Surfaces and Essences: Analogy as the Fuel and Fire of Thinking (95 page)

Sometimes one finds much charm in definitions that were drawn up in a long-gone era when today’s technologies could not have been imagined by anyone. For example, the 1932 Funk & Wagnalls New Standard Dictionary, previously quoted in
Chapter 4
, defines the words “browse”, “folder”, and “hacker” as follows:

It’s rather astonishing to think that less than a decade before the invention of the digital computer, in an era when radios and movies, cars and airplanes, telephones and record players already existed and were even commonplace, the three concepts above were conceived of in ways that today strike us as incredibly concrete, limited, and quaint; indeed, if someone from back then could visit us today, it would take us a little while to explain to them the modern technological senses of these words. There is no doubt that the old and new meanings are cousins, and yet it would take a nontrivial crash course and some major leaps of the imagination to spell out some of the analogies that bridge the gap, but once one sees the common abstract core, it is very clear.

Consider, for instance, this possible contemporary definition of “hacker”: “A person who invades a data base to permit the illicit flow of information”. The analogy with the 1932 definition has deliberately been made salient, and yet the conceptual gap is still a huge mental stretch, somewhat reminiscent of the vast conceptual gap between California’s rural and picturesque Santa Clara Valley in the post-Depression years, dotted with scenic orchards and small farms, and what it became just a few decades later: the ultra-modern bustling metropolitan area known as Silicon Valley, jam-packed with high-tech firms, criss-crossing freeways, upscale housing developments, and more Thai, Indian, and Chinese restaurants than you can shake a stick at.

The emergence of modern computer-oriented meanings for many words not only has given rise to new categories, but has helped to make the abstract essences behind
the terms in the above list become clearer (as was discussed in
Chapter 4
). As we already saw in the case of the word “desk”, the original meaning of a term gives rise to an analogical extension, allowing the old concept to apply to virtual (or “software”) entities; this extension in turn simplifies and refines the old category, in the sense that some of its once-central qualities are now seen as superficial and thus merely optional. For instance, the virtual concept of
address
has made us understand that
physical location
, standardly tied tightly to a postal address, is not crucial to this concept, but that what matters, in fact, is
accessibility through a symbolic label.
When a message is sent to an electronic address, what matters is simply that it should reach the person concerned, and the geographical whereabouts of said person are of as little import as whether a desktop is made of wood or pixels or whether it has drawers or not. In short, the recent appearance of a digital version of the category
address
has made us see that geography is not relevant to the term, even though up till then, geography had been absolutely indispensable to the notion.

Certain curious phrases that crop up now and then on Web sites, such as “Your trip has been placed in your shopping cart”, reflect this tendency of categories’ essences to be revealed ever more clearly as categories get analogically extended. Thus an online shopping cart shares with a physical shopping cart the property of accumulating potential purchases, about which one can change one’s mind at any moment until one comes to the “checkout stand”. In real life (
i.e.
, in “the good old days” before computers became a central reality in our lives), it would seem absurd, to say the least, to speak of “placing a trip in one’s shopping cart”, but what would be absurd for the old and narrow categories of
trip
and
cart
need not be so for the new and broader categories, because once a
trip
has become a
potential purchase
, and once a
shopping cart
has become a
repository for potential purchases
, then the phrase “to place a trip in a shopping cart” seems perfectly innocuous — indeed, quite sensible.

Finding a new category that gracefully combines the pre-technological and the technological versions of a concept often requires one to jump to a relatively high level of abstraction, as is the case, for example, for
firewall, hacker, peripheral
, and
port.
Thus in the on-line world, a
firewall
is a protection against
hackers
, whereas in a more traditional context, the phrase “a firewall against hackers” has little or no meaning. Similarly, in a computer context, it makes perfect sense to speak of “plugging a peripheral into a port”, whereas in the pre-computer world, such a phrase would merely have sounded like meaningless gibberish.

And yet there are other categories, such as
address book, keyboard, move, screen, delete
, and
send
, which scarcely seem to have been extended at all to incorporate their new aspects. One might think that these categories have emerged unscathed from all the technological upheavals, that they have withstood the earthquake without being touched in the least. However, the truth of the matter is a bit subtler than that. Thus, for instance, an electronic
address book
, unlike its paper forebear, is not a concrete object. One “writes” entries in it not with a pen but by typing (or possibly by speaking aloud!), and it allows one to do certain things in a flash that would be very laborious with an old-fashioned address book, such as finding a person given their telephone number.

Asking whether a virtual address book
belongs to the category
called “address book” or simply
is analogous to
an old-style address book is a question that needn’t be answered, because the dichotomy it presumes is a false one. Belonging to a familiar category and being analogous to a familiar thing are not black-and-white matters, and should not be thought of as opposites, or as excluding each other; both have sliding scales (or shades of gray) that depend on both perceiver and context; indeed,
strength of category membership
comes down to nothing but
strength of analogousness.
The gradual and natural extension of technological terms provides an excellent illustration of the fact that analogy and categorization are just two sides of the same coin.

What about the concepts expressed by verbs such as “move”, “erase”, and “send”? It might seem that their meanings haven’t budged an inch as a result of the computer revolution. And yet, would you say that “moving” a file from one folder (on a hard disk) to another is
exactly the same thing
as moving a paper file from one cardboard folder (in a wooden drawer) to another? Or would you say that highlighting (on a screen) the set of pixels forming an alphabetic character and then hitting the “delete” key is
exactly the same thing
as quickly and forcibly rubbing the pink end of a pencil back and forth across some marks (on a piece of paper) until they are barely visible any longer? And would you say that sending an electronic message is
exactly the same thing
as sending a letter via “ordinary” mail? It’s easy to forget that when one sends a material letter, one has to relinquish it physically, whereas when one sends a message electronically, the original remains on one’s hard disk. And so we are reminded once again that even when it comes to terms that, on first blush, seem completely unextended by their computer versions, the truth is that the categories in question have indeed been broadened, and it is only thanks to analogy-making with previously familiar everyday categories that these extensions could take place.

Do the hundred-odd terms given a few pages back, all resulting from analogical extensions, coexist with all sorts of other new terms that did
not
come from analogical extensions? The fact is that whenever a new technology comes along, the standard way of devising a new set of terms that work naturally with it is to borrow pre-technological terms and to rely on the predictable naïve analogies that most people tend to make. Anyone who doubts this should just listen to computer people talking and take note of terms that didn’t exist fifty years ago. They will discover that everyday down-to-earth words are ubiquitous, while terms that are unique to the new technologies and that are nowhere to be found in old dictionaries, such as “motherboard” or “pixel”, are not all that frequent. To be sure, if one were to transcribe a technical conversation between two computer specialists, one would find a rich harvest of acronyms and other narrow technological terms, just as one would for any specialized discipline, but there can be no doubt that ordinary people, in speaking about their computers, are constantly exploiting terms that hark back to a day long before anyone could have dreamed of the abstract uses to which those terms would be put, decades or centuries hence.

Thus concepts from the world of computers now permeate our daily lives because our down-to-earth concepts, through hundreds of naïve analogies, have permeated the technology itself. These naïve analogies, building as they do on extremely familiar categories rooted in mundane daily activities, allow us to endow complex and mysterious technological entities with all sorts of simple and unmysterious properties, and they do so at minimal cognitive cost. Terms based on naïve analogies catch on easily because they naturally bring out qualities that otherwise would be highly elusive.

Thus thanks to our naïve analogies, we all speak with ease and accuracy of “placing objects on the desktop”, of “inserting documents into a folder”, of “opening or closing a window”, of “moving, copying, or throwing away a file”, and so forth — and we didn’t need a course or an instruction manual to gain this skill. Moreover, this down-to-earth lexicon for describing the behavior of abstract technological entities (such as
desks, files, windows
, and
documents)
is not just a rough-and-ready set of linguistic tools, serving solely to help novices get their feet wet but then to be summarily dropped; to the contrary, even the most technically savvy people speak in this concrete manner. They, too, open and close windows, files, and folders — and when using such phrases, they feel they are expressing themselves perfectly straightforwardly and non-metaphorically.

Experts in the field of human–machine interface design have stated that the best possible interface should be invisible — indeed, that “the best interface is no interface at all”. Such assertions, stressing the importance of natural and intuitive interfaces, mean essentially that designers should always try to exploit analogies to familiar things. Only if this is done will the interface become “transparent”, which means that users will feel as if they are manipulating everyday objects, a feeling that frees them up to concentrate on their main goals. Interfaces designed in this felicitous manner convert the computer into an easy-to-use tool for accomplishing a particular type of task. Instead of working up a sweat figuring out how to use the tool, one simply concentrates on the task itself.

Interfaces carefully based on analogies to familiar activities do not suffer from the great awkwardness of poorly-designed interfaces. Donald Norman, who is not only a distinguished cognitive psychologist and error-collector, -classifier, and -modeler, but also a pioneer in human–machine interface design, has stated this idea succinctly: “The real problem with the interface is that it is an interface. Interfaces get in the way. I don’t want to focus my energies on an interface. I want to focus on the job… I don’t want to think of myself as using a computer, I want to think of myself as doing my job.”

Although indispensable, naïve analogies that help us to relate to new technologies have their limitations, because members of the new category will sometimes behave differently from those of the older, more familiar category. In such cases, the naïve analogy is likely to lead one down a garden path. After all, when one depends on a naïve analogy, one does so, by definition, naïvely — that is, lock, stock, and barrel. For better or for worse, the naïve analogy is one’s only guide — and on occasion it will mislead. In a word, we are back again in the land of categorical blinders.

When the differences between a virtual category and its old-fashioned analogue do not involve the categories’ most central aspects, there generally is no problem. For example, it’s obvious that virtual desktops, files, and folders have no volume or mass, cannot get dirty, and aren’t subject to wear and tear. This is because categories such as
virtual file
and
virtual folder
are immaterial, just as is the category
virtual desk
(or
desk
₂
).