Making Core Memory: Design Inquiry Into Gendered
Making Core Memory: Design Inquiry into GenderedLegacies of Engineering and CraftworkDaniela K. Rosner1, Samantha Shorey2, Brock Craft1, Helen Remick3123Dept. of HCDEDept. of CommunicationSeattle, WAUniversity of WashingtonUniversity of Washingtonhremick@msn.comSeattle, WASeattle, WA{dkrosner, bcraft}@uw.edusshorey@uw.eduABSTRACTThis paper describes the Making Core Memory project, adesign inquiry into the invisible work that went into assembling core memory, an early form of computer informationstorage initially woven by hand. Drawing on feminist traditions of situated knowing, we designed an electronic quiltand a series of participatory workshops that materialize thework of the core memory weavers. With this case we notonly broaden dominant stories of design, but we also reflecton the entanglement of predominantly male, high statuslabor with the ostensibly low-status work of women’shands. By integrating design and archival research as ameans of cultural analysis, we further expand conversationson design research methods within human-computer interaction (HCI), using design to reveal legacies of practiceelided by contemporary technology cultures. In doing so,this paper highlights for HCI scholars that worlds of handwork and computing, or weaving and space travel, are notas separate as we might imagine them to be.Author KeywordsWoven memory; gendered labor; craft; handwork; computing history; participatory workshops.ACM Classification KeywordsK.4.0 Computers in Society: general.INTRODUCTIONOur methods of inquiry shape not only how we do designand technology development but also how we come to understand who counts as a designer and what counts as design practice, extending the very definition of design. Justas HCI’s methodological toolkits have continued to expand— enrolling a wider array of instruments, techniques, 58,60,67] — our definitions of technology havePaste the appropriate copyright/license statement here. ACM now supportsthree different publication options: ACM copyright: ACM holds the copyright on the work. This is thehistorical approach. License: The author(s) retain copyright, but ACM receives an exclusive publication license. Open Access: The author(s) wish to pay for the work to be open access. The additional fee must be paid to ACM.This text field is large enough to hold the appropriate release statementassuming it is single-spaced in Times New Roman 8-point font. Please donot change or modify the size of this text box.Each submission will be assigned a DOI string to be included here.Figure 1: Close up view of the Core Memory Quilt.broadened, too. From establishing the Jacquard loom as aprecursor to the Babbage Analytical Engine [66] to recalling that the first “computers” were young women[11,19,31,40], gendered narratives of craftwork and engineering both haunt and inform HCI’s ideas of technologicalbelonging, participation, and differentiation [44].Our focus in this paper is the gendered forms of craftworkunderlying digital production and their valuation as technical work. We explore our own handwork as HCI designers in relation to the people we acknowledge as central contributors to engineering innovations. We describe our process of collaboratively making a historically-informed design artifact that reconstructs the story of magnetic-corememory. Core memory constitutes one of the principalmechanisms by which computers stored and retrieved information during the first two decades of the Cold War. TheApollo mission computers, for example, stored informationin core memory ropes: threaded wires, passed through oraround magnetized rings. NASA engineers nicknamed thishardware “LOL memory” for the “Little Old Ladies” whocarefully wove wires around small electro-magnetic ferritecores by hand. Later versions of hardware required evensmaller weaving instruments and microscopes. But scholarsstill know little about the core memory weavers: what theirwork looked like or what they contributed to space travel.Drawing on feminist, historical, and designerly approaches[18,27,53,55,63], this paper details the development ofMaking Core Memory, a project of design inquiry that un-
folds in two parts: first, the making of an electronic quilt(Figures 1 and 2) composed of core memory planes that actas quilt patches; and second, the presentation and engagement of the quilt in workshops. During three workshopsacross the west coast of the United States (Mountain View,CA, Seattle, WA, and Los Angeles, CA) we asked a rangeof historians of technology, design educators, and membersof the public to help us materialize the work of the corememory weavers by weaving core memory “patches” andrevisiting the weavers’ history together. We gave participants “patch kits” comprising a simple metal matrix, beadsand conductive threads (in place of ferrite core and wire).Plugging the completed patch kits into the electronic quilt(Figure 2) triggered the quilt to play firsthand accounts of1960s core memory production while sending tweets viaour @lolweavers account. This approach allowed us tobring hidden stories of innovation work to the people building tools, histories, and pedagogies for design—recontextualizing their own practices within stories of handwork heretofore eclipsed.The paper that follows contributes to HCI scholarship alongthree key dimensions. First, by recognizing the hidden,feminized work involved in specific contexts of 1960s corememory production, we comment on the particular legaciesof expertise and knowledge on which core memory workrelied. This analysis moves beyond HCI’s dominant focuson material capacities (what a technology can do) to focusinstead on the histories of practice that make those capacities possible. Second, our work contributes new perspectives on the process of building embodied, experientialknowledge. Specifically, we offer a different conceptualframe that broadens ongoing HCI conversations on digitallabor and handwork. We shift debates from an assumedseparation between cognitive (masculine, innovative, highstatus) and manual (feminine, menial, low-status) labortoward an active examination of their entanglement. Third,this paper highlights the centrality of feminist historicalperspectives, and argues for historical-participatory inquiryas a core approach by which different practices, ideas, andethical stances get recognized and examined. Sites of digitalmanufacturing require new forms of inquiry that challengethe current valorization of individual innovators and users.This observation acts as a direct call for HCI scholars totake up new collective methods that expand the ethical andhistorical horizons of contemporary design research.LITERATURE REVIEWCraft practices like weaving, sewing and other forms oftextile making embody processes of creative productionhistorically associated with women [2]. Within public narratives of technology and engineering, journalists and pundits tend to depict craft expertise as menial labor traditionally associated with women that is inherently different from(and often less valuable than) the sophisticated cognitivelabor of engineering associated with men. Despite this continued bifurcation of manual and cognitive work, the development of textile crafts has long occupied a crucial place inFigure 2: Plugging a woven patch onto our Quilt.the disruption of women’s exclusion from science and technology cultures, both presently and in historical narrative.Notably, women have encoded critical messages duringtimes of distress, particularly around quilting [43,64,68].Using handwork as a device for encryption frames textileartifacts as complex forms through which to catalyze technical and strategic innovations.Within CSCW and HCI, similar techniques for technicalinvention and encoding through craft have informed a robust body of work on digital manufacturing. One strand ofthis work examines the forms of human-machine collaboration that emerge with the introduction of digital fabricationtools within studio-based arts practice [10,16] and othersites of traditional craftwork [3,35,54,57,59]. Devendorfand Ryokai’s Being the Machine project [16], for instance,challenges a stable division of labor between maker andmachine by altering the coordination between them. In other work, HCI researchers explore contributions of craftwithin emerging forms of digital cultural heritage[24,51,52], including core memory [15]. Work by Petrelli,Ciolfi and others [12,51,52] draws on textile traditions toexplicitly identify embodied practices as key modes of HCIresearch.A related strand of HCI work has examined techniques oftextile production as innovation work, broadening the historical resonances and present-day possibilities of dexteroushandwork traditions. While some position historical machines such as the groundbreaking Jacquard loom runningon punch card technology as important sites for investigating embodied interaction [21], others consider platforms forelectronic-textiles making as mechanisms for interveninginto [48–50] and even improving [7,8] environments ofSTEM learning. Leah Buechley, for example — known forher origination of a sewable microcontroller that connectscircuitry with conductive thread instead of wire — has argued that sewing circuitry provides a case for shifting metaphors of engineering development from brittle and mechanical solutions toward open-ended possibilities [8]. Thisis especially true for women, 40% of whom come to technology innovation spaces from a background in arts &crafts (rather than engineering) [20].In what follows, through the confluence of historical analysis, quilting, and participatory workshops, we draw togetherthese separate strands of work to investigate the influence
of craftwork on the methods, processes, and narratives ofengineering that continue to inform programs of HCI research. We show that revisiting craft-based practices withinand as part of engineering histories may challenge thosehistories, enlivening new features of the technological past(revealing links between weaving and engineering inventions, for example). But it may also expand conversationson design research methods more broadly. This work involves engaging new histories of production that set hardware inventions in motion—reconsidering not only whatcounts as innovation work but also who does it and how.BACKGROUNDHCI’s cases of hardware development tend to focus on cutting edge tools, methods, and infrastructures. Yet, sites ofcollective digital manufacturing have a much longer historyand reveal deeper insights about how the innovation ofthings unfolds [25]. Apollo 8, the first “manned” mission tothe moon and back, required the work of over 400,000 people—a collaboration that popular discourse often elides inits tendency to valorize the work of individuals [22]. Belowwe contextualize the development of our Core MemoryQuilt in this history, beginning with the NASA mission thatcatalyzed a new reliance on weaving techniques for digitalinformation storage.Apollo MissionThe design of the Apollo Guidance Computer (AGC) beganin 1961 [69]. Among the greatest challenges faced by theteam of Apollo engineers was building one of the world’sfirst portable computers [41]. The Apollo missions neededan on-board guidance system that could direct the spacecraft, independent of mission control stations on Earth [42].Room-size machines running on punch cards, equipmentthat was too heavy and too large to fit in the cone of a rocket, dominated computer technology in the 1960s. One of thekey solutions was a form of information storage called“core rope memory.”Core rope memory is a mechanism for information storagethat uses wires running through or around magnetic ferritecores to create binary “zeros” or “ones.” During the early1960s, female line workers (the “Little Old Ladies,” as engineers called them) assembled the AGC code by hand inWaltham, Massachusetts. Employed by the Raytheon Corporation, they sat across from one another at long desks,passing wires back and forth through a matrix of eyeletholes, each comprising a magnetic core bead. Passing awire through the core created a “one,” while bypassing thecore created a “zero” (see Figure 3) [13].Although astronauts wouldn’t touch the moon’s surfaceuntil Apollo 11, each of the Apollo missions was defined bythe need to leave Earth’s atmosphere and travel an extraordinary distance with living humans onboard [69]. Amongthe many unique environmental factors facing a computerleaving the Earth’s atmosphere, the AGC had to withstandthe extreme vibrations of takeoff and potential power lossduring the mission [69]. Rope memory answered the needfor information storage that fit the temporal and spatial constraints of the storage environment—extreme cold, intensevibrations—while also responding to the limited resourcesof electricity and weight load [22,69]. It minimized thenumber of circuits required, the number of componentsused, and “packed them as tightly as possible” [70].Because the core ropes in the Apollo Guidance Computer(AGC) “hardwired” binary code, engineers promoted thetechnology as “a permanent storage device” [38] and thusideal for storing navigation information. In archival interviews with Apollo engineers, the AGC is repeatedly praisedfor being extremely “robust.” Don Eyles, a programmer forthe Lunar Module, went so far as to say “that code probablystill exists, despite being left on the moon” [22]. Indeed,hobbyist engineers presently tinker with recovered surplusAGC modules bought on the scrap market with the hope ofreading the original AGC code [46].Despite this, Apollo engineers faced trepidations about thereliability of computers, in general, and fears around thepossibilities of human error because of the handmade natureof core rope, in particular. Modern computing was in itsearliest stages of development and was not seen as a dependable technology. Eldon Hall, hardware designer for theApollo missions, expressed this clearly: “The biggest problem was convincing people that a computer could be reliable. That was harder than designing it” [22]. The handmadenature of many of the components amplified this persistentawareness of potential error in the manufacturing process.Despite relying on the hand-weaving of cores by the “LittleOld Ladies,” an early ACM publication introducing corerope memory assured readers, “this process has been automated to the fullest extent possible in order to minimizethe chance for human error” [38]. At each stage of assembly engineers meticulously tested the rope to ensure that ithad been wired correctly—three separate tests in total. Additionally, program managers continuously inspected thework of the weavers. Apollo line worker Mary Lou Rogersrecounts 50 years after her experience: “the componentshad to be looked at by three or four people before it wasstamped off. We had a group of inspectors in from the Federal Government to check our work all the time” [22].As David Mindell argues in his book Digital Apollo, theApollo moon missions represent one of the earliest forms ofcontemporary human-computer interaction [42]. The software (manifested materially in core rope memory) was collaboratively designed and built in a complex web of institutional arrangements amounting to hundreds of thousands ofpeople. And, perhaps more importantly, the flight of theApollo mission required collaboration between pilots andthe first automated flight systems. Along with pioneeringsoftware, the Apollo missions also required the invention ofmethods of “software verification”—reducing the chance oferrors in the ropes to 1 in 3 billion [65]. The core memoryweavers were implicated in all of these achievements. Insum, the weaving of cores offers a compelling early case of
know technical labor, but also from our concern for the roleof historical investigation in design research.MethodologyFigure 3: Rope memory (left), core memory plane (right).digital technology development in its response to a designconstraint: offering an extremely dense, light mode of storing information.Core Memory TechnologyCore memory comprises an array of tiny donut-shaped ferrite magnets (or “cores”) that each store one bit of memory.Although core rope memory stored information using aphysical distinction—passing a wire around or through amagnetized core—core planes relied on electronically controlled changes in polarity due to the direction and amountof current flowing through the ferrite cores. This meant thatrope memory comprised read-only memory (ROM) whereas core planes consisted of re-writeable memory.Both core rope memory and core memory planes shared amobility and reliability, storing information indefinitelywithout the use of electrical current — once the cores werepolarized they would remain so. However, unlike core rope,core planes comprised at least two types of wire that madethe machinery readable and writable: vertical and horizontalwrite wires (also called “X” and “Y” lines) that polarizedeach core and read wires (also called “sense” lines) that randiagonally through each core and detected voltage changes.To write to a core, the computer sent just enough electriccurrent through an X and Y wire so that only the core at theintersection changed polarity. To read that core, the machine first had to write: sending that same electrical currentthrough each core while sensing for a polarization change(signaled by a spike in voltage). The core memory weaverscritically assembled those sense lines: weaving through oraround cores in the case of core rope memory, and weavingthrough each core in the case of core memory plans (theprocess we would introduce in workshops, outlined below).OUR PROJECTWe developed the Making Core Memory project with thehope of examining the forms of technical labor performedin early processes of core memory production. We alsoaimed to use core memory production to expand HCI methods of design research: positioning embodied, personallysituated, and historically inspired encounters as valuableprocesses of investigation. We wanted to explore how ourown valuations of technical labor might inform understandings of craftwork: exploring and even challenging the prevailing purification of high status cognitive labor associatedwith male engineers from the ostensibly low-status practices of women’s hands. Our development of the CoreMemory Quilt (hereafter, the Quilt), in this sense, stemmedfrom an interest in examining how HCI scholars come toOur methodological approach draws from feminist approaches to situated inquiry [27,28,63] and strands of interventionist inquiry within traditions of critical and speculative design [17,18,47], a position summarized elsewhere ascritical fabulations [55] (see also [29]). Where feminist approaches to intervention foreground the generative andhighly situated forms of knowledge produced by frictions,complications, and breakdowns, programs of speculativedesign cast designed objects as tools for interrogating alternative futures.Our project draws from the above perspectives to explorethe specific bodies, practices, and narratives underacknowledged in historical accounts of innovation and contemporary understandings of engineering work. Specifically, we harness our design process to explore two centralquestions. First, how do craft legacies of innovation informHCI’s ideas of technical labor? Second, how might historically-informed objects expand existing instruments of design research?Design processOur process of developing the Core Memory Quilt comprised two phases: 1) making the Quilt, and 2) organizingworkshops for people to engage with the Quilt. We beganthe first phase by reviewing existing historical resources oncore memory production (a selection of which we describein the prior background section). We then ordered corememory planes made in the 1960s from e-Bay and noticedthe samples held a surprising resemblance to quilt blocks,the square pieces
Making Core Memory: Design Inquiry into Gendered Legacies of Engineering and Craftwork Daniela K. Rosner1, Samantha Shorey2, Brock Craft1, Helen Remick3 1Dept. of H
Guided inquiry in the Chemistry lAborAtory exPerienCe 15 The degree of inquiry is based on the amount of self-direction by the student compared to directions provided by the teacher. This concept is summed up in the following table: essential Features of Classroom inquiry and Their Variations Essential Feature Open Inquiry Guided Inquiry Structured
Module 1: Introduction to Scientific Inquiry Lesson 1: Scientific Inquiry Activity 4: Misconceptions and Challenges of Scientific Inquiry Estimated Time: 10 minutes What aspect of scientific inquiry are you already doing or what appeals to you about scientific inquiry? Record your ideas below. Modu
John Dewey's Dual Theory of Inquiry and Its Value for the Creation of an Alternative Curriculum Fred Harris, independent scholar Abstract Dewey's theory of inquiry cannot be reduced to the pattern of inquiry common to both common-sense inquiry and scientific inquiry, which is grounded in the human life process, since such a reduction ignores
In memory of Paul Laliberte In memory of Raymond Proulx In memory of Robert G. Jones In memory of Jim Walsh In memory of Jay Kronan In memory of Beth Ann Findlen In memory of Richard L. Small, Jr. In memory of Amalia Phillips In honor of Volunteers (9) In honor of Andrew Dowgiert In memory of
Memory Management Ideally programmers want memory that is o large o fast o non volatile o and cheap Memory hierarchy o small amount of fast, expensive memory -cache o some medium-speed, medium price main memory o gigabytes of slow, cheap disk storage Memory management tasks o Allocate and de-allocate memory for processes o Keep track of used memory and by whom
Chapter 2 Memory Hierarchy Design 2 Introduction Goal: unlimited amount of memory with low latency Fast memory technology is more expensive per bit than slower memory –Use principle of locality (spatial and temporal) Solution: organize memory system into a hierarchy –Entire addressable memory space available in largest, slowest memory –Incrementally smaller and faster memories, each .
Memory -- Chapter 6 2 virtual memory, memory segmentation, paging and address translation. Introduction Memory lies at the heart of the stored-program computer (Von Neumann model) . In previous chapters, we studied the ways in which memory is accessed by various ISAs. In this chapter, we focus on memory organization or memory hierarchy systems.
Solutions: AMC Prep for ACHS: Counting and Probability ACHS Math Competition Team 5 Jan 2009. Problem 1 What is the probability that a randomly drawn positive factor of 60 is less than 7? Problem 1 What is the probability that a randomly drawn positive factor of 60 is less than 7? The factors of 60 are 1,2,3,4,5,6,10,12,15,20,30, and 60. Six of the twelve factors are less than 7, so the .
