Chapter Seven Excerpt

Pictures


It is said that a picture is worth a thousand words. I have often pondered that ancient bit of wisdom. The fact that this Chinese proverb has survived for centuries must be considered testimony to its innate truth. Probably all of our ancestors who said that a picture was worth a hundred words or a million words were forgotten. Somehow a thousand is a number that hits a resonant chord with us. We say to ourselves that a picture is worth a thousand words, and we nod our imaginary heads in agreement. Yet I wonder. Throughout this chapter I shall be preoccupied with the question. Is it? Is a picture really worth a thousand words? In order to manage information transfer, we should know.

Before we go any further, let me set the stage at the outset by saying that, whether or not a picture is worth a thousand words, it seems that in electronic transmission and storage pictures cost the equivalent of a thousand words. For example, we discussed earlier the fact that voice is transmitted in the telephone network digitally at a data rate of 64 kbps. When network-quality television is transmitted in digital format on telephone facilities, a rate of 45 million bits per second is required--almost exactly a factor of a thousand more than voice. Here we are all awash with bits necessary to create the picture all for some inane sitcom or such entertainment. Why so many bits into the picture, and so few into our minds?

Since pictures are so greedy in their information requirements, it seems reasonable to ask if they give equal value in return. Confronting one of my friends who has devoted his professional career to picture processing I asked, “What is so wonderful about pictures?” It was the wrong thing to ask. “Pictures are a great way to get a lot of information into a human mind very quickly,” he replied defensively. (I left out some adjectives in his response.) I conceded the point. Personally, I often feel that I can gorge myself with bits in only a single glance at a picture. That is what minds are designed for, I think--parallel processing of images; pictures are clearly the way to leap tall buildings of information in only a single bound. It is, after all, a common feeling. Turgenev wrote that “A picture shows me at a glance what it takes dozens of pages of a book to expound.” A poem by Alexander Pope contains the phrase, “Or where the pictures for the page atone...” But does this intuitive feeling really stand examination? That will be one of our areas of concern in this chapter.

Briefly, there are two issues: How many bits does it take to create a picture, and how much information can a human extract from it? For the first of these, we will discuss some of the approaches that are used in computer technology for picture coding. We will also be concerned briefly with the abilities of computers to create, recognize, and understand pictures. Though these capabilities are rather primitive, the knowledge in those fields seems more substantial than the knowledge of our own, human, picture processing abilities. Is a picture worth a thousand words? There is no accepted answer. Read on, and decide for yourself.

The Prevalence of Pictures

At the moment I am writing at the terminal in my office. Surrounding me are the stacks of papers, books, and memoranda accumulated from a quarter of a century of technology--the information that has stuck to me, for whatever reason. On my desk there is a giant pile of to-be-processed mail. Behind my terminal a small blue cable snakes through a little mouse hole in the wall out to its freedom in an electronic world of endless computer files to which I have access. Everywhere I look and touch, information. How many pictures do you think there are in this melange? Right. I am sure that you thought very few, especially in proportion to the text. There is one glaring exception. Built into the wall behind my desk are a television set and videotape recorder. I never use them. Yet the small rack of video tapes contains more bits than all the books that line my shelves. If pictures are such an efficient way to convey information, why are they not more prevalent in this office where the sole preoccupation is the exchange and processing of information?

Beyond this cloistered domain of research, the world of the information age buzzes with pictures clamoring for our attention in a competitive struggle. “Look at me!” they plead. They can only hope for a glance, so they have to be effective. It would be hopeless to try to sell anything without using pictures (at least anything ordinary--multimillion dollar programs become the exception, since for that kind of venture people stop looking at pictures and start searching for logic). Magazines, newspapers, advertisements, posters, products in beguiling packages--we swim through waves of color. At home, television mesmerizes us and fills our time with the parade of passing images. We sink into bed, bloated with pictorial bits, starved for information.

When I think of the most prototypical sources of “information,” I think of libraries and encyclopedias. How many pictures are there in those sources? Proportionately, there are very few, but the trend is towards more and more. In the famous 11th version of the Encyclopedia Britannica, published in 1911 (said to be the last attempt to encapsulate all of human knowledge) there is about one picture per four pages of text. The pictures are very small, and about 80 percent of the pictures are line drawings, representing maps, instruments and machines, and mathematical and scientific constructs. The remaining pictures are a few scattered sets of photographic plates. The text font is extraordinarily small, so the proportional amount of actual text is overwhelming. By contrast, the 1985 edition of the same encyclopedia contains about two pictures per page. Almost all the pictures are photographic and in color, and they are mostly pictures of people and places.

It is interesting again to note that if the Encyclopedia Britannica were digitized, most of the bits would be devoted to pictures. However, consider for a moment, that without those pictures there would be almost no loss of information value. On the other hand, without the text, which consumes the minority of the bits, the encyclopedia would be useless. But it is also fair to say that without the pictures the encyclopedia would be a much duller set of books. They would have a hard time selling it. Moreover, we would be much less satisfied with its content. When we read about Beethoven, for example, we want to know what he looked like, for some perverse human reason that a computer would never understand. Whether or not we have seen the picture, we would achieve the same score on a factual test of our knowledge of Beethoven. The information is mainly in the text, but much of the humanity and the attention-keeping components are in the pictures.

We have all “read” picture magazines in doctors’ offices. I can “read” National Geographic in about ten minutes. In so doing, I fly off into bit heaven, or have I photographed all those exquisite pictures and placed them in a gallery in my mind, where later I can go and browse in my imagination? Presently we will consider what is known of this matter, for it is important from an informational standpoint. For the moment my concern is more with the pictorial milieu. More and more we replace sources of textual information with pictorial sources. Picture magazines and “coffee table” books proliferate. Comic books, in some countries, are more popular than textbooks. I even find myself leafing ahead in books to find how many pictures are forthcoming, so as to judge how easy or hard it will be to get through a given number of pages. Quick and easy; pictures are the fast food, the junk food, of the information age.