The Newest Medium

e’ve been seeing computer art in picture books for about ten years now — an eon in new technology terms — but until recently it was easily recognizable and something of an oddity. Often the art had been manipulated to death, all vitality smoothed away. The illustrations usually lacked focus and the color was apt to look artificial, even a bit hallucinatory. One was reminded of the early days of synthesizers when some bands released tunes with all-synth back up: the sound was intriguing for the first few bars, but soon one longed for some texture or a little mistake — anything to show there was a human behind that sound. In the same way, the first picture books illustrated with computers were obvious novelties but certainly didn’t stand up to their paint-on-paper counterparts. It’s no wonder that the words “computer art” in a review usually turn out to be derogatory. But we are finally seeing some progress (and, surprisingly, it’s not necessarily coming from the youngest illustrators).

I remember seeing an early computer-illustrated book around 1990. We all gazed in awe at the odd colors and flat shapes outlined in black with that now-recognizable stair-step look of visible pixels. When I showed the book to a friend who had designed some of the early computer imaging systems, he predicted miraculous advances in computer art and assured me that in the future I’d be painting with my mouse just the way I did with a brush. There would be different “textures” available simulating canvas and watercolor paper. I would choose among several virtual brushes and wouldn’t have to wait for the paint to dry. And best of all, my problems with ventilation when using oils in my small apartment would be a thing of the past. It all sounded vaguely sinister and sacrilegious.

Well, the future is here, and naturally it’s not quite what my friend predicted, nor what I expected. Rather than drawing with a mouse, I sit with a large plastic tablet in my lap, “drawing” on its smooth beige surface with a cordless plastic stylus the size and heft of a ball-point pen. As I draw, the marks appear not on the tablet but on the monitor. It’s like learning to type after writing longhand: at first you think you’ll never get used to resting your hands on keys instead of holding a pencil. Will it ever feel natural? And if it does, will you be able write creatively, so disconnected from the pencil and pad? For some, the answer is no, and the pencil or pen remains the writing medium of choice. But most of us have made that transition painlessly. Getting used to a stylus and tablet was even easier. Surprisingly, the eye-mind-hand connection is what’s changed the least. After a few minutes, the fact that my hand was drawing on a tablet while my eyes were watching a screen became a non-issue. As with paints or pencils, the subconscious soon takes over, allowing one to work in a sort of trance. I suppose it’s the same for people who play sports. It would be tedious and distracting to have to think constantly about how to run or hit or catch. Instead, one’s muscle memory kicks in and most of the thinking is subliminal, allowing the athlete or artist to think about the product — the game, the painting — enjoying the process without dwelling on it.

So ten years (and three computers) later, I have not 16 grays, not 256 colors, but millions of colors. Instead of one megabyte of memory, I have 256. Nine gigabytes of space on my hard drive has replaced my original 20 MBs — that’s 450 times more.

With a software program called Painter, I can use my computer to draw or paint in virtually (as well as “virtually”) any medium I want. I can choose canvas (millions of widths of warp and woof) and watercolor paper (hot press, cold press, or create-your-own texture) without traveling to the art supply store and handing out $20 or $30 for the good stuff. Of course, the computer, monitor, software, drawing tablet, and color printer cost $7000. I get a surprisingly good approximation of oils, watercolors, pen, pencil, markers, and pastels. But drawing with the tablet, I have less control than I do holding a pencil in my hand and drawing on paper. Ventilation’s not a problem because nothing smells (how I miss the linseed oil!), but my computer’s so sensitive to heat and humidity that I had to protect my investment by buying a large air conditioner. Even so, now I work without the fear of making a mistake on expensive paper. I can copy and paste, save, save as, and even undo. But by far the most exciting development is the color palette. I create whatever color I want instantly by mixing digitally — or I can just point to a color on a wheel and, voila, there it is on my virtual brush. What I see in front of me is as vividly colored as anything by Van Gogh. But just try to print it so it looks like what’s on your monitor and you’ve opened Pandora’s paint box. More about that later.

So what’s it like for picture book illustrators who are making the same transition? Recently, I talked to three established illustrators, each of whom learned only lately to use a computer for the express purpose of employing the technology in her next book. These three are representative of a number of good illustrators who are now using computers to create books that hold their own beside traditionally illustrated books. Since the beginning, of course, there have been a few — including J. Otto Seibold (Olive, the Other Reindeer and the Mr. Lunch books) and David Pelletier (The Graphic Alphabet) — who skillfully used the computer without making us long for live music. (For the most part, they have used Adobe Illustrator, a program that, unlike Photoshop or Painter, doesn’t attempt to mimic brush-strokes, paints, and textures. Instead, it allows the user to create shapes using perfect curves and straight lines; fill these in with flat areas of color, gradients, or patterns; and manipulate type so that it’s curved or skewed.)

Before she used a computer to augment her collages in To Market, to Market, Janet Stevens says she “handwrote everything. . . . I didn’t know how to turn our computer on!” Since then, she has fine-tuned her technique in Cook-a-Doodle-Doo! and My Big Dog, now devoting a portion of her website to a description of her new method. Rachel Isadora became fed up with how the colors from her original paintings changed in the printed book. She decided a computer would provide more color control, so she hired a high school student to teach her Photoshop. Before that, she says, “I didn’t know anything about computers. Nothing, nothing, nothing.” Using the computer to create ABC Pop! and 1 2 3 Pop!, she developed a radically different style. When Ora Eitan received the text for Christine Loomis’s Astro Bunnies, an upcoming companion to their Cowboy Bunnies, she too decided it was time to learn to use a computer. She claims she is “not very good with instruments” and doesn’t drive, saying “what I can operate successfully is a washing machine, dryer, maybe microwave.” Still, like Janet Stevens and Rachel Isadora, she approached the computer as she would any new medium, with excitement rather than fear: “It wasn’t frightening. I was just aware of the fact that it would be very demanding.”

All three illustrators use a Macintosh with Photoshop, powerful image-editing software that allows the user either to create art from scratch, by painting electronically, or to edit and manipulate images from a scanner or digital camera. They all prefer to start with a simple drawing on paper that they then scan and manipulate on the computer. One of the most appealing aspects of Photoshop is its “layers,” which allow the artist to create a sort of digital collage. Imagine a stack of clear plastic sheets, each containing a different part of the image. Each layer can be turned on (visible) or off (invisible), moved in front of or behind other layers, or moved to a different part of the image. It can be made bigger or smaller, more or less transparent, rotated, stretched, and skewed. The color can be changed in hue or saturation, darkened, or lightened. For example, to create a shadow of an image on one layer, that layer can be copied, and the copy can be offset, skewed, and blurred. The possibilities are as endless as the artist’s imagination. The latest version of Photoshop even has multiple undos and redos stored in what it calls a history. Of course this takes a lot of RAM (memory) — as much as you’re willing to buy. All the layers and palettes also require the user to have a large, twenty-one-inch monitor. Once the file is done, the layers are “flattened” so the image can be saved in a format that is compatible with publishing software such as Quark or PageMaker. Flattening also reduces the file size to a “manageable” 80 megabytes or so (that’s about 55 times bigger than a floppy disk). Janet Stevens’s spreads are often 600 megabytes while they are in layers.

Stevens uses a Mac G3, a large Wacom tablet and stylus, an HP scanner, and a wide Epson 3000 color printer that allows her to print her creations on watercolor paper, spray them with fixative, and continue to paint into them with real paint. She loves using the computer for collage:

With books you’ve got the gutter, type — everything’s in your way. Now you can play with that. It’s so much easier to control because you can move your characters around and compose within that block which is your spread. Before it was just “Gosh, I hope all this fits!” And “Oops, the butt’s off the page! ” But now you can take elements and make them larger, push them into the background, put layers on top of each other, try him in the back, him in the front. That is so liberating!

On the other hand, using a computer can create a new set of problems. One is knowing when to stop. Stevens says it has taken her about two years to fully understand her tools and develop a way to work productively. It’s easy to become addicted to all the possibilities the computer affords at each step. “More isn’t necessarily better. You have to set up a whole new set of instincts. Before, when you had a piece of art that you were working on, you tried to make yourself stop before it was overworked.”

When Stevens began working on Cook-a-Doodle-Doo! in which Rooster, Iguana, Turtle, and Pig make strawberry shortcake, she knew she wanted to scan her original drawings and continue working on them on the computer. Then she thought she would scan chicken wire for the opening spread “because the thought of painting the chicken wire did not excite me.” She discovered she could remove the lid on her scanner and place larger objects on the glass, and providentially noticed that her cheese grater resembled an iguana’s skin:

I put the cheese grater [on the scanner] along with some Doritos and sponges and some other textures that were interesting to me. And then I manipulated them and made the texture of Iguana the cheese grater and sponge together. I actually used the rubber stamp tool [which allows one to copy one part of an image onto another] like a paintbrush, and painted cheese grater onto him. And you can choose light cheese grater and dark cheese grater, big cheese grater and little cheese grater. I used raspberries for his comb and Doritos for the beak of the rooster.

Several layers in a drawing of Turtle from Janet Stevens's Cook-a-Doddle-Doo! Number 7 shows the finished illustration, including hand painting over the computer images. A larger interactive version of this transition can be found on Stevens's website (www.janetstevens.com/graphics/Turtle_15/Turtle_15_layers.html)

The final and most crucial step for Stevens is printing out her digital art and painting back into it. But even printing with a high-quality printer will not deliver the same vibrant colors that were on the monitor. While it is important to calibrate the monitor and printer, the color in the printed image will never quite match the monitor’s image. In part, this is because the monitor’s luminous image shows colors by overlapping tiny red, green, and blue lights while the printer’s reflective image is made by mixing inks. But the problem goes beyond this to the question of how each brand of printer translates the digital information it receives from the computer. Stevens gets around this problem by printing out a page of brushstrokes in different colors to see how the monitor’s colors will change when printed. “I’m not paying attention to what it looks like on the monitor. The composition and rough values, yes. But then I always have the option of painting back into it.” Since the non-fading ink cartridges she prefers are water soluble, Stevens has experimented with different fixatives, trying polyurethane, gel medium, and an acrylic polymer called GAC. Choosing the right paper is crucial. The flecked beige handmade paper she used for Cook-a-Doodle-Doo! held the ink well and looked great with the original art, but it didn’t reproduce well when the book was printed.

Professional print reproduction has been an annoyance for illustrators ever since process printing became widely used in the late nineteenth century. Printers now use four standard inks, known as CMYK: cyan (a bright sky blue), magenta (reddish pink), yellow, and black. The problem is that these inks can’t reproduce every color. They do pretty well with mid-range, mid-value colors, but bright greens and oranges are impossible, as are most pastels. A good press operator can improve the color, but adjusting on press can be expensive. Many children’s book publishers take their books to printers in Italy or Singapore because printers there allow for extra time to adjust color on press while keeping the book within budget.

Back in the late nineteenth century, Edmund Evans, the pre-eminent color wood engraver and printer of children’s books, used as many as eight separate color blocks to print illustrations by Kate Greenaway, Randolph Caldecott, and Walter Crane. In Greenaway’s books, he sometimes used two separate greens if combining the blue and yellow blocks did not match the brightness of the original art. But wood engraving was time-consuming and costly, only possible with child labor and underpaid engravers who ran the risk of becoming blind after many years of the arduous, eye-straining work. So photographic process printing gradually took over and is now the standard for commercial book printing. We use four inks because it’s cost-effective and because it’s what printers and press operators are used to. While some printers now use hexachrome, a system that adds two more inks (green and orange) to the original CMYK, this process is too expensive for most trade book publishers. Art directors and printers alike are skeptical that hexachrome could ever become a picture-book printing standard.

Rachel Isadora turned to the computer because of her frustration with the printing process.

I was getting so sick of the fact that — and everybody I know says the same thing — when you see your book printed it never looks like the original artwork. So disappointing. I heard people saying that when you work with a computer, whatever values you punch in, that’s what you get. You get what you give, you know? And so I said, “Ah, that sounds wonderful! I have to explore this.”

While Stevens used the computer to help her create more interesting collages in the same style as her previous books, Isadora’s computer art has a completely new look (see this issue’s cover). She insists that the computer did not dictate her new style, it just helped facilitate it. Her Pop books and Listen to the City were created in a Pop Art style, but she says the books she’s working on now employ yet a different style — still a departure from her earlier books in watercolor, pen, or pencil. “It’s not that I’m using the computer any differently. It’s just that my style is changing,” she explains. “I don’t want the computer to be an overwhelming factor.”

Isadora uses a Mac, like Stevens, but is content with a mouse rather than a stylus for input. She doesn’t need to print her art because she submits it to the publisher on disk. Instead, she chooses colors much as color separators did when books were printed in just two or three colors. Back then, the illustrator created separate black-and-white drawings for each color and couldn’t see how they would look until the book was printed. Of course, with a little practice and some printed samples showing how the colors mixed, an illustrator became adept at imagining the combined colors as he drew. It’s similar for Isadora:

I don’t go by what my computer [monitor] shows. I bought one of those little books that have all the colors and then I’ve also gotten to know the bounds. And this is great: I can put a red in and there’s red! . . . The publisher was really skeptical, but when they printed the first thing, I said, “Wow, that’s great! Every color I put in is right there!” There are no secrets, no magic. It was terrific.

She’s currently working on a book about jazz entitled Bring On That Beat. The base illustrations are black-and-white oils while abstract computer-created colored shapes represent the music. Long attracted to a signature style of Disney’s animated films, Isadora wanted to show a similar contrast between lushly painted backgrounds and more simply drawn animated shapes. But her first art files for this book — hours and hours of work — had to be discarded because the scans of the oil paintings didn’t retain enough detail. For her previous books, she had started by scanning bold black-and-white drawings, but scanning detailed gray-scale objects requires a different set of skills and a better scanner. To remedy this, the publisher sent the oils out to be scanned professionally, giving Isadora the new scan files so she could once again add her music shapes.

This sort of trial and error is common for the pioneers of computer art. Kathy Dawson, senior editor at Putnam and Isadora’s editor for Bring On That Beat, says, “I really feel as if we’re making all the mistakes for the next group of people to come in and do it right.” Dawson and many of her colleagues feel that the vocabulary of computers is often counter-intuitive and therefore difficult to master. “Each of us has developed her own kind of language for how we describe things. We often think we’re saying the same thing and we aren’t. There’s a lot of miscommunication that goes on because we’re not trained.”

Another new frontier for publishers is the contractual situation that arises when artists submit their final art on disk. Traditionally, the second part of an advance is paid on receipt of the final artwork. But, as Dawson asks,

What is final artwork? Is it a disk that’s been scanned at the wrong resolution? Is it a disk that looks pretty good but the lines are fuzzy? People often say, “This just needs to be cleaned up.” They say that all the time with this computer stuff. Well, the final delivered art should be clean. We shouldn’t have to clean it up. . . . But then, is the disk final art? Everyone’s screen will make it look different. Everyone’s printer will make it look different.

Traditionally, the publisher agrees to create a book that matches the original art as closely as possible. They send the original art out to be photographed or scanned and have what is called an Iris print created. The Iris should be calibrated to the offset printer that will eventually print the entire book, showing the actual color possible on that particular press. If the first Iris isn’t good enough, the scan can be adjusted in Photoshop and output again. When the print is acceptable, the art director signs the Iris. This is known as a contract proof because the printer agrees to print the book to match the quality in the signed print within a reasonable tolerance. So should computer illustrators pay to have Iris prints made? Typically, an Iris for a medium-sized picture book spread would cost about fifty dollars. Multiply that by sixteen spreads and you’re spending eight hundred dollars — more if you’re not happy with the first Iris of each spread. Some illustrators could find themselves spending half of their advance on Iris prints.

But when artists submit computer files without also submitting a print they like, it leaves the art director in the dark at printing time. Cecilia Yung, Putnam’s art director, has often found herself in such situations. If she hasn’t seen computer art in a print — or on the artist’s home monitor — how can she tell if the Iris has the right colors? When the artist looks at a printer’s proof and says, “Well, that’s not what I did,” all she can do is relay the artist’s comments to the printer, saying “it needs to be brighter” or “it needs to be cleaner.”

Luckily, there is a technology on the horizon that should clear up some of these problems. Color profiling of monitors, scanners, and printers is already here, but it is rather complex and not as widely used as the industry had hoped it would be. But apparently profiling is about to become more user-friendly, especially regarding home color inkjet printers. Soon, an illustrator with a reasonably good color printer — one that costs under fifteen hundred dollars — will be able to print Iris-quality proofs at home. This will be possible through color profiling. The illustrator will actually be able to calibrate her home printer by using the color profile of the large offset printer in Italy or Singapore or wherever her book will eventually be printed. So if this technology develops as promised, the Iris problem may become passé. The illustrator can submit art on disk accompanied by a proof made at home.

This technique should be especially useful to Ora Eitan, who lives and works in Israel and submitted her art for Astro Bunnies to Putnam on disk. First she had some sample Irises made at a service bureau near her home and calibrated her monitor — as much as possible — to match their output. Eventually, some of her Irises came from Israel and some were created in New York, and she was not happy with the quality of all of them.

Astro Bunnies was something of a marathon for Eitan. She says that when she began, she knew nothing about computers. Like Isadora, Eitan hired a Photoshop expert to sit with her and do the scanning and some of the manipulating. In this way, she learned the parts of the program she needed for that book. Why did she choose to use a computer for this book when she had no previous experience? She said the subject seemed to dictate it. Like Stevenson, though, she did not want her signature style to change. So she drew or painted her characters on paper and then scanned them. Soon, she became enthralled with the possibilities: “I was scanning many other things. I was using a lot of textures and I was having lots of fun creating them. There were many things I could scan and then manipulate with color and layers.”

Eitan now feels more at home with Photoshop and has used it to create another book [Dance, Sing, Remember, reviewed on p. 769] using more visible computer techniques such as repeating parts of an image using the rubber stamp tool. She finds an interesting correlation between her progressing computer techniques and the evolution of painting styles.

The fantastic thing about it is you really can produce so many effects. It’s really fantastic! It’s a tool, you know. It’s like a brush. You can paint like Botticelli and you can paint like Van Gogh. In one, you can’t see the brush strokes. And in the other, the brush strokes are what matters. And so with the computer. It’s very rich. . . .

Besides, it has two wonderful options that I wish I had in life. One is save and the other is undo.

So where do we go from here? What kinds of policies will publishers create to help the work flow? Is this even possible to predict, given the technology’s rapid change? Like Dawson, Putnam art director Cecilia Yung feels they are still learning from their mistakes. For example, the complexity of the technology and the cost of the equipment are apt to take an illustrator by surprise, and the publisher may be put in the position of picking up the pieces. Yung has found herself having to ask established artists, “How long did it take you to master watercolor? How long did it take you to learn how to draw? Why did you think you could do this in three months?”

When Yung was working on some of the earlier computer books, she had to decide how to handle artist’s requests for leniency because of the special demands of the medium.

I’ve actually had people say, “If you want me to do this, I need more money in advance because I’m going to need to expand my hard drive.” And I’m thinking, But I’ve never sponsored somebody’s purchase of watercolor paint. So I think the biggest dilemma on the artist’s side is how to solve the technical and financial and business end while they’re learning a new medium.

Yung thinks that it is imperative to judge computer art by the same standards as other art. At the same time, she understands that any illustrator who wants to learn to use the computer is in a dilemma. She says, “Its been very rough on them because not only is it very frustrating — they’re in the process of learning a brand new language — but in the meantime it’s costing them a bundle.” But as an art director, she has to make sure the finished book looks as good as it can.

One of the things I always say to artists is, I really don’t care if you did it blindfolded standing on one leg with your left hand. Because if it’s a fabulous book, it’s just an interesting aside. But the thing is, it has to be a fabulous book.

What about the next generation of illustrators — the ones born in the last ten years who learned how to manipulate a mouse before they could draw with a pincer grip?

Ora Eitan, who teaches at Betzalel Academy of Art and Design in Jerusalem, works with young students who feel completely at home using a computer. “So for them it’s natural. It’s like using pencils. . . . I believe that it will be much more popular [in the future].” But she doesn’t think this means the publishers’ job will become easier:

I believe it will be much, much harder. Because it will develop on and on and on, and when you think that you know it, then they’ll come up with something new. It’s like the Red Queen in Alice through the Looking Glass: you have to run as hard as you can just to stay in the same place. So I’m running! Like hell!

What about some of the brand-new technologies that could have applications in picture books? I solicited predictions from high-tech innovators who suggested all sorts of Star Trek–ish scenarios. One said holograms would become big in picture books — not the kind Marcus Pfister already uses in Rainbow Fish, but the kind that really appear to be taking up 3-D space. Another suggested that nanotechnology — the development of microscopic atomic-level devices — could help create picture books that recognize a child’s facial expressions and load different story paths depending on the child’s reaction to the story so far. And then there’s e-paper (now called Electronic Reusable Paper), the flexible plastic sheet that is already being used for signage and could be made into a book that is blank until a new story is loaded into the spine. As in e-books, text and art could be downloaded from the Internet and stored in part of the spine or perhaps the cover. Just press a button and the text would format itself onto the pages. But when I queried Xerox about the applications of this technology to the under-two set, they admitted that they weren’t yet sure their e-paper was safe to chew on.

If those now-defunct CD-ROM divisions are anything to go by, though, it seems clear that picture book editors and art directors will not be asked to oversee these tech-heavy projects. Instead, Yung thinks a separate tech department could be developed, with editors and art directors who are well versed in new technology. But everyone agrees that in a world where more and more of a child’s diversions come with an on/off switch, the safe comfort of books will remain essential to a child’s sanity.

Eitan, for one, enjoys creating three-dimensional art and is excited about the artistic possibilities of some of this technology. She is not at all afraid that it would endanger traditional children’s books: ”It’s just another thing on top of everything we have. Like the theater was not gone because of the cinema. And the cinema was not gone because of television. So right now I don’t think that it will become instead of.”

It’s time the term computer art lost its stigma. Any book ought to be judged on its own merits, regardless of the medium used to create it. Undoubtedly, we will continue to see many bad computer books, just as we see bad books illustrated with watercolors, acrylics, and oils. Kathy Dawson says, “I think the biggest problem is that it’s like any other art form. You can be a master, and you can be a hack. But the difference is, it’s easier to be a hack in computers than it is in painting.” In the end, it is just another medium, and one to watch with interest as it develops and comes into its own. Janet Stevens sums it up:

It has to become yours. You have to almost wear it like a skin the way you wear watercolor or whatever you really have come to know. And then it’s incorporated into your being, almost. . . . There are so many misconceptions. It is just another tool that is really and truly fabulous. It’s frustrating and has all the drawbacks of another medium that you might just be learning. But it’s so vast, it’s scary in what it can do. When you get into it, it’s almost like printmaking: you never really know what’s going to come out.

Lolly Robinson is the designer and production manager for The Horn Book, Inc. She illustrated Mama Bear by Chyng Feng Sun (Houghton, 1994) using real watercolors.