Figure 3.37
Ian Campbell: Push: Work Class

3.4 Animated, Interactive, Multimedia, and Randomized Collage

With animation, interaction, sound, and random selection of fragments, the concept of collage finally comes unglued. Here collage becomes fully digital and native to the Web. Each of these overturns the basic definition of collage as an assemblage of pieces of other images and pages selected, shaped, and glued to a backing to make a finished work of art. Inserting an animated fragment may not seem like a radical change, but it is remarkable how even two animated sections creates a sense of activity on the page which requires alert attention. Figure 3.37 links to one piece in a suite by Ian Campbell, who sets a standard for subtlety for the Web that few even attempt to meet. He has another, hypnotic collage with a clock and the sound of ticking that simply animates change of focus from an apple to the clock. Technically, these are very simple—no JavaScript—which for some time was good, because the development of JavaScript has been torturous.

Figure 3.38
Shirin Kouladjie: When Nobody Was Looking

Far more flexible and extensive animation effects are possible even with HTML (and Javascript), but Macromedia Director and then Flash have become very popular for their ability to integrate sound and scripted animations. A good bit of Shirin Kouladjie's recent work has been finding uses for these powers. If you click on Figure 3.38, and your browser is equipped with a Flash 4.0+ plugin, you can see some of the abilities of Flash to define areas and move panels and layers. The idea of a face parts of which slip "When Nobody Was Looking" might seem rather out-of-the-blue if you just wandered across it, but in relation to the themes and practices we have been tracing, it is fairly clear: it animates a process of strip- and plane-shifting that goes all the way back to Gris, many of whose Cubist displacements can be restored, at least partially, by realigning their strips. (See Poggi: 112-116).

Figure 3.39
Shirin Kouladjie: Faces

"Faces" (Figure 3.39) is a more subtle variation on the collaged face. These looks like an animation of Bernie Stephanus making a face graft; I find myself looking very closely as if I could reach out and pin it in the best position. These four variations on the face are so photorealistic that they invite contextualizations as images in various kinds of mirrors or water and the slight motions give an uncanny sense of living, breathing people. BTW—you can drag "your" viewing eye anywhere in the window you like in all four versions.

Macromedia Flash uses scalable graphics, so Flash collages also are scalable. As delivered on the Web, some run so frenetically fast on an average system that they challenge the eye and seem to dissolve into a flurry of rushing about like a Monty Python sequence. However, if you load them into a larger window, they slow down considerably (though if they have music with a strong beat, it may synchronize only roughly with the video changes). Even when slowed by enlargement, The Polaroid is a somewhat frenzied hyperadvertisement that mocks the Polaroid commodification of family life with phrases like "Plastic Moments of the Past" and "Kitsch Family" running through a display changing vertically and horizontally and accompanied by the sound of a child's piano piece.

Figure 3.40
Martha Rosler: frame from Beauty Knows No Pain (1966/72)

A page begins to be interactive when input from the viewer changes what is displayed. Any interactive feature moves viewers from passively observing scripted screen events to actively seeking and choosing what they see. Thus defined, interactive is very broad and includes such mundane activities as scrolling, placing ("touching") the cursor over various parts of the page, and clicking on links. For our purposes, a narrower focus is useful, which is the property of a page to change its images according to viewer input. This includes "mouseOver" events where you can change part or all of an image by passing the mouse over it. (The "part" effect arises from making a hidden layer or division visible, or from changing one cell in a compound image.) Figure 3.39links to one such JavaScript driven "rollover." The image is not native to the Web but is an adaptation of a large collage done by Martha Rosler around 1972. Bearing no title, it a large (104 x 71 cm.) collage of Playboy bunnies all in a reclining position so that the array appears to be like a beach. In the retrospective collection (Positions in the Life World), this piece is reproduced as a double fold-out 4 page spread, but the Institute of Contemporary Art site gave the work this rendering with a rollover giving different segments of the center portion of the work. This scales down the work, of course, but it simulates the eye darting from place to place, always finding more of the same. This would seem to fit in nicely with Rosler's strategy of supplying so much that its essential repetition and sameness are overwhelming. A recent site devoted to this idea, and even theorizing it, is Jean-Luc Bohin's Labyrinthes du Désir which, however, is not interactive.

Figure 3.41
Ian Campbell: Sweepings

Even more recent developments in JavaScript allow you to move parts of a compound image (i. e., DIV- or LAYER-tagged elements) by clicking on them and dragging them with the mouse button down. They usually stay where they are when you release the mouse button.

Various applications were made as soon as this script became available. Ian Campbell concluded his "Push" series of images with a Sweepings page (Figdure 3.41 is a snapshot), in which fragments from the preceding collages (all rectangular "boxes") can be rearranged by the viewer in a sort of "now you try" gesture. These fragments are all fairly well-defined, with edges; the "good forms" of objects represented in them are not broken by juxtaposition, although you can further fragment them by dragging another box over to partially cover them (if the dragged box has a higher z-index; the z-index (order of stacking) is fixed in this composition.) They are fragments of letters, photographs and other things many of which have appeared more completely in earlier pages.

Figure 3.42
Friederike Paetzold: I-section

Another early application is Friederike Paetzold's rather grisly "I-section" where you can open a virtual torso and remove the vital organs, in return for an emblem and saying for each organ. Paetzold uses GIFs with transparent backgrounds for the inner organs that you click and pull aside, so that you seem to be seizing and moving the organ itself (though in fact it is still in a rectangle). In this snapshot (Figure 3.42), the torso has been opened and the intestine and ribcage removed. The secondary window with text and surgical apparatus pops open when you click on the rib cage. Here moving the parts is as it were exploratory surgery; one is not contemplating an esthetic rearrangement of the human torso. Another artist who found early use for the moveable GIF objects is Christy Sheffield Sanford. (See her Rock-Garden: garden-path.) For a veritable showcase of the possibilities opened by moveable panels, see her "The Roots of Nonlinearity: Toward a Theory of Web-Specific Art-Writing" in Beehive, 3.1 (2000)

Figure 3.43
M.D. Coverley: Tidelands:Swim/high

Figure 3.43 is s snapshot of one of the pages of "Tide-Land", a work by Marjorie Coverley Luesebrink writing as M. D. Coverley which appeared in Beehive, 3:4 (2000). This work of fiction combines panels of lyrical text evoking memories associated with tide water coast with old pictures and with nature sounds and guitar music. The audio component contributes to the immersive reverie or quasi-state induced by the pages. Most pages have several transparent panels of text and open with the panels overlapping each other so as to make them only partly legible. The viewer thus has to pull them apart in order to read them—that is the primary reason for moving the panels, not arranging them into a design, since the panels are mainly text. Moving the old family snapshots around has an effect similar to going through a drawer or box of old photographs which speak their various texts to you. With the panels with transparent background and text, however, we have moved beyond the idea of moving objects around at all; the effect is more like moving transparencies prepared for an overhead projector. We move them to find contexts and captions for the pictures, and so have entered a realm where the writer does not fix any one relation of text and image but leaves it up to us and the semiotic conventions we may choose to apply.

Figure 3.44
Curt Cloninger: One numbered fragment from the Mechanical bubblegum card

Once authors have begun to turn over control of the location of the fragments in a collage, they can next give up final choice of which fragments appear on the page. Curt Cloninger has sets of "bubblegum cards"—flash cards for different words. At present he has sets for seven adjectives (electronic, mechanical, postmodern, etc.). The "cards" (pages) in each set are autogenerating, by which he means they are written with component image and placement selected by a random number generator; they are re-written every few seconds (also randomized). The images are smallish .gifs "cut" with transparent parts (Figure 3.44 is one example). They are given different layer (z-index) values, and stacked a few deep on each writing of the page. The look of the combinations is quite complex, unpredictable, and shifting, and because of the eroded shapes of the components, it requires some patience and study to make out what is going on. Each card is accompanied by an appropriate, somewhat sound-collaged track with a beat that very roughly syncs with the refreshing of the page. A similar randomizing technique is used by Robin Michals in "e-arcades.com" to juxtapose randomly selected fragments of text with randomly selected backgrounds, again, selected from thematically coherent sets of texts and images, and I have adapted it in my reworking of an online fragment of Benjamin's Arcades Project.

To remove even this degree of artistic selection, Amy Alexander's Multicultural Recycler gets its images from on-line Webcams, selected from a list either randomly by the computer or by the viewer. Here we are approaching the extreme end of a theory of art. Our last example pulls back slightly from this theoretical pole.

Figure 3.45
Dirk Hine: from Photomontage

Dirk Hine calls what he does "dynamic photomontage," but I place it under collage because it works by movable, semi-transparent objects which (in Internet Explorer only) can be dragged around with the mouse. At times, when several overlap, it may not be clear what the component images are, but dragging separates them. The transparency sounds like a photomontage technique, but no space or world is represented: the background is simply black. If one wanted to argue the point narrowly, it could be noted that fairly transparent fragments were first used in collage by Braque, and the dress pattern fragment in Pat Street's "Effect in Yellow" is also nearly transparent. But of course they are the exception rather than the rule, as with Hine's work, and they don't give the effects of juxtaposition or cutting off so fundamental to collage. The images used certainly lean toward the collage side, since they are made from old photographs, diagrams, pages of books, letters, stamps.

Figure 3.45
Dirk Hine: Dynamic Photomontage

The truly remarkable and as far as I know unique property of "dynamic photomontage" is that the images that appear are chosen at random from a fairly large bank (100); their position on the page, their z-index (place in the stack of layers), and their opacity is set randomly. Further, the viewer can rearrange the page (though sometimes the image moved is set adrift on the page). The crucial role of the artist in arranging the fragments is thus given over to chance and the viewer. The shaping role of the artist is still there as the selector and preparer of the fragments and in the conception and execution of the amazing JavaScript. One could carry on a bit about the Postmodern artist and so on, but to me the special quality of this work is the experience of the viewer. One is in effect dealt a hand of images and can happily set about arranging them. However, when some of them begin to drift, you realize the product of your work only lasts an instant. You could of course fill up your drive with screen captures or try to produce an animation. "For some time now," Hine says, "I have thought of art as a verb, rather than a noun." ( Subterranean notes, April 00)

Conclusion to Part I: With moveable, digital collage, we have left behind four of the main characteristics of papiér colle, including those that give it its name— paper and paste (and fixity), as well as being broken forms of images and breaking the form of the images over which they are pasted. Collage with its hard edges is almost the opposite of photomontage with its soft edges and variable opacity. They are the two great ways of making images which are multiple and not representations of single views into a coherent, naturalistic space. Photomontage softly blended multiple images into a single space, thereby requiring some new rules for interpreting the non-naturalistic world that resulted. The lack of any space except the black void in Hine's work was our reason for withholding the photomontage label, although in fact this blending without regard to a prevailing world has become a major practice of photoshopmontage. For illustrating complex concepts and multifaceted experiences, pasting into a single space and world has little to recommend it. As collage becomes less material, and photomontage releases its anchor in worlds organized by vanishing-point perspective, the two quite different traditions begin to overlap. When the materials and material acts of cutting, tearing, and pasting are replaced with the Selection Tool and Pasting, a change in sensibility is also likely to occur, since the main art of selecting figures digitally is to get just the figure and all of the figure. (To be sure, digital tools also give other, quite novel ways of breaking the surfaces and contours of things, such as making transparent all the pixels of a certain color or range of colors in an image.) Also, the cutting and tearing of collage destroys the original image, which is one reason collage suggests repudiation of the world of the original images. Where collage is violent and abrupt, photomontage (as we have defined it) is gentle and flowing. Those contrasting connotations, which arise not just from the material acts of fabricating but as well from the traditions of use that we have traced, are part of our visual culture. They will doubtless be supplemented and modified as film negatives and enlargers join scissors and paste as increasingly retro tools of the purist and nostalgic, like typewriters and sliderules. On the other hand, scissors-and-paste collage is a very powerful expressive tool to which virtually everyone has access. The technology does not require electricity, and is introduced in every primary and middle school I have ever known. The feel for papiér colle may endure as long as there are magazines and photographs and maps and notices to cut and tear. After all, we still draw with charcoal.