin partial condition, lacking essentialperipherals, software, and/ordocumentation, or they are too fragile ordamaged to be handled or used.Collecting a piece of a system such aspart of a CPU or even a set of individualcomponents does not allow the entiresystem to be studied, understood, or

experienced. In this instance, CMU wasoffering to donate all hardware, spareparts, software, and documentation fora system that was in running condition.1

The Sigma 5 could perfectly meet theMuseum’s desire to collect andpreserve artifacts that would provide anaccurate and complete picture ofcomputing technology.

I proposed to the Collections Committeethat we accept the donation of theSigma 5 and all related pieces thatwould provide a complete picture of theSigma 5 system and 1960s mainframecomputing, including items such as themetal file cabinet used to storepunched cards. The committee saw theSigma 5 as a valuable addition to thecollection and agreed that we shouldaccept the system as a whole. Now allthat was left was to arrange de-installation of the system and shipmentto the Museum in California.

TRANSPORTATION AND LOGISTICS

Initial conversations with Bothner-By andDadok indicated there was no rush tomove the system. CMU would wait untiladequate space became available atthe Museum. I began researching thelogistics of actually getting the donationto California.

Then one day I received an urgent emailfrom Bothner-By: the system must bemoved as soon as possible. Themachine room housing the Sigma 5 hadbeen transferred from the chemistry toa different department and wasscheduled to be demolished andremodeled. The Sigma 5, along with allother contents of the NMR Lab, neededto be removed immediately!

This put the transportation planningprocess into hyper-drive. Two significantchallenges had to be faced:transportation costs and preparing thesystem for pickup. Based on our historyof transporting similar systems, weestimated that it would cost about$7,500 to prepare and ship the Sigma5. Unfortunately, the Museum budgetdid not have an allocation for this.However, I knew that Max Palevsky, oneof the founders of Scientific DataSystems, had generously contributed tothe initial founding of the Museum. Iwrote asking if he would sponsor themove and soon received a phone callindicating he would be happy to. Withfunds in hand, we began the task ofplanning the actual move.

For transportation of high-value artifactsthat are fragile, heavy, and bulky, theMuseum uses a carrier with experiencein transporting computers, electronics,and similar equipment. Usually theMuseum receives donations of largesystems that have been de-installedwithout planning for future use or study.Cables are often cut instead of beingunplugged and carefully packed;software and documentation aremissing; and integral pieces of thesystem have been abandoned ordisposed. Even when an item isdonated “intact,” it is often removed orpacked with an eye towards expediencyrather than preservation. For example,packing tape may be directly applied tosurfaces, leaving damaging residue orpulling off paint and surface materialwhen removed.

Given the excellent condition andcompleteness of the Sigma 5, avoiding

PA G E 1 8

C O M P U T E R H I S T O R Y M U S E U MC O M P U T E R H I S T O R Y M U S E U M

PA G E 1 9

C O R E 4 . 1

these mistakes was a very high priority.Having facilitated several systemmoves, I knew that a Museumrepresentative needed to do the actualpreparation of the system for shippingto California. We worked out a date andlogistics for pick-up and began to planthe de-installation process.

THE DE-INSTALLATION

Again my travel plans coincided withMuseum needs and I made a trip toPittsburgh in June 2002. To help the de-install of the Sigma 5 proceed smoothly,a preliminary written plan was createdto document the process with bothphotographs and a log kept in mynotebook computer.

I planned to spend the day documentingthe system configuration, uncabling,packing, and staging the donation forpick-up. Consulting with staff at theMuseum and others who had directexperience with SDS and XDS—Xeroxpurchased SDS in 1969 and named thecompany XDS, or Xerox Data Systems—machines, I learned how to best preparethe system, i.e., what to do or not, whatwas important, and what could be leftbehind. For example, a former SDShardware engineer told me thatconnectors on peripheral cables werevery fragile and would often break whenbeing removed. He provided advice onhow to remove a cable and avoid strainthat could damage it. This advice provedto be invaluable and insured that manyfragile items were handled correctly andwithout damage.

I met Bothner-By in Pittsburgh onThursday morning, May 17, about8:30am. I found the computer room in

the same condition as the year before.The NMR Spectrometer had alreadybeen disassembled, so the control roomequipment and Sigma 5 were left.The plan called to first map the systemin situ, take photographs, and inventoryall the cabinets, spare parts, software,documentation, and other donateditems. I inventoried the major piecesthat would need to be handled bymovers: hardware cabinets, peripherals,and several cabinets holding punchedcards and spare parts. We also did aquick inventory of software stored onmagnetic tape. Tools for disassemblywere located in a helium generationroom down the hall, along with somesystem documentation.

The next step was to pull floor tiles anddo a quick assessment of whichcabinets to uncable first. Yikes!Removing more and more floor tilesrevealed an ever-more complicatedGordian Knot2 of data and power cablessnaking all over the machine room.Usually when de-installing, one wouldtake Alexander the Great’s approach toloosen the original Gordian Knot, butbecause we wanted to preserve theSigma 5 intact, that approach wasavoided. I had an inkling this might takea little longer than I initially thought.

Uncabling the CPU presented severalproblems. I had hoped to just unplugthe power cables and keep them withthe system. However, after tracing thepaths of the primary and peripheralpower cables, I found them runningthrough a hole in the wall into the nextroom that was (of course) locked andinaccessible. If I couldn’t unplug fromthe building power, I’d just disconnect

the hard-wired cables. Although thepower was hard-wired into the main CPUcabinet, it fed power to other cabinetsin the system, which simplified that partof the disassembly.

Fortunately, the breaker box was locatedin the computer room. Beforedisconnecting power, I double-checkedthat the three-phase 408V power to thesystem was turned off at the breakerbox. Then I checked again.

Separating the cabinets from each otherwas even more complicated. Unlikemodern systems where elements suchas disc, memory, tape, and CPU arephysically present in a single cabinet,the Sigma 5 CPU was composed ofthree large interconnected cabinetshousing the CPU, floating point unit, andcore memory. Tens of cables were lacedbetween, through, and under thesecabinets. In addition, cables connectedthe CPU and peripherals, some of whichwere physically next to the CPU, andsome of which were located at adistance across the room.

Today’s cables have connectors on eachend that allow them to be disconnectedquickly and easily. On the Sigma 5, inaddition to the mass of tangled cablessnaking between cabinets, each cablehad a 4x4-inch printed circuit boardhardwired to each end. I soondiscovered it got worse. Some cableshad not just two, but up to five cardshardwired at various points in-betweenthe ends and were 20 or more feetlong. These cards enabled the cables toplug into backplanes or card cages inone or more cabinets.

I saw why most systems of this type areremoved from service by cutting thecables. Finding, untangling, andremoving what looked to be severalhundred cables seemed an impossibletask. It made sense to start where theperipherals attached to the CPU. With alot of patience and gently working withcables that had not been moved in over30 years, I was provided a detailedlesson in 1960s-era mainframepackaging and interconnect technology.

In addition to untangling the cablesbetween the cabinets, it was importantto record exactly where each cable wasplugged, in order to facilitate theeventual reconstruction of the Sigma 5.I defined a labeling protocol based ondesignations present in each hardwarecabinet. Each card plugged into asocket and was tagged with a code thatindicated 1) cabinet, 2) frame, 3) row,and 4) slot.

By the end of the day on Thursday, I hadremoved only about a third of thecables, no software or spare parts hadbeen packed, and I was beginning toworry about completing the task by theend of Friday. By working until 3am, Ifinally disconnected and packed mostdata cables. The next day, Bothner-Byand I started working at 9am with agoal to finish preparing the system by1pm so I could catch my plane thatafternoon. Hope springs eternal. About3pm, it was obvious that there wereseveral more hours of work and Ichanged my departure to Saturdaymorning.

The Computer History Gods must havebeen smiling on us because the

process of separating the system in anorderly fashion became easier andeasier with each cable. The corememory cabinet was the first to becompletely decabled. Once several boltsholding the cabinets together wereremoved, it was moved aside for thefirst time in 30 years. I made a mentalnote that rubber wheels sitting in oneposition for that period of time also tendto flatten, so it was as much pushed asit was rolled.

While I concentrated on physicallyseparating the cabinets, Bothner-By waspacking away documentation and spareparts. Since the lab closure called forremoving all furniture and systems, hejust packed all the spares in place intheir storage cabinets to be shipped toCalifornia. Cabinets of punched cardswere likewise secured and locked. Thesystem documentation was packed intosix large boxes. I looked through a tapelibrary of about 200 half-inch magnetictapes, picked those that appeared tocontain system software, and packedthem into another four or five boxes. In the late afternoon, all cables weredisconnected except those in the CPUcabinet with the operator front panel. Inthe interest of time, these were simplyrolled up and packed in a large box ontop of the cabinet, leaving one end stillconnected. This was OK for transport,and the cables would be removed onarrival at the Museum. When this wascompleted, all cabinets had beenseparated, positioned, and labeled forpick-up the following week. An inventorywas created for the shippers. The de-installation and preparation of theSigma 5 was complete by about 5pm on Friday.

It looked like there would be someunused room in the shipment. Inaddition to the Sigma 5, the machineroom contained a PDP-8 with two seven-track 1/2-inch tape drives. While theMuseum has several PDP-8 systems, Iknew of no operational 7-track tapedrives. Since these are extremely rare,they would be very useful for readingand converting the Museum’s collectionof 7-track tape media. Bothner-By and Itracked down Professor Mort Kaplanwho owned the PDP-8. He confirmed itwas no longer being used and agreed todonate it to the Museum. After a phonecheck with the Collections Staff toconfirm we would accept donation of thePDP-8, it was quickly prepped andmoved into place to piggy-back on theshipment.

At 9pm Friday, I left the Mellon Instituteat CMU, my original eight-hour de-installtask completed after almost 31 hours ofwork in a 48-hour time period. I wasexhausted as I drove to the Pittsburghairport, but still very excited thinkingthat soon the Museum would be inpossession of an almost mint-conditionartifact for the collection.

LESSONS LEARNED

What was learned from this de-install?In addition to learning aboutinterconnect technology on the Sigma 5and navigating the inner sub-basementsanctum of the Mellon Institute, I alsolearned that assumptions based onexperience with a contemporary systemare not always applicable to oldermainframe technology. I was remindedhow important it is to plan your workand work your plan. A standard set oftools can make the job go much faster.

PA G E 2 0

C O M P U T E R H I S T O R Y M U S E U M

PA G E 2 1

C O R E 4 . 1

Talking with subject matter experts forthe Sigma hardware, scoping out thedonation and logistics ahead of time,and lots of patience paid off in spadeswhen it came time to actually de-install,prepare, and move the system. Takingnotes in real-time on a laptop andannotating with photos from a digitalcamera was a big help in gauging myprogress and should greatly facilitatereconstruction of the system in thefuture and with other SDS/XDShardware donated to the Museum. I also learned a lot of dust and debriscan accumulate under a computer room false floor over a 30-year period.

THE SIGMA 5 AT THE COMPUTER

HISTORY MUSEUM

The Sigma 5 was delivered to theMuseum in California about a monthafter I completed the de-installation. Itwas initially stored in the Museum’swarehouse at Moffett Field. Uponarrival, the different pieces in theshipment were entered into theMuseum’s collection database. Thisprocess records a description of eachartifact, physical characteristics, andassigns it a permanent, uniqueaccession number for future reference.

On April 4, 2003, the CPU cabinet withoperator panel, card reader, teletypeconsole, and tape drive were moved tothe Museum’s new building in MountainView California. A display area wasprepared for the Sigma 5 describing thesystem and its capabilities. The Sigma5 is on display for visitors in the newVisible Storage eshibit area along withthe CDC 6600, IBM System 360, SAGE,and 600 other artifacts from thecollection.

WHAT’S NEXT FOR THE SIGMA 5, AND

HOW CAN YOU HELP?

With the Sigma 5 acquisition safely inCalifornia, many possible projects areenvisioned for it. Immediate projectsinclude photographing and making amore detailed catalog of the Sigma 5artifacts. This will allow the Sigma 5 to be displayed via the CyberMuseum at the component and system level. A more detailed catalog needs to becompleted of the Sigma 5documentation, software, and spare parts.

Longer term projects might includereconstituting the Sigma 5 as a runningsystem, incorporating the system aspart of a larger exhibit, scanning theSDS/XDS documentation and ephemeraand making them available via the Web,or creating a working Sigma simulatorwhich could run the software donatedwith the Sigma 5. Completing an oralhistory project of Scientific DataSystems would provide significantinsight into the state of the computerindustry in the 1960s when manybusiness and technology innovationswere realized.

A more formal set of de-installationguidelines for artifacts based on thenotes and log from the Sigma 5 andother artifact donations need to becreated and adopted for futureacquisitions.

As with other Museum projects, thelikelihood of these projects beingimplemented relies on knowledgeable,reliable, and motivated volunteers. Ifyou have an idea for or would like toparticipate in a project involving the

Sigma 5, or other facet of the Museum,as the volunteer coordinator, I will bevery happy to hear from you. Please feelfree to send me an email atcourtney@computerhistory.org.

And as always, the CollectionsCommittee is interested in consideringadditional SDS and XDS artifactdonations that complement the Sigma5, from components to software toephemera to other SDS/XDS systems.

The Computer History Museum isentering a new and exciting phase of itslife. Many pieces are coming together toallow the Museum to build a world-classinstitution documenting the history ofthe information age. Now the Sigma 5,and other artifacts like it, can beproperly preserved for study andenjoyment for past, present, and futuregenerations.

AFTERWARD

Many expectations and goals for thisacquisition were exceeded because ofadvanced planning, foresight of donors,generous contributions, and hard workon the part of volunteers and othersinterested in seeing computing historyrecognized and preserved. Thisacquisition was the result of the effortsof many people.

Thanks go to: Drs. Aksel Bothner-By andJoseph Dadok of the CMU ChemistryDepartment for their time and effort tokeep the system running, recognizingthe value of the system to documentingthe history of computing, and working tofacilitate its contribution to theMuseum; numerous volunteers on theWeb who provided invaluable

Sigma advertisements promoted the machine’s ability to handle real-timeapplications while running background processes (far left) and their abilityto handle more input and output than existing technologies required.“Anything you deliver we can handle,” the ad concludes (immediate left).

SIGMA 5 SPECIFICATIONS

Instruction set: 90 instructions

Word length: 32 bits plus parity, EBCDIC character encoding

Memory: Up to 512K bytes multiport core memory with 2- or 4-way interleaving and 950 nsec cycle time

I/O: Multiplexed IO processor: 8 to 24 channels each with bandwidth of 450 or 900KB/sec and supporting up to 32 standard speed devices each

Optional selector IO processor: 8 to 32 channels each with bandwidth of 3.3MB/sec

Performance Min – Max (all times in usec)AW Add Word: 2.0 – 3.36

LW Load Word: 2.0 – 3.36STW Store Word: 2.5 – 3.94

AWM Add Word to Memory: 3.3 – 5.0BAL Branch and Link: 1.3 – 3.22

Technology: Discrete transistors

Number customer installations: Approximately 250 as of 1972

Purchase price: $90-500K

Installed base markets: Defense, scientific/engineering R&D, space, universities

Primary competitive systems: IBM 360/40, /44, /50, 370/135,/145; DECSystem 1040, 1050, 1055; CDC 3300, 1700; Univac 418-III

System software: Basic Control Monitor, Batch Processing Monitor, and Batch Timesharing Monitor; FORTRAN IV; Assembler; COBOL; scientific libraries

Peripherals: RAD files (fixed head disk), disk, 7- and 9-track magnetic tape, hardcopy TTY and CRT terminals, unit record equipment, paper tape, plotters, datacomm, and A/D & D/A converters

Sources: Sigma 5/8 Sales Guide, Xerox Data Systems, January 20, 1972, and the SDS Sigma 5 Computer Reference Manual, September 1968

PA G E 2 2

C O M P U T E R H I S T O R Y M U S E U M

information allowing me to successfullycomplete this project—in particularKeith Calkins, George Plue, and EdBryan; and Max Palevsky for hisgenerous support of this project andthe Museum in general. Special thanksto the staff in the CollectionsDepartment and Collections Committeewho approved the donation and helpedwith arrangements for its safe transportto California.

When not serving as the Chair of the Museum'sVolunteer Steering Committee, participating innumerous Museum initiatives and projects, andpestering the staff about SDS/XDS items, LeeCourtney is an Engineering Manager at MontaVistaSoftware and harried father of a very energeticthree-year-old boy, who also happens to be theMuseum’s youngest volunteer.

1When the Chemistry Department acquired theSigma 5 in the late 1960s, SDS, and subsequentlyXerox Data Systems (XDS), were contracted toperform periodic maintenance on the system. In1975, Xerox exited the mainframe computerbusiness and sold their customer base toHoneywell. After some false starts, CMU, like manyother Sigma sites, opted for self-maintenance. Thismeant accumulating a supply of spare parts andcollecting the documentation needed to maintainthe system. In addition to self-maintenance ofhardware, CMU opted to maintain its own systemsoftware. This was fortuitous for the Museumbecause it provided a treasure trove of informationto accompany the Sigma 5. Additionally, becauseBothner-By and Dadok took such care inmaintaining and documenting the system, theMuseum and its visitors will benefit from a breadthand depth of equipment and information rarelyavailable for this class and age of system.

2The Gordian knot has come to represent a mostdifficult puzzle. According to Greek legend, anoracle prophesied that the future king of Phrygiawould come into town riding in a wagon. When thepeasant Gordius and his wife did just that, theelders made Gordius king. He dedicated his wagonto Zeus, tying it in front of his palace with anintricate “Gordian” knot as a reminder of hishumble beginnings. Decades later, an oracleforetold that the person who finally unraveled theGordian knot would rule all of Asia. Many, manymen tried to untie the famous Gordian knot untilAlexander the Great drew his sword, slicing theknot in half.

In 1961, Scientific Data Systems (SDS)started with an objective of providingcomputers for the scientific,engineering, and education markets.Many of the founders of SDS hadpioneered this market while at theoriginal Packard Bell Computer Corp. Byusing solid-state but serial logic and aninnovative memory technology cheaperthan core, Packard Bell was able tooffer a machine starting at $40,000.Until then, a price below about$100,000 was difficult formanufacturers to offer so a computerpriced at less than half that opened upnew markets. Crossing a lower pricethreshold—thus attracting newcustomers—has happened severaltimes in the history of computers. Oncehooked, their demand for capabilityincreases, and prices rise once again.

SDS participated in this trend andprovided more capabilities to that samemarket. The Series 9 machines usedcore memory with serial solid-state logicfor an initial price of $54,000 for theSDS 910 in 1962. However, when thecustomer finished acquiring necessaryperipherals and additional memory, theprice would often be two to four timesthat amount. The Series 9 machineswere successfully used in a variety of“embedded” applications where theyprovided control and captured data fromsystems, experimental environments,and other real-time situations.

A desire for computers that couldcombine these tasks with businessapplications led SDS, with its Sigmaline, to focus on multi-use andtimesharing capabilities. To meet theserequirements, SDS invented a memory

map to allow dynamic relocation ofprograms that were being run at thesame time. A fixed-head Rapid AccessDisk with storage of up to 3MBsupported the high-speed swapping ofprograms in and out of memory.Reading or writing up to eight heads inparallel on some models gave a transferrate of 3MB/sec.

In April 1964, IBM announced theSystem/360, which provided many newcharacteristics but not capability fortimesharing or any significant real-timeapplications. The 360 incorporated baseregisters to make programs lesssensitive to their location in memory butdid not provide a means to relocateprograms dynamically. Theseshortcomings gave SDS an opportunitythat it exploited well.

After Sigma architecture was set andthe Sigma 7 announced, othertimesharing endeavors came to light.IBM started a project working withuniversities (Michigan, Carnegie Mellon)to modify a 360/65. UC Berkeleymodified an SDS-930 to provide amemory map and larger memory. SDSproductized and sold that modificationas the 940 to the growing timesharingservice bureau market until the softwarefor the Sigma 7 could be finished.

The Sigma 7 was first delivered inDecember 1966, and the Sigma 5 wasdelivered in August 1967. Thesemachines still used core memory butmore of it, up to 512K bytes. The logicwas now bit parallel for more speed andearly integrated circuits were used forimplementing the working registers. The goal of the Sigma 5 was to provide

real-time and batch simultaneously,allowing batch capability to be used forsoftware development, data processingfrom real-time capture, businessapplications, or processing studentjobs—all while the machine remainedresponsive to real-time needs. TheSigma 5 had the same processingcapability as the Sigma 7 althoughsome of it was optional. It had thesame interrupt system, flexible I/O, andRAD, but it did not have the memorymap. And it found its way intoapplications similar to many of theoriginal 9 series machines but with thebenefit of multi-use.

After a few years of trying to supportthe minimally-configured Sigma 5 at itsstarting price of $90-100,000, basicrequirements were increased and thebase price was more like $160,000.But, still, a typical configuration couldget to $500,000. The Sigma 7, fullyconfigured, could be well over$1,000,000.

This particular cycle started withcomputers that cost only $40,000.Another cycle started when DECintroduced the PDP-8s for $10,000.Even with limited capability, theyenabled a new set of customers toobtain computers for the first time. As time passed, customer demandsincreased and the Minicomputer’scapabilities and cost grew until it toopassed $1,000,000 for someconfigurations. And then came theMicrocomputer or PC…

Pete England was the architect of the SDS Sigmaseries of machines. Ed Bryan and Wendell Shultzdeveloped the operating systems for the series.

THE SDS SIGMA 5IN CONTEXTBY PETE ENGLAND

WITH ED BRYAN AND WENDELL SHULTZ

PA G E 2 3

C O R E 4 . 1

PA G E 24

C O M P U T E R H I S T O R Y M U S E U M

WE HAVE MADE GREAT STRIDES

With the purchase of our new building inOctober 2002 came an exciting changeof direction from the Museum’sprevious plan to build at NASA Ames, achange that enabled us to “step up thepace” and offer more to our membersand to the public than ever before.

And step up the pace we did. InDecember 2002, less than two monthsafter the completed purchase, with helpfrom volunteers and many others, theMuseum relocated to the new building.Construction on a Visible Storageexhibit area and a new auditoriumbegan the following month and wascompleted in May 2003. On June 2,over 600 people gathered to celebratethe opening of the Alpha Phase. See thearticle on page two for more details. Itwas a wonderful evening and gave usall a chance to celebrate how far wehave come in so short a time.

Museum lectures and events continueto feature innovators and champions ofcomputing history. Here are just someof our recent offerings.

ADOBE SYSTEMS—THE FOUNDERS’

PERSPECTIVE

On November 22, 2002, the 20thanniversary year of Adobe Systems,more than 300 people gathered atMoffett Field to hear Adobe foundersJohn Warnock and Charles Geschke in atalk facilitated by Bernard Peuto. Thetwo spoke about the company’ssuccess throughout the years andshared key philosophies and strategiesthat enabled the company torevolutionize desktop publishing.

From PostScript to Illustrator toPhotoshop to Acrobat, Adobe repeatedlyintroduced products for which there wasno market. Warnock said that, “it’sreally important...not to try essentiallyfor today’s market, but to look a coupleyears out and shoot ahead of the duck.”Geschke explained policies and culturethat they believe helped the company toexperience success and longevity. Hesaid, “We wanted to build…a place that,frankly, we would like to work…And wefelt if we did that, we could attract…thegreat engineers, the insightful marketingpeople, the dedicated sales people.”Over the years, when asked to delineate the “Adobe way,” he would reply,

“There’s really only one rule: if you areconfused about how to deal with[someone]…just treat that individual theway you’d like to be treated…and thatwill be the ‘Adobe way.’”

AN EVENING WITH STEVE WOZNIAK

Apple co-founder Steve “Woz” Wozniakengaged an audience of 300 people atMoffett Field on December 10, 2002with personal stories about his childhood—including pranks he used toplay. He recalled his interest inelectronics, the inspiration of Tom Swift books, his success in science fairs, andpositive feedback he received from histeachers and his parents. He relayedhow he became a licensed ham radiooperator in 6th grade. He said he “wentall the way through high school thinking,‘I’m designing computers right and left,but I don’t think they ever have jobsdesigning computers. I mean engineers,which I want to be, you know, theydesign TVs and radios and things.’” Hedescribed how early on he “measuredhimself by how few chips” he used anddiscussed the sequence of break-throughs and events that eventually ledto early Apple designs. To learn more,check out the events section of theMuseum Web site.

REPORT ON MUSEUM ACTIVITIESBY KAREN TUCKER

Karen Tucker is Vice President ofDevelopment, Marketing, and PRat the Computer History Museum

Board Chairman Len Shustek (left) and ExecutiveDirector and CEO John Toole officially ushered in theMuseum’s Alpha Phase in a ribbon cutting ceremonywith the city of Mountain View at the new building.

Adobe co-founder Charles Geschke described hisand Warnock’s efforts to create a company culturethat was positive and respectful of employees,customers, shareholders, and the community alike.

Adobe co-founder John Warnock expressed theimportance of “shooting ahead of the duck” whenintroducing new technology.

Steve Wozniak autographs a BYTE magazine.

Steve Wozniak told stories about a childhood full ofpranks and programming and discussed earlyinnovations at Apple.

Phot

o by

Ear

nest

J.

Ray

PA G E 2 5

C O R E 4 . 1

25 YEARS OF HENNESSY AND

PATTERSON

On January 7, the Museum opened thenew year with a lecture at PARC withJohn Hennessy and David Patterson andfacilitated by John Mashey. More than200 people heard these legendary menspeak about early RISC developmentand their work since that time. Asassistant professors, both men taught“brainstorming” classes to explore newtechnology directions. Hennessyrecalled the class goal of designing aprocessor where they “almost naively”assumed, “given that VLSI is going tobecome the implementation technology,we need to re-look at the question ofhow processors should be architectedwhen they’re not being built from gatearrays or bits-wise kinds of technology.”He added, “the fact that we had alimited number of transistors forced usto think really hard about what belongedin hardware and what belonged insoftware.”

Patterson recalled being visited early onby John Cocke, who encouraged bothhim and Hennessy in their respectivework. Patterson surmised that, althoughCocke never said it, he visited the westcoast to “communicate ideas so theycould figure out how” to create the

chips IBM was somehow not managingto create. “IBM was super-secretive withideas, except Cocke would just go talkwith a bunch of faculty and gradstudents…and get us excited....” “Thiswas very controversial stuff. We did thisas assistant professors….It was heresy.It got emotional reactions.”

Together the two authored ComputerArchitecture: A Quantitative Approach,considered for over a decade to beessential reading for every seriousstudent and practitioner of computerdesign.

HOW DATABASES CHANGED THE WORLD

Chris Date, Herb Edelstein, Bob Epstein,Ken Jacobs, Pat Selinger, Roger Sippland Michael Stonebraker, withmoderator George Schussel gathered onFebruary 10 to share database storiesand lessons learned.

After introductions, Date started thepanel with a memorable acknowledg-ment of relational database pioneerEdgar “Ted” Codd, who was unable toattend (and sadly, passed away just twomonths later on April 18). Date said,“We are all here…in this room becauseof what Ted Codd did back in the 70s”at IBM when he proposed the relationaldatabase model that “basically put thewhole field of database managementonto a solid scientific footing.”

The panel noted the paradigm shift ofrelational databases and told stories ofthe companies and people that developed, used, and marketeddatabase technologies over the pastthree decades. Marketing was noted asof primary importance to successfulproducts and Selinger discussed some of the user testing that IBM underwent in its decision to stick with SQL, apowerful language criticized asproblematic and sometimescumbersome. Epstein predicted thatdatabases will be used more and moreas monitoring systems to processstreaming data, where companies willbe “passing data through queriesinstead of passing queries through data.” Sippl said, “There is a newrevolution coming, not of the algebra ofhow to deal with tables of data, but how

to deal with combinations of businessprocesses….There’s going to be a simple, powerful model for doingthat…that will have as big an impact asthe relational database did. It’s dealingwith our processes, not just our data.”Edelstein concluded that there is andwill continue to be a “vast increase inthe scale of information” beingprocessed, where great “complexitiescome from the nature of data and therules associated with data.” We aregoing to “need a way to deal with thesecomplexities….Someone [will surfacewho will] abstract these newcomplexities into a new paradigm.”

NATURE OR NURTURE: ESTRIN’S LIFE

IN TECHNOLOGY, SO FAR

Judy Estrin, second-generation computerscientist thrice named to Fortunemagazine’s list of the 50 most powerfulwomen in American business, spoke at a Museum event hosted by Microsoft onMarch 5. Long-time friend and venturecapitalist Yogen Dalal masterfullyfacilitated the conversation. Estrindescribed her “rich beginnings” and the“incredible role models in terms ofvalues, ethics, and love of learning” shefound in her parents, Thelma and GeraldEstrin, who worked together to buildIsrael’s first mainframe computer, theWeizac. She said, “most of the peoplewho grew up with me knew me as a

John Hennessy recollected how he got intocomputers in the first place, how RISC developed,and how the technology evolved over time.

Dave Patterson was first introduced to computersone semester in college when he took aprogramming class because all of the mathsections were full.

The database panel discussed the paradigm shiftof relational databases, the incredible growth ofthe database industry, and ideas about the futureof data management.

George Schussel moderated the database panel.

Phot

o by

Ear

nest

J.

Ray

Phot

o by

Ear

nest

J.

Ray

people person and maybe a facilitator,but not necessarily a leader….I’m notsure that anybody would have guessedthat I would have ended up being anentrepreneur. On the other hand,” shecountered, “I think that I perhaps wasdestined to be a technologist” becauseof her family environment, among otherthings.

Estrin, who has co-founded multiplestartup companies and become anexpert in organizational management,described her experiences at severalcompanies over the years. Sheremembers writing her first businessplan on a TRS-80 with cassette tapestorage. At Zilog, she “learned thatmarketing matters.” She had beenquoted from those days as saying that“marketing is an unnecessary evil,” butcame to believe that “technology fortechnology’s sake doesn’t solveanybody’s problems because it nevergets to the customer.” Estrin co-foundedher first startup at age 26, her thirdstartup was acquired by Cisco Systems,and she is presently CEO of PacketDesign, her fourth startup company.

THE ORIGINS AND IMPACT OF VISICALC

On April 8, Dan Bricklin, Bob Frankston,and Mitch Kapor along with CharlesSimonyi discussed the invention ofVisiCalc, the first electronic spreadsheetprogram. Bricklin and Frankston createdVisiCalc and Kapor followed theirinnovation with Lotus 1-2-3.

Bricklin came to understand the needfor changeable content and prototypingthrough observing his dad’s printingbusiness. In college, he daydreamedabout a “magic, typeable blackboardthat would do what I couldn’t, which is,when I made a mistake…in my[spreadsheet] homework, I could erase one number and have [the blackboard]change all the numbers.”

Frankston reminisced aboutprogramming much of the software,creating final code that was just 20KB,including operating system, screen buffer, and disk utilities. Simonyipointed out that today’s Microsoft Excelrequires 8.7MB for the software alone.

Kapor admitted that when Frankstongave him the first demo of VisiCalc, hesaid, “huh?” Then added, “Fortunatelythat was not the last demo I got!” Increating Lotus 1-2-3, Kapor wanted to“design something that could just standnext to VisiCalc without embarrass-ment.” Bricklin observed that in Lotus1-2-3, Kapor kept “all the things [fromVisiCalc] that ended up being the rightthings, as opposed to making itdifferent for the sake of different, andthen added new features” and changedthings that were worth changing.

The lecture was co-hosted and held atMicrosoft in Mountain View. Visitwww.bricklin.com for a personal historyof VisiCalc.

CELEBRATING THE ETHERNET’S 30TH

ANNIVERSARY

On May 22, the Museum and PARC co-hosted a special celebration of“Ethernet at 30.” The event wassponsored by 3Com, Cisco, HP, andIntel, and held at PARC in Palo Alto,Calif. The first panel of speakersincluded inventor Bob Metcalfe andearly Ethernet pioneers Gordon Bell,Judy Estrin, and David Liddle.

Metcalfe remembered that, in 1982, hethought with amazement, “there arepeople buying Ethernet whom I havenever met.” He remembers thinking justfour years later that “there are peopleinventing Ethernet whom I have never met!” He went on to define severalelements that made Ethernetsuccessful, including packets, layering,distribution, the ether itself, and theEthernet “business model,” whichconsists of de jure standards,proprietary implementations, fiercecompetition moderated by a marketcommitted to interoperability, and anevolution over time that preserves thesame base technology. Estrin pointedout the commitment to being a “best”technology, instead of a deterministicone limited by the lowest commondenominator. Liddle gave an interestinghistory of competing standards thatenabled Ethernet, although barely attimes, to continue to succeed. Bell, whowas instrumental in developingcorporate alliances, told stories of hisadvocacy for the technology.

Ann Winblad moderated the secondpanel discussion on the future ofnetworking with industry thought-leadersAndy Bechtolsheim, Eric Benhamou, W. Eric Mentzer, and Stephen Squires.Bechtolsheim expressed his belief thatthe hardware and technology evolution

PA G E 2 6

C O M P U T E R H I S T O R Y M U S E U M

Venture capitalist Yogen Dalal facilitated aconversation with Judy Estrin. The first time he mether, he was “struck by her incredible amount ofenergy...and adventuresome spirit.”

Mitch Kapor followed on VisiCalc with Lotus 1-2-3,wanting to do something that “took it to the nextlevel.”

Dan Bricklin (left) and Bob Frankston (right) created VisiCalc, the first electronic spreadsheet program.

Judy Estrin has been a role model for women inbusiness, and always just “assumed” she wouldsucceed in her efforts. She gives credit to her “richbeginnings” within a creative technology family.

Phot

o by

Rob

ert

Hul

lPh

oto

by R

ober

t H

ull

Phot

o by

Mic

hael

Bax

ter

Phot

o by

Mic

hael

Bax

ter

PA G E 2 7

C O R E 4 . 1

will continue, but that “the challengebefore us is in software, at theintersection of computing andconnectivity.” Benhamou reminded usthat Aristotle defined ether as thesubstance that surrounded the earth.This vision of “pervasive connectivitywill be ever more relevant as we headinto the 21st century,” he said. Squirescharacterized Ethernet as an “enduringand multidimensional invention” andenvisioned the next 30 years supportingthe successful implementation of nano-technology in every aspect of our lives.It “won’t be just a little more of thesame,” he said, “but a leveraging” ofcurrent advances—molecularelectronics, scalable modules, newabstractions and visualizations to hidethe details, systems growth, andapplications development—to createcomprehensive change.

Mentzer reflected on the continuing “convergence of the digital and realworlds” and concluded the program withhigh praise for Ethernet, asking, “Howmany technologies do you know that are30 years old that you still want to beworking on?”

FIST LECTURE EVENT IN THE

MUSEUM’S NEW BUILDING

On June 10, the Museum christenedthe Hahn Auditorium with its first panelevent in the new building: “JurassicSoftware: A Look Back at TheBeginnings of Consumer Software.” Onthe panel, Intuit co-founder Scott Cook;Broderbund Software co-founder DougCarlston; and Electronic Arts and 3DOfounder Trip Hawkins reminisced aboutthe early days and recalled lessonslearned in the founding of a newindustry. Stewart Alsop, venturecapitalist, former P.C. Letter publisher,and Demo and Agenda conferencefounder, moderated the livelydiscussion.

Attendee Bob Glass remarked, “It wasexciting to hear the ‘grand masters’ talkabout the anecdotes that formed thebeginning of personal computing! It puta human touch to my knowledge of theevolution of consumer software.”

John Wharton said, “What fascinatedme at the Consumer Software panelwas hearing the inventors themselvesdescribe the context of their work—whatthey’d been doing when inspiration hit,how they fleshed out their basicconcepts, and who helped them honetheir ideas to create a successfulproduct. You can’t get stuff like thatfrom books”

VOLUNTEERS AND THE MUSEUM: CAN’T

HAVE ONE WITHOUT THE OTHER

The Computer History Museum’svolunteers are the lifeblood of theorganization! Behind-the-scenesvolunteers contribute many hours oftheir time helping with a diverse array ofprojects. Some help plan new displays,

assist with the Web site, or arrange forfuture exhibits. Some help withadministration at the Museum’s office.Still others open the Museum to sixpublic tours each week as well as manyspecial tours and scheduled events.Since moving to the new Shorelinelocation and opening the Alpha Phase,more than 50 dedicated volunteershave been involved with the staff andtheir projects.

Also since our move to the newbuilding, 20 new docents have enabledus to increase the number of tours fromtwo to six per week. Tours are nowavailable Wednesday, Friday, andSaturday afternoons. Thanks toMuseum volunteers who have taken thetraining to become docents andgreeters—they independently run publictours, welcome visitors, handle sales inour Museum store, and make thecollection come alive.

The Volunteer Steering Committee, arepresentative group that meets withstaff regularly, is the voice of thevolunteers. The group is headed by LeeCourtney, who serves as chairman.Send requests for information orcomments on the volunteer program toLee Courtney or to Volunteer ProgramManager Betsy Toole. For thoseinterested in helping, please visitwww.computerhistory.org/volunteer andclick on “Become a Volunteer.”

YOUR SUPPORT COUNTS

Our members and supporters areproviding faithful and generous supportfor the Museum in spite of thechallenges of the current economy. Seethe list of people on page 28 who arecurrent members at the $100 andabove level.

If you have been contemplating joiningor lending your financial support to theMuseum, we encourage you to “takethe plunge!” Not only can you enjoy thebenefits of being associated with thisgreat institution, you can take pleasurein your support of important work topreserve and present key objects andstories of this amazing time in history.Please contact me if I can help you inany way. Thank you!!

Bob Metcalfe celebrated the 30th anniversary ofEthernet on May 22. Other presenters includedGordon Bell, Judy Estrin, David Liddle, AndyBechtolsheim, Eric Benhamou, W. Eric Mentzer, andStephen Squires.

The Museum held its first lecture in the HahnAuditorium on June 10, 2003. A panel on earlysoftware included Scott Cook, Doug Carlston, andTrip Hawkins, with moderator Stewart Alsop.

Cake to celebrate the anniversary of this long-livedand successful technology

Phot

o by

Mic

hael

Bax

ter

PA G E 2 8

C O M P U T E R H I S T O R Y M U S E U M

THE CORE CIRCLECORE BENEFACTOR64K ($65,536+)Gwen & C Gordon BellL John Doerr & Ann DoerrDonna DubinskyElaine & Eric HahnJeff Hawkins & Janet StraussGardner Hendrie & Karen JohansenDavid & Karla HouseLeonard J ShustekCharles Simonyi

CORE PATRON32K ($32,768+)Bill & Roberta CampbellPierre & Pam OmidyarJohn & Sheree Shoch

CORE SPONSOR16K ($16,384+)Grant & Dorrit SaviersPeter HirshbergMike & Kristina HomerThe Krause FoundationJohn Mashey & Angela Hey

CORE PARTNER8K ($8,192+)Eric BrewerLisa & Ed ColliganDavid N CutlerSonja & William DavidowSusan & Phil GoldmanDraper Fisher JurvetsonSheldon Laube & Dr Nancy EngelAudrey MacLean & Mike ClairThe Markkula FoundationIke & Ronee NassiJim & Stephanie NisbetSuhas & Jayashree Patil Family FundSteve Perlman & S&i DobrowolskyBernard L PeutoPierluigi Zappacosta & Enrica D'Ettorre

THE CORE CLUB CORE INVESTOR4K ($4,096+)Paul & Evelyn BaranJohn Seely Brown & Susan HavilandPeggy Burke & Dennis BoyleRobert Garner & Robin BeresfordPeter & Louise HammNancy & Tim HowesJ Burgess JamiesonDeborah Meredith & Curtis ColeElizabeth & Paul MockapetrisDave & Jan RossettiRudolph & Sletten, IncCharles L SeitzJohn & Elizabeth TooleWakerly Family FoundationWade & Brenda WoodsonCindy & Peter Ziebelman

CORE CONTRIBUTOR2K ($2,048+)AnonymousSally M Abel & Mogens LauritzenAllan AlcornAllegis CapitalFrances AllenDr Gene & Marian AmdahlAsset Management CompanyCarol A BartzCarolyn Bell & Forest BaskettMichael & Sharon BlasgenJack & Casey CarstenNed & June ChapinBruce & Gail ChizenEd & Patti ClussJohn & Norma CrawfordYogen & Peggy DalalSharon Brunzel & Stephen DeeringWhitfield Diffie & Mary Lynn FischerJudy Estrin & Bill CarricoLori & Joe FabrisFederico & Elvia FagginJudi & Mike FalarskiTracey Stewart & Barry James FolsomEd Frank & Sarah RatchyeBob FrankstonBuck GeeCharles & Nancy GeschkeLorraine Hariton & Stephen WeylChristine Hughes & Abe OstrovskyMatthew & Connie IvesErnest E Keet

Steve & Michele Kirsch FoundationSandra Kurtzig & Carl BrunstingLarry Kwicinski & Nanci CaldwellRonni & Bernie LaCrouteJan & Sofia LaskowskiLaura & Gary LauderMarc & Marian LeBrunShawn & Douglas MacKenzieJames & Patricia MarkevitchRobert & Katherine MaxfieldThe Avram Miller Family FoundationGerhard & Carol ParkerBill & Rochelle PrattDennis RitchieToni & Arthur RockSaal Family FoundationJean SammetAlvy Ray & Zu SmithStephen Squires & Ann Marmor-SquiresKaren & Mark TuckerFloy & Willis WareGio & Voy WiederholdPaul Winalski

CORE FRIEND1K ($1,024+)AnonymousDavid & Robin AndersonThe Dennis & Janet Austin FundConnie & Charlie BachmanSheila & John BanningAllen Baum & Donya WhiteBruce G & Leona D BaumgartBarry & Sharla BoehmGary BooneJerry BrenholzSusan & David BurwenCheck Point Software Technologies IncRichard J ClaytonR T CosletAndrea Cunningham & Rand SiegfriedBill DanielsonEleanor & Lloyd DickmanLes & Marian EarnestDavid EmersonAmy Roth & Robert EpsteinThelma & Gerald EstrinJohn GallowayThe Bill & Melinda Gates FoundationE Michael GreenawaltGregory GretschDave & Dee GustavsonJudy & Marcian HoffGeorge & Emily JaquetteCraig JensenJerry JensenDerry & Charlene KabcenellJerry & Judy KleinDr & Mrs Leonard KleinrockDonald & Jill KnuthTom Kopec & Leah CarneiroDavid & Grayson LaneLorie Family FundRon & Deb MarianettiJames McElweeLore Harp McGovern & Patrick McGovernRobert MetcalfeTashia & John MorgridgeJoan Myers & Stanley MyersJeffrey Berg & Debra PagetPaul PierceJames N PorterBrian D QuintRobins, Kaplan, Miller & Ciresi LLPThe Rockwell International Corporation

TrustPeter & Valerie SamsonSecure ComputingStephen & Paula Smith Family FoundationSteve SommerBob & Verna SpinradGeorge SpixRobert & Lee SproullSRI InternationalNaoya & Masako UkaiLarry & Dawn WeberSusan & Anthony WoodXerox Palo Alto Research CenterJanet & Del YocamWarren Yogi

GENERAL SUPPORTERSCHAMPION MEMBERS ($500+)Kathleen & Duane AdamsAutodesk, IncDavid BabcockGene P CarterLori Kulvin CrawfordStephen Crocker

Claudette & Henry CrouseVicki & Drew DuseboutBarbara & Joseph FredrickOliver FriedrichsFrank S GreeneGeri Hadley & Ron CoteWilliam R Hearst IIIWilliam HoltzmanJennifer & Chuck HouseJohn ImpagliazzoElizabeth & Dan JickRobert Kahn & Patrice LyonsMarlene & Jeffrey KalbLaurel & Ray KaledaFrank McConnellRandall & Julaine NeffDonald & Helen NielsonPerforceJay PerryArati Prabhakar & Patrick WindhamJudith & Steven PuthuffCatherine E Rossi-RoosJohn & Christine SanguinettiVirginia ShulerZalman SternBob SupnikBrad TempletonDuane & Lorna WadsworthMichael WeaverJim & Sylvia Work

SUPPORTING MEMBERS ($100+)AnonymousRhonda AbramsGeorge B Adams IIIJim AitkenSaul & Irene AmarelMelissa Anderson & Howard LookMary Artibee & Milt MalloryGreg AstfalkRebecca Gurley BaceJohn BackusBob & Jean BanningEric BarbourRoss BassettDavid & Bonnie BellAnn & Sanford BenettRobert N Blair, M.Inst.PDaniel BobrowFrederick & Nancy BrooksJack BurnessJerome CalvoPaul CeruzziAlison Chaiken & Wolfgang RupprechtBrian ChaseGeorge Comstock & Anne HillmanHarlan CrowderWilliam CulbertsonDeborah & Michael CussenPaul DaliBarton DarnellLillian & Alfred DubinskyMarlene & Duane DunwoodieJohn EhrmanDavid EllsworthSarah & David EpsteinMatthew EttusJoe EykholtElizabeth FeinlerDaniel FentonLon FialaChris FralicAlan FrindellEnid & Bernard GallerVictor GalvezKaren Tierney-Gay & Riki GayLance & Wendy GlasserSteve GolsonPaul & Beatrice GomoryGeorgia R Grant & Paul V AtkinsDouglas GreigRobert & Marion GrimmLaura & Stephen GrubbForrest GunnisonKathleen & Philip GustMichael GustafsonRollin C HardingAlys HayJohn & Andrea HennessyDaniel HillWinston HindleHarold HolbrookMarc W HowardDave & Sharon HoytV Bruce & Debra M HuntHyperion Solutions CorpDina & Neil JacobsonSteve JasikCathryn Jenkins

Luanne JohnsonCurtis Jones & Lucille BooneWilliam JordanLily JowRandy KatzJakob NielsenThomas & Mark KorneiEric J KorpelaRobert KresekKevin KrewellWinston KrigerThomas KurtzRobert H LashleyDavid A LawsDeborah LevingerJefferson LillyEva & Michael LoebLiza LoopAlice & Donald LoughryCarl & Claudia LowensteinJim & Virginia LyonsJohn Maloney & Roxanne Guilhamet

MaloneyPatricia & Dwight ManleyJulius MarcusRobert MarshallCraig MathiasGeorge MaulJames P MayerStanley & Maurine MazorBrion McDonaldPaul & Raquel McJonesPatricia & Patrick McVeighWilliam D & Dianne Mensch, JrDavid MillerCelia & Gary MillerJim MitchellTim MitchellJeff & Jennifer MoffattMichael MorgansternCarol NewtonDean NicholsRonald NicholsonMarilee J NiemiLandon NollDavid NovakTim OrenTed PanofskyDonn B ParkerDoug & Shirley PearsonMichael PiqueNancy Mason & Christopher PodaPolytechnic Alumni Association of

Northern CaliforniaJames & Betty QuinnMichael M RobertsRon RohnerHeidi Roizen & David MohlerLynn & George RossmannJeffrey RudyKathleen L RydarRita Seplowitz SaltzScott & Jennifer SandellRex SandersSandon AssociatesMorey & Barbara SchapiraJudith Schwartz & William ConlonCatherine Povejsil & William ScullEric ShepherdRichard ShoupHoward S ShubsAngela SidddallDan & Karon SiewiorekSilicon Valley Forth Interest GroupHorst SimonChris & Jade SimonsonPeter SkillmanMatt Chew SpenceDavid M StearnsNina & James SteinerSteven StepanekGreg StikeleatherRichard SweetLarry Tesler & Colleen BartonGeoffrey ThompsonIlene Chester & Frank TobinPhilippe ToblerJoseph Traub & Pamela McCorduckLydia VarmazisGreg & Susan WagemanRoy & Verna WeaverCarol WelshSusan & Stephen WildstromKevin Wilke & Valerie ShawMalcolm WingDonna YobsJohn G ZabolitzkyJohn & Mary ZolnowskyGeorge D Zuras

ENTERPRISE50K ($50,000+)HP

SYSTEMS20K ($20,000+)MicrosoftRambusSun Microsystems

NETWORK10K ($10,000+)3COMAdobe SystemsCisco SystemsConvergysIntel MaxtorPeopleSoftSigma PartnersThomas Weisel Venture Partners

INFRASTRUCTURE5K ($5,000+)Analog DevicesEHDDFish & NeaveGoogleIBM Mid Peninsula BankRudolph and SlettenSendmail

PETA DONORS$10 MILL IONElaine & Eric Hahn

TERA DONORS$5—9.99 MILL IONL John Doerr & Ann DoerrDonna DubinskyJeff Hawkins & Janet StraussGardner Hendrie & Karen JohansenDave House Family FoundationIntuit, in honor of founder Scott CookLeonard J Shustek

GIGA DONORS$1—4.99 MILL IONGwen & C Gordon BellMike & Kristina HomerJohn & Sheree Shoch

MEGA DONORS100K—999K ($100,000—999,999)Steve Blank & Alison ElliottBill & Roberta CampbellSheldon Laube & Dr Nancy EngelJohn Mashey & Angela HeyIke & Ronee NassiBernard L PeutoSaal Family FoundationGrant & Dorrit SaviersPierluigi Zappacosta & Enrica D'Ettorre

KILO DONORSTO 99K (TO $99,999)Sally M Abel & Mogens LauritzenDavid & Robin AndersonPeggy Burke & Dennis BoyleLori Kulvin CrawfordAndrea Cunningham & Rand SiegfriedDavid EmersonCarol & Chris EspinosaSamuel H FullerJennifer & Chuck HouseChristine Hughes & Abe OstrovskyPierre & Pam Omidyar

Current as of September 1, 2003. To correct your listing, please contacthenderson@computerhistory.org.

For information about individual andcorporate memberships, sponsorships,and other giving opportunities, pleasecontact Karen Tucker.

THANKS TO OUR SUPPORTERS We thank our members and donors for their loyalty and enthusiasm and look forward to working with new friends as we build for the future.

ANNUAL DONORS CORPORATE MEMBERS

CAPITAL CAMPAIGN

PA G E 2 9

CONTACT INFORMATIONEXECUTIVE STAFF

JOHN C TOOLE

Executive Director & CEO+1 650 810 1000toole@computerhistor y.org

MICHAEL FALARSKI

Vice President of Operations & Facilities+1 650 810 1001falarski@computerhistor y.org

KIRSTEN TASHEV

Director of Collections and Exhibitions+1 650 810 1037tashev@computerhistor y.org

KAREN (MATHEWS) TUCKER

Vice President of Development,Marketing, & PR+1 650 810 1011tucker@computerhistor y.org

MICHAEL R WILLIAMS

Head Curator+1 650 810 1024williams@computerhistor y.org

OTHER FULL-TIME STAFF

SHARON BRUNZEL

Archivist+1 650 810 1016brunzel@computerhistor y.org

MARLEN CANTRELL

Of fice Manager+1 650 810 1018cantrell@computerhistor y.org

JENNIFER CHENG

Events Coordinator+1 650 810 1019cheng@computerhistor y.org

JEREMY CLARK

Registrar+1 650 810 1020clark@computerhistor y.org

PAM CLEVELAND

Events Manager +1 650 810 1021cleveland@computerhistor y.org

WENDY-ANN FRANCIS

Accountant+1 650 810 1023francis@computerhistor y.org

CHRIS GARCIA

Historical Collections Coordinator+1 650 810 1041garcia@computerhistor y.org

CATRIONA HARRIS

Development & PR Associate+1 650 810 1036harris@computerhistor y.org

LAURA HENDERSON

Associate Director of Development+1 650 810 1029henderson@computerhistory.org

DAG SPICER

Curator of Exhibits+1 650 810 1035spicer@computerhistor y.org

MIKE WALTON

IT Director+1 650 810 1040walton@computerhistor y.org

PART-TIME STAFF

LEE COURTNEY

Volunteer Coordinator cour tney@computerhistor y.org

DAVID CANTRELL

Oral Histories Intern+1 650 810 1017dcantrell@computerhistory.org

ALEX CHAN

Accounting Intern+1 650 810 1025chan@computerhistory.org

SOWMYA KRISHNASWAMY

Database Services Intern+1 650 810 1028krishnaswamy@computerhistory.org

BETSY TOOLE

Volunteer Program Manager+1 650 810 1038etoole@computerhistor y.org

VINH QUACH

IT Suppor t+1 650 810 1039quach@computerhistor y.org

COMPUTER HISTORY MUSEUM

1401 N Shoreline BlvdMountain View, CA 94043-1311, USA+1 650 810 1010+1 650 810 1055 (fax)

WWW.COMPUTERHISTORY.ORG

Current staf f openings can be foundat www.computerhistor y.org/jobs.

UPCOMING EVENTSPlease visit www.computerhistory.org/events formore event details and to RSVP for all events. More information is available at +1 650 810 1013.

THU, SEPTEMBER 25THREE DECADES OF INNOVATION:PHIL IPPE KAHN’S PERSONAL STORIES Hahn Auditorium, Computer History Museum

TUE, OCTOBER 21FELLOW AWARDS CELEBRATIONHahn Auditorium, Computer History Museum

WED, OCTOBER 22TIM BERNERS-LEEHahn Auditorium, Computer History Museum

WED, NOVEMBER 12SUHAS PATILHahn Auditorium, Computer History Museum

WED, DECEMBER 3DON KNUTH: A DOZEN PRECURSORS OFFORTRANHahn Auditorium, Computer History Museum

MUSEUM ARTIFACTSON LOAN

ONGOING (2003)“THE COMPUTING REVOLUTION”

Museum of ScienceBoston, Massachusettswww.mos.org

TOUR THE MUSEUMTours of Visible Storage are nowavailable on Wednesday, Friday, andSaturday afternoons at 1:00pm and2:30pm. Tours take about an hour.

Please make your reservation by calling+1 650 810 1013 or emailing:tours@computerhistory.org.

VOLUNTEER OPPORTUNITIESThe Museum tries to match its needswith the skills and interests of itsvolunteers and relies on regularvolunteer support for events andprojects. In addition to special projects,monthly work parties generally occur onthe 2nd Saturday of each month,including:

OCTOBER 11, NOVEMBER 8,DECEMBER 13

Please RSVP at least 48 hours inadvance to Betsy Toole for work parties,and contact us if you are interested inlending a hand in other ways! For moreinformation, visit our web page atwww.computerhistory.org/volunteers.

C O R E 4 . 1

FELLOW AWARDS CELEBRATION 2003

Explained from CORE 3.3

This small section of the British DEUCE(Digital Electronic Universal ComputingEngine) computer constructed at theEnglish Electric Company is approx-imately 2 lbs. and 4 1/8" x 8" x

2 15/16" (HWD). This three-vacuum-tube module formed part of themachine’s mercury delay line memory,which translated a digital pulse traininto sound waves, sent these wavesdown a tube, then recirculated thewaves back through the tube.

The English Electric Deuce was ageneral-purpose vacuum tube digitalcomputer, with a serial organization anda 1 MHz clock rate. It was a re-engineered Pilot ACE (AutomaticComputing Engine), a landmark machineconceived but unrealized by Alan Turingand developed by the UK NationalPhysical Laboratory. The ACE can beseen today at the Science Museum,London. The DEUCE contained 1,450vacuum tubes and was nearly twice thesize of the ACE prototype.

The DEUCE’s word length was 32 bits,and its arithmetic units were capable ofperforming single, double, and mixedprecision binary integer arithmetic. Thefast main memory was comprised of 12mercury delay lines. Eight delay lines

held executable instructions, and fourdelay lines comprised auxiliary storage.The magnetic recording drum (anexample of which resides at theComputer History Museum) containedone block of 16 read heads and aseparate block of 16 write heads. Eachhead provided access to a track of 32words and both blocks could be movedindependently to any of 16 positions.

The DEUCE used standard Hollerith(IBM) 80-column punched cardmachines. Reading and punchingtransferred one binary word per row of acard and conversion to and fromdecimal was performed by software.

The first machine was delivered in thespring of 1955. From late 1955onwards, English Electric began sellingthe DEUCE 2, followed in 1957 by aDEUCE 2A; these featured, among otherthings, re-engineered input/outputsystems. The company sold about 31DEUCE 1 and 2 machines between1955 and 1964, priced at around50,000 UK Pounds in 1958.

Please send your best guess tomystery@computerhistory.org before12/15/03 along with your name, andshipping address. The first three correctentries will each receive a free travelmug with the Museum’s logo.

1401 N Shoreline Blvd, Mountain View, CA 94043 USA

Address Service Requested

NONPROFIT ORG

U.S. POSTAGE

PAID

MOUNTAIN VIEW, CA

PERMIT NO. 50

MYSTERY ITEMSFROM THE COLLECTION OF

THE COMPUTER HISTORY MUSEUM

WHAT ISTHIS?THIS ITEM WILL BE EXPLAINED IN THE

NEXT ISSUE OF CORE.

English Electric, DEUCE Mercury Delay LineAmplifier Circuit (1955), XD 4.75, Gift of MurrayAllen.