A Bird in the Hand: Index Cards and the Handcraft of Creative Thinking
Haig Armen and I presented this paper (actually, we wrote a lot of index cards and stuck them on the wall in front of a projector!) at Congress 2013 in Victoria this past June. The talk was part of a session called Mediating Creative Practice that was put together by Frederik Lesage and Ben Woo. I don’t know why it took so long for me to put it online, but here it is.
* * *
A Bird in the Hand:
Index Cards and the Handcraft of Creative Thinking
John W. Maxwell
Publishing @ SFU
Simon Fraser University
Faculty of Design & Dynamic Media
Emily Carr University of Art and Design
The humble index card (and card index) has a rich history. It is the precursor to the electronic database. Its role in many a writer’s practices has been celebrated. Cards and card sorting are popular too in design research and agile software methodologies. Cards are protean artifacts in that they are indexical, iconic, and textual; different card practices privilege these aspects differently. This paper looks into personal and ephemeral uses of cards in creative practice. We explore cards as “personal dynamic media” in both individual and collaborative settings, and question the extent to which these practices can be modeled in software.
This paper is about paper–and its role in the creative process. It is easy to think of paper in superficial terms: it is the traditional home of the written word and it provides a convenient and wonderfully portable storage medium for text. And yet paper has become so cheap and ubiquitous in contemporary life that it is easy to take it for granted. Indeed the advent of the ebook in recent years has cast the role of paper in particular relief, and we are ironically reminded of its physical qualities, its relative permanence, and its centrality to literary practices. The book’s traditional manifestation on paper at times seems almost poised for a renaissance, at least in bookish circles (Bosman 2011). And our stubborn tendency to print out electronic drafts for editing, articles for leisurely reading, and even correspondence for filing speaks volumes on our inability to leave it behind (See Sellen & Harper 2001). But, easy as it is to imagine so many sheets, sheafs, and codices that we often overlook the subtler ways we use paper beyond the obvious document forms, and underestimate the complexity of our relationship to paper. As Ian Sansom points out in his remarkable book, Paper: an Elegy, “We are … paper fanatics and paper fundamentalists: even when it’s not there, when it has been shown to be unnceccesary or not to exist, we continue to imagine it, to honour it, and to wish it into being.” (Sansom 2012, xvi)
One of our favourite paper forms is the index card. We (the authors) have been interested in the culture of index cards for some time. Index cards–ubiqitous and wonderfully low-tech–have a long and complex history that extends from the beginnings of organized print culture right into the midst of digital media. The extent and details of cards’ role in the mediation of creative practice is woefully under-represented; like the blind men and the elephant, cards and card practices look very different depending on whom you talk to. Writers tell a certain set of stories about how they use cards (e.g., Nabokov 1967; Benjamin 1928); researchers tell another (e.g., Niklas Luhmann; see MK 2007a); designers and software developers still more. To a certain extent, we have found a number of recognizable traditions of practices with cards, but it is also true that individual card practices can be very idiosyncratic. UBC English prof Janet Giltrow even suggests that note-taking and such “supplementary practices around research and writing” are, by and large, not taught; they are not “craft practices” that are deliberately passed on across generation of scholars (Giltrow 2012). And yet card practices persist; every university bookstore is well stocked with decks of lined and unlined cards, not to mention blocks of similarly sized sticky notes. Cards and card practices turn up again and again when we look at creative practices. So what are we doing with them?
Our study—into cards and card practices on the threshold of the digital revolution—is only in its embryonic stages. We have yet to present quantities of empirical data on card use, we have yet to create new prototypes for card use in the digital age; though we intend to do both. But even a general survey of card use in recent history, which this paper does present, reveals interesting patterns in how people move thoughts, ideas, and chunks of information between their minds and their hands. Much of the more practical fundings presented herein are drawn from our first-hand practices with cards, as researchers, scholars, designers, and software developers. We have approached this initial report as “reflective practitioners.”
Card play (including note taking, sorting, shuffling, etc.) is one of a variety of high-level methods creative people use to organize thoughts and generate ideas. The card practices we focus on here can be compared with a number of other organizational strategies which include the production of conceptual sketches or diagrams (including mind-maps), writing in Moleskine notebooks, compiling big (non-)linear MSWord documents (whose structure may be opaque to all but the creator), and even spreadsheets–not of numbers, but words and thoughts.1
What distinguishes cards from other kinds of documents (whether paper or software) is that cards can be and are used in at least three distinct modes. First, a card is a textual document; a writer could compose a novel on a deck of index cards (as Nabokov did) and you could read that novel by reading the cards as if they were pages in a book (or, presumably, as hypertext). Second, a card can be indexical–hence the “index card”, and even more so, the “card index” that forms the backbone of all library catalogues (even though the cards are now electronic). In the indexical mode, a card stands in for another object, providing a means of storing, sorting, and random access much more easily than manipulating the collection itself. The indexical mode of card practice is, we argue, the very prototype for the software database.
But a third modality of card practice is an iconic mode, in which the visual (not to mention handy tactile) characteristics of cards make possible arbitrary arrangement in two or more dimensions—on a table, say, or a corkboard. Here, cards are manipulables. This is the modality that makes possible most card games, from Tarot to Solitaire,2 and also a good deal of contemporary card practice in design and software development. Icons work differently than both text and index in that they are immediately recognizable; they work visually rather than linguistically (see Peirce 1955). The visual/iconic aspects of card practices allow us to move ideas around in physical space in an almost unique way. This is also, ironically, the piece that almost all attempts to mimic index cards in software miss.
Card play, then, is multifaceted. We look at cards in one or more of these modalities: we can read them textually, like a document, by attending to the words written on the card. We can read them indexically, as in a library catalogue. We treat the cards iconically, as visual stand-ins for their concepts, and we are able to push these around a surface, organizing thoughts as though they were baseball players on bubblegum cards. Not surprisingly (Bruner 1966, 10ff), different people (and different practices) put different emphases on different parts of card play; perhaps no two people do it exactly alike.
The history of card play is a long one. Evidently, its origins are in playing cards – as games (like Poker, Hearts, Solitaire, etc.) and as cartomancy (like the Tarot). Cards allow both sorting and randomizing (pick a card, any card); they often have two sides: a presentation side and an anonymous side, and thus can present as visible or hidden. Random access to cards is a key idea in card play. It makes games possible, obviously, but also allows a single card to take focus from the rest of the deck. There are myriad variations on this theme. An example is the card deck created by famed producer Brian Eno and artist Peter Schmidt. The cards featured printed aphorisms, and were used as a means of breaking creative blockages. Eno apparently played these cartomantic games in the recording studio with David Bowie in the 1970s, encouraging lateral thinking. Bowie, for his part, apparently wrote lyrics by the “cut-up” technique of arranging snippets of text on cards—a technique initially popularized by William Burroughs.3
Markus Krajewski’s 2011 book, Paper Machines traces the history of cards and card catalogues from the 16th century to the 20th. Krajewski credits Swiss physician Konrad Gessner with the earliest published method for card play in organizing information; Gessner’s book Bibliotheca Universalis published in 1548, presents a system of copying out excerpts and ideas, one per line, cutting them into individual slips, and organizing them in small boxes. (Krajewski 2011, 12–13). Krajewski pays especial attention to the kind of boxes Gessner used, and compares them to the wooden type-cases developed by Gutenberg and other early printers. The suggestion is that the card index is part of the re-organization of knowledge and expression that print brought about (and the “segmentation” that Marshall McLuhan noted). Later, the development of a national bibliography and unified catalogue of French books in the swirl of the French revolution in the early 1790s led to the standardization of the cards themselves, as uniformly sized and formatted cards rather than mere slips of paper—drawing directly from the example of playing cards (p45ff).
Krajewski goes on to trace two subsequent historical threads. The first is the development of a somewhat standardized set of practices around what he calls “the scholar’s box,” a means of organizing and filing notes, references, and excerpts by an individual scholar. Within a collection of cards emerges a set of cross-references, and Krajewski points out that the value of an individual card is nil, but an inter-connected collection becomes something like a book, if not hypermedia itself (64ff). Through the 17th century the practice of keeping massive card indexes flourished. And yet, this practice was oddly curtailed by other cultural movements. Krajewski writes:
In their time, men like [18th-century lawyer Johann Jacob] Moser could proudly refer to their index cards as a text-generating technology, contributing to the Enlightenment with an almost uncanny production rate. yet around 1800, with the blossoming or the European idea of genius, the light dims, and production aesthetics undergo a fundamentalk change. From now on, painstakingly produced drafts go unmentioned, veiling the writing process in the darkness of a productive sleep. Darkness keenly protects the trade secret of textual genius. (62)
Krajewski’s other interest is the development of card-based filing systems in library catalogues, a practice with its roots in the French Revolution, but which comes of age in Melvil Dewey’s masterful standardization of the American Library Association, (87ff) which gave rise to the card catalogue boxes that still grace many libraries today (or which are nostalgically missed by those of us old enough to remember them).
But Krajewski’s focus is on systems of organization, whether individual or institutional; the role of the card here is primarily indexical, and his vision of cards is in stacks or carefully constructed filing boxes. Krajewski’s purview does not include spreading cards out on a table and grouping and sorting them visually, though it would appear that card-sorting practices have roots in the same playing card traditions that gave rise to the index card itself.
“Card sorting” as a non-recreational practice is traceable, according to the Interaction Design Foundation, to the use of playing cards in experimental psychology. Card sorting was a means of measuring memory and cognitive functions (Hudson 2012). Later, card sorting because used in various methodologies for qualitative data analysis. Interaction & usability pioneer Jakob Nielsen wrote about using card sorting in the design of Sun Microsystems’ website in 1994 (Nielsen 1994), possibly the first published account of the technique being used in service of a design methodology.
Today, card-sorting practices of a wide variety are part of a good deal of contemporary design methodology: as a research technique, as a strategy for increasing end-user engagement, and as a way of conducting the design process collaboratively. Card sorting is widely used by information architects and designers to gather and understand structure for a variety of purposes. A typical use of cards might be to map the information for a website onto cards, and the sorting—a process in which many different people may participate—helps create categories for navigation and the overall architecture. Interaction designer Dave Gray and colleagues distinguish between card-sorting and other common design research tasks with cards or post-it notes:
The applications of card sorting are numerous, and in use it works similarly to Post-Up and affinity mapping. Card sorting can differ from these methods, however. First, the cards are generally prepared in advance, although participants should be allowed to create their own while sorting. Second, the cards are a semi-permanent artifact and can be used as a control over several exercises with different participants to find patterns among them. (Gray, Brown, & Macanufo 2010)
Contemporary software design methodologies similarly employ card play. In particular, the Agile method features cards and card sorting as a collaborative design technique, in which the various parts of a overall system—whether human, software, or infrastructure—can be arranged and manipulated by a team of designers. One such practice, the “CRC Card” modeling technique (Beck & Cunningham 1989) specifies a set of formats and mechanisms by which a team can work out the complex interrelationships of software modules; a form of structured brainstorming by a group, around a large workspace. Interestingly, the CRC Card technique was originally implemented in software, using Apple Computer’s HyperCard. However, the authors note,
we were surprised at the value of physically moving the cards around. When learners pick up an object they seem to more readily identify with it, and are prepared to deal with the remainder of the design from its perspective. It is the value of this physical interaction that has led us to resist a computerization of the cards. (Beck & Cunningham 1989)
How then do we work with cards—physically moving them around, writing on them, having them stand in as ideas in play? The ways in which cards can be manipulated in space are myriad. Organizational schemes include linear (or sequential); grouped (like affinity diagrams); structured as grid, tree or radial, stacked; fanned (like a hand of playing cards, as a horizontally splayed a stack); and, more generally, laid out arbitrarily across a table.
Related to these kinds of organization are ways of working within a card: colour coding (paper colour, and colour tagging) and a variety of structured formats for how and where to write on a card. Card play often involves the general practice of “jotting”—that is, fast, impermanent, and small/constrained by the size of the card. Only a few words will fit, especially if they are to be readable from any distance; many practitioners recommend the use of a heavy marker (e.g., a Sharpie).
Two-dimensional cardplay provides a macro view, however they may be organized. The macro view persists even when examining or focusing on one card—and herein may be the unique quality of index cards (often 8x13cm or thereabouts) over other common paper forms. In their small size, they share available space on a surface more easily than larger documents. They are easily handled at this size, and so one’s attention can be shifted from one individual card to the larger arrangement and then to another individual card without requiring re-arrangement. This practice seems harder to achieve with letter-sized paper or books. We may indeed arrange multiple books or documents in order to peruse them in concert and shift our attention from one to another, but not with the same facility as small cards.
The affordances of the workspace are critical. Visual-spatial arrangement requires a relatively large surface, be it a desk or table (for index cards) or a wall (if sticky notes are used in place of cards). A stack of cards lends itself to easy manipulation in the hands, or in one’s lap. Interestingly, software incarnations of card systems tend to assume the stack as the primary organizing form, as computer screens are rarely as big as a desk or table, and therefor limit the macro view of a card arrangement (unless the cards are shrunk in size). Screens afford an excellent micro view of an individual card, and of course computers allow all manner of sorting operations, but screens—at least at contemporary sizes—do not deliver the macro view that seems essential to a great deal of card play.
In practice, the arrangement of cards is messy. Mess is not a bad thing in and of itself; it speaks to complexity. The bigger problem is noise in a complex arrangement, that which prevents one from seeing either the details or the patterns of arrangement. Interestingly, though, index cards provide us a set of ‘manual’ tools for sorting through, or for getting out of the mess, and reducing the noise. By moving cards around on a surface, making and unmaking groupings, we gain topsight on both the details and the possible arrangements. We are reminded of Ariadne’s thread, which allows one to go into the labyrinth and return.
Cards are also collected, piled, sorted into stacks. A stack of cards is an encapsulation of the cards in the stack. It is an abstract representation of the cards, for their order or logic is no longer visually accessible. There may be a significant order to the stack–it is sorted in one way or another–but that order is black-boxed, abstracted away when the cards are stacked. In contrast, the cards on the table are ordered in an immediate, visual sense. We might think of stacks and spreads of cards as ‘closed’ and ‘open’ arrangements, respectively.
So, there is a two-dimensional arrangement or cards on a table or a wall, and there is a different arrangement in a card stack. By putting these together, we—potentially, at least—get three dimensions. But regular card play rarely goes to three dimensions. We say rarely, because most of the obvious card processes are in one or other of these basic modes. But we do sometimes go to three dimensions. A good example is in the game of Solitaire, where we move from a two-dimensional arrangement to a set of stacks. By building stacks, we win the game.
In design practice (e.g., Hudson 2012), card sorting takes a generic sequence: see, sort, and distill. The modes are divergent and exploratory thinking (like brainstorming or requirements capture) and then convergent thinking (grouping, sorting, culling, distilling), and the cycle is often repeated. What follows is called “ideation,” where the thinking that is synthesized by the divergent and convergent modes produce insight (or at least provoke) new ideas (Rutter 2010).
Indeed, we “win the game” of creative or design practice when we go to the ideation stage. At this point the thinking seems to move off the two-dimensional surface and hinges toward a new plane. This is the goal, the point, the culmination. Often, this means moving the practice to another mode or medium entirely: to a diagram or sketch, to a written composition (like this one).
Index cards have long been represented in software. In fact, the stack of cards is one of the oldest metaphors in software design. One one hand, the library card index is a direct ancestor of the electronic database; punched cards replacing hand-written ones. On the other hand, the index card is also an appealing metaphor for user-interface design. In the early 1980s, a software tool called Notecards was developed at Xerox’ Palo Alto Research Centre (Halasz, Moran, & Trigg 1987). Much more famously, Apple Computer’s HyperCard software (distributed freely with Macintosh computers in the late 1980s) provided the first ‘mass-market’ exposure to hypertext and hypermedia, via a staightforward index-card metaphor. While HyperCard came well before the World-Wide Web, one of the Web’s most successful applications, Wikipedia, derives directly from it. Ward Cunningham, who wrote the first wiki software in 1995, reports that his design was taken directly from HyperCard and cardplay more generally (Cunningham 2003).4
Almost all card-inspired software mimics the individual card as an interface. A ‘stack’ of such cards can then be sorted, re-arranged, and accessed randomly. Such interfaces have repeatedly proved successful for note-taking and other information collection practices (see MK 2007b). Software-based card decks connected to the Web can then be used collaboratively by small or even large groups of people. Again, Wikipedia is the easy example to point to. Today, scores of software products feature a card-like metaphor for managing and filing short documents.
Software design has primarily succeeded in representing the individual note card in a sortable stack, however. Less successful is the multidimensional, visual sorting aspects of cardplay.5 To a great extent, this is the result of screen limitations in contemporary computer interfaces. Rarely do we have access to a computer screen much larger than a letter-sized sheet of paper. As laptops replaced destop machines, the majority of available screens became smaller rather than larger. And, as tablets and mobile devices proliferate, screens get smaller still. While there is some chance that large (tabletop-sized) touchscreens may become popular in the future, prevailing trends are towards portability.
The trend in consumer electronics may indeed result be the continued popularity of the lowly index card. For now, nothing electronic seriously challenges the combination of facilities card play presents: the immediate, tactile, visually dynamic, and collaborative manipulation of ideas in space. Cards are inexpensive, emphemeral, portable, and satisfying to use. The study of cards and card play in creative practice is a fresh new field. The present paper has served as only an introductory framing of the directions of our research, and we would invite contributions and interested collaborators in this work.
Beck, K. & W. Cunningham. 1989.“A Laboratory For Teaching Object-Oriented Thinking” In Proceedings of the OOPSLA 1989 Conference, New Orleans.
Benjamin, W. 1997. One-Way Street and Other Writings. New York: Verso.
Bosman, J. 2011 “Selling Books by their Gilded Covers.” *the New York Times, Dec 3, 2011.
Bruner, J. S. 1966. Toward a Theory of Instruction. Cambridge: Harvard University Press.
Cunningham, W. with B. Venners. 2003. “Exploring with Wiki: A Conversation with Ward Cunningham.” In Artima Developer.
Giltrow, J. 2012. Comments made at the ReadingDigital Symposium, University of British Columbia Library, Sept 20, 2012.
Gray, D., Brown, S. & J. Macanufo. 2010. Gamestorming
Halasz. F.G., Moran, T. P., & R. H. Trigg. 1987. “NoteCards in a Nulshell” In Proceedings of the 1987 Conference of Human Factors in Cam- puter Systems (CHI+GI ’87). Toronto. Ontario, Apr. 5–9, 1987, pp. 45–52.
Hudson, W. 2012. “Card Sorting.” In The Encyclopedia of Human-Computer Interaction 2md Edition, ed. by M. Soegaard and R. Friis Dam.
Markus Krajewski. 2011. Paper Machines: About Cards & Catalogs, 1548–1929. Trans. by Peter Krapp. Cambridge: MIT Press.
MK. 2007a. “Luhmann’s Zettelkasten” TakingNoteNow blog, Dec 16, 2007.
MK. 2007b. “A Faithful Electronic Version of Luhmann’s Zettelkasten.” TakingNoteNow blog, Dec 16, 2007.
Nabokov, V. 1967. “The Art of Fiction No. 40” The Paris Review. 41. Winter/Spring 1967. <http://www.theparisreview.org/interviews/4310/the-art-of-fiction-no–40-vladimir-nabokov>
Nielsen, J. 1994. “The 1994 Design of Sun Microsystems’ Intranet”. Jakob Nielsen’s Alertbox, December 31, 1994.
Peirce, C. S. 1955. “Logic as Semiotic: The Theory of Signs,” in Philosophical Writings. New York: Dover.
Sansom, I. 2012. Paper: An Elegy. New York: Fourth Estate.
Scerri, E. 2012. “The Periodic Table” OUP Blog. August 8th, 2012. <http://blog.oup.com/2012/08/how-exactly-did-mendeleev-discover-his-periodic-table-of–1869/>
Scrivener writing software. See
Sellen, A. J. & R Harper. 2001. The Myth of the Paperless Office. Cambridge: MIT Press.
David Straker. 1997. Rapid Problem-Solving with Post-it® Notes. Boston: Da Capo Press.
Trello task-management software. See
- The use of spreadsheet software to organize ideas–not just numbers–is, in our experience, widespread in today’s office culture. Microsoft Excel is a powerful data-processing tool installed on vast numbers of personal and office computers. Indeed, shy of going to the trouble of developing an actual database, Excel is the tool at hand when people need to organize things. Hence, schedules, duty rosters, content inventories, and so on are often found in the rows and columns originally designed for calculations and bookkeeping.↩
- The pioneering chemist Mendeleev, it is said, played “chemical solitaire” on the way to developing the Periodic Table of the Elements. He wrote the names and properties of known elements on cards and sorted and re-sorted them on his way to discovering the schema we know today as the Periodic Table. See Scerri 2012.↩
- The Eno/Schmidt card deck is still available. See Oblique Strategies:
- Interestingly, software designer and Agile Methodology pioneer Ward Cunningham is the common force behind both the “CRC Card” modelling technique and wiki software. Both draw inspiration directly from card play; both were originally implemented in Apple’s Hypercard. See Cunningham 2003.↩
- Some notable examples of writing and personal organization software attempt to provide a visual sorting mode. The writing software Scrivener features a “corkboard” mode in which snippets can be dragged around onscreen like so many cards. The web-based task manager Trello works on a visual sorting metaphor, moving card-based tasks from one column to another.↩