Images

In this age of environmental and health awareness, science pervades the media. Amidst this pervasion, the American Museum of Natural History's "Science Bulletins" program is unusual: it consists of perhaps the only High-Definition science media made in a museum and for museum display.

Two designers from Science Bulletins discuss the antecedents of their product, its medium, and its setting-- from pre-eighteenth century illustrations and dioramas, to early motion picture projection, to the wunderkammer (cabinet of wonder) from which natural history museums sprung. With these forebears in mind, they analyze the challenges in designing current content for Science Bulletins and its subscribers across several countries.

These challenges exist both in designing the content of the medium and in acknowledging the medium's presence as a light source, backdrop, and/or theater within an exhibition space. Much of Science Bulletins' source imagery is data collected from objects invisible to the naked eye-- either their size or time-span is too small or too big to see, or the objects exist outside the spectrum of visible wavelengths. It is important to convey the reality of these objects while maintaining scientific fidelity. Considering the sequential nature of time-based media, avoiding information over-saturation, and predicting subtle effects like the "afterimage" in the viewer's mind-- these are all inherent steps in the design process.

One general issue persists: amidst the crowded cabinet of a museum's tangible artifacts, how does one design a movie (something inherently intangible) that acts as a true "curio"-- an unusual article that inspires a sense of interest, even wonderment? And is wonderment something to be desired by the designer as a targeted effect of his/her work?

Introduction

The history and news of the moving image in the cinema is well documented, but we encounter moving images in many settings beyond the black box: billboards, public and private transportation, hospitals and municipal services. Most of us keep a television, not to mention an Ipod, cell phone, camera, and computer that all play moving images. And moving images have become an integral part of cultural institutions. Often coupled with sound, they guide us to/through and expound upon the cultural artifacts we wish to see-- often, they are the cultural artifacts themselves! Especially with the durability of digital hardware to display it, the moving image has become its own cultural institution.

We are two animators of a kind of frequently updated moving image on permanent exhibition in the American Museum of Natural History (AMNH). This program is called Science Bulletins, and as far as we know, it is unique as perhaps the only High-Definition science media made in a museum and for museum display at AMNH and about forty other subscriber institutions throughout the USA, Canada, Asia, and Australia. Science Bulletins’ team is small (11 full-time members plus several part-time scientist advisors) but self-contained, consisting of a few producers, video-shooter / editors, production / distribution manager / engineers, writers, and animator / designers.

Each of Science Bulletins’ projects are developed by at least three people (a writer producer, an editor/animator, and a scientist) in an effort to create something with more expository impact than each could make alone. We are drawn to this multidisciplinary project, particularly in a setting as wondrous and jumbled as a natural history museum. We wonder about the antecedents of this project as much as we strive to make its products desirable and engaging. Tracing through AMNH’s own history as an exhibitor of moving imagery, we noticed several significant eras in its use. Examining these eras and our own everyday design challenges has helped us think about what we do in a fresh perspective.

The 1890’s/1900's: first moving image at AMNH.

The history of the moving image in the natural history or science museum is as poorly documented as its presence in the cinema is well documented. For this section, we are deeply indebted to Allison Griffiths, a professor at Baruch College and one of the few scholars studying the history of the moving image in the museum.

AMNH opened in 1869. The following year, the museum’s president Harry Osborn talked of this vision for a museum that will "bring visitors directly under the spell of nature.” His speech implied idealized versions of science and the natural world—in other words, dioramas.

In 1892, AMNH president Morus Jesup spoke of "rational amusement in the form of respectable recreational pursuits.” Both Osborn’s and Jesup’s speeches had somewhat elitist undertones—while the museum was ostensibly welcome to all, it was nevertheless a “refuge” from cheap entertainment like the penny arcade.

Expeditions in the early 1900’s help to fill the Hall of North American Indians’ habitat group dioramas. In 1908, Popular Science referred to the diorama mannequins as “doll people.”

The highly organized African Hall from 1909. This experiment in “scientific correctness and aesthetic pleasure” was a descendant of the "wunderkammer" (cabinet of wonder), large jumble-rooms of beauty cultivated in the late 1800’s by the naturalist-minded trends in Europe and America.

“Anthropometrics” was one early combination of photography with the scientific method. It was used on this Aboriginal woman in 1870. Its rationale was that “mechanical reproduction obtained via standardized photometric methods permit recovery of reliable data” and is more objective than a verbal or textual description. However, anthropometrics was only conducted on non-white, non-Western people by white Westerners.

Motion pictures made their debut at AMNH’s Public Programs in the 1910s. Paul J. Rainey, a playboy/adventurer, brought the self-financed African Hunt to Manhattan cinemas and auditoriums in 1912. While AMNH’s exhibitions took are more decorous tone, African Hunt was so wildly popular that some members were infuriated by their lack of access to the sold-out show!

The play “Hiawatha” was performed in front of a film projection the next year, again as part of Public Programs. It was a high-profile production with a Native American cast and props taken from the Hall of North American Indians.

The turn of the twentieth century encompassed a growing experimentation with photographic/filmic methods and scientific (mostly subjects. Partly because of technical hurdles and, perhaps, partly because film had such lowbrow connotations, the only inroads the moving image made into AMNH was through its Public Programs—a well established format that safely paralleled the proscenium/audience format found in film theaters around town. AMNH’s roots lie in an anthropological mission, which is why anthropology may have figured more prominently in its motion picture experiments.

1930s/1940s: Hayden Planetarium, AMNH Film Dept.

Information in this section is from the AMNH photo archives.

In the mid-1930s, AMNH began construction on its first planetarium. It may have been not-quite astonishing to museum visitors, considering that only forty blocks down the street in Times Square, the latest display gadgetry had been evolving in a much more ostentatious way. (at left: Hayden Planetarium, 1936.)

Nevertheless, the Hayden Planetarium was and still is New York City’s major planetarium. Its presentation format was not the proscenium/audience that film watchers were used to. Audiences found themselves watching projections on the domed ceiling that seemed to surround them on all sides. At left: the construction of the Planetarium dome, 1935.

A cutting-edge projector was installed (at left, 1935). It bore some resemblance to the advanced Zeiss planetarium projector that would follow in 1970. Planetarium projectors work by simultaneously projecting several flat images, and content for this projector format has always been limited. AMNH began to generate some of its own content for the planetarium, whose complicated technical demands continue to be a driving force for AMNH content. This is still true today.

The planetarium was also a significant development for the moving image in the museum setting because not only was it a technical projection challenge, it was a major display of scientific (astronomical) data in the moving image.

In the 1940’s, AMNH opened a film department that created field-based nature movies for school and TV. The movies from this department were a kind of precursor to the field-based National Geographic shows we see today. This particular photo was taken during the filming of a documentary on Harriman State Park.

If one thinks of the science and natural history museum partly as a campus of research scientists and partly as a display of current research, then it follows that current technology would be used to display this research, with the moving image playing a key role.

1960's: Bicentennial Exhibit, "Can Man Survive?"

Information in this section is from the AMNH photo archives. Thanks to AMNH head archivist Barbara Matthe for directing us to this show.

The museum turned 100 years old in 1969, and its board and curators aimed to mark its birthday with a major blockbuster show. The result was “Can Man Survive,” a half-million dollar (quite high for the time) show that stayed up for two years (quite long for a temporary exhibit). Current notions of technology, environmentalism, and postmodernism threaded its way through the show.

Far from the usual AMNH exhibit, “Can Man Survive” abrasively challenged visitors to examine the sustainability of industrialized life. A truss designed by Osaka engineer Masao Tanaka conspicuously contained the exhibit in AMNH’s main foyer.

The show was divided into societal trends that degrade the environment: littering, large cars, noise pollution, water contamination, industrial soot, cruelty to animals, and overpopulation. Increasingly narrow, decreasingly ventilated corridors served to physically impact the show’s points on visitors. A mix of slides, film, and video were projected onto randomly slanted walls already full of text, texture, and graphics.

“Can Man Survive” opened just two years after the Sony Portapak, the first portable video recorder, was introduced to the market in 1967. Perhaps it is consequential that while "video art" is said to have begun with Nam June Paik taping the Pope in downtown New York, a more institutional kind of "video art" was being prepared uptown for the city's natural history museum.

It would make sense that portable video technology would change the face of all kinds of exhibitions, not just that of the fine art gallery or museum. Film and its projectors were too expensive and high-maintenance to use in multiple sites unattended, and broadcast television did not allow for closed circuit looping.

But portable video technology was cheap and durable enough for multiple systems to be set throughout the museum, independently, without the need of a person to unload or rewind the image medium. Then as now nevertheless, audiovisual technicians (one is seen at left) had to spend many after-hours maintaining equipment and media.

Almost smack in the middle of the Soviet-US space race, 1969 was also the year that humans first stepped onto the moon. The space race drove much technological invention and fueled the popular zeitgeist. In fact, one curator of “Can Man Survive” said that AMNH wanted its exhibit to be able to compete with effects-laden blockbuster movies like “2001.”

Though the popularity of “Can Man Survive” reached the levels of African Hunt, that early movie shown at AMNH, it received a harsh critical response. Many journalists were affected by the show’s own harshness, and some conflated their response into their evaluation of the show’s conceptual quality. Some articles, like the New York Daily News piece shown at left, played their rage tongue-in-cheek.

2000's: the New Space Center.

AMNH demolished and rebuilt its astronomy division and planetarium in 1998, calling it the Rose Center for Earth and Space. This renovation sparked the creation of Science Bulletins, as AMNH sought its own high definition content to match its imposing new building.

These current-science video loops were first designed for the Earth and Space halls, and have also been created for AMNH’s other renovated halls, that of Biodiversity and Human Origins. Each of these loops is updated every week and includes a 2-minute news-snapshot, a 3-minute data visualization, and a seven-minute documentary following current researchers in the lab and field. Each Science Bulletins product is a collaboration betweendesigners / animators / editors, producers / writers, and scientists.

Desktop computers, in cheaply allowing previously disparate data sets to be easily synthesized, transformed organization and visualization in scientific research. Digital technology has changed the face of science exhibition, allowing for HD output and interactive kiosks that are so widely implemented.

But after working at AMNH for 3 years, we have noticed that while media technology may be popular among exhibit makers, it is unevenly used by exhibit visitors. With so many stimuli to take in, they are seldom compelled to watch more than a 1 or 2 minute movie, even in a black box.

Unlike in the cinema or even in front of a TV, visitors construct fragmentary narratives of an exhibit's objects. "Museum-fatigued" bodies are more willing to submit to a movie if it's at the end of an exhibit and if it includes benches. Adults are more likely to have the patience and fatigue to watch these movies.

Discussions that arise from past conflicts between the educational role of a museum such as the American Museum of Natural History and the “light” entertaining character of the cabinet of wonders seem to arise in our design discussions today.

Our Duties as Designers in the Science Bulletins.

A. Ducao focuses mainly on 3D (three-dimensional) elements in the bi-monthly visualization and documentary. I. Koen's duties focus mainly on the weekly news-snapshot production. We have both done work for all parts of the loop and in all content categories.

While our duties may be divided by format, our (often shared) challenges would probably be best divided by conceptual approach: the design of text templates and color palettes; the illustration of phenomena using the conventions of drawing and cinema; and the visualization (an actual TRANSLATION) of scientific data using computer programming and the mining of external databases.

Amidst this fascinating yet frustratingly mercurial setting, we face the challenge of translating large amounts of scientific information into a design that the swiftly passing visitor can be drawn to and accurately understand. Much of this information is collected form objects invisible to the naked eye; they are too big/small to experience in the time, space, or light that humans can perceive. Consequently, many of our major (and most interesting) design challenges involve translating this data and making the results appealing and understandable.

Sloan Digital Sky Survey: Showing how two-dimensional tools can translate three-dimensional space too large and far for our eyes to see.

Many of the astronomical visualizations we make are based on a large AMNH-maintained data set called Digital Universe (homepage at left). Updated once or twice a year, Digital Universe includes star, planet, nebula, and quasar data for the Milky Way and nearby galaxies. Digital Universe can be run on the Partiview (short for Particle Viewer) and Uniview software packages.

Digital Universe is divided into several file sets with the extension “.speck” (also referring to particles). Each line of these files gives attributes (like location, size, color, brightness, albedo) for one object (like a star, nebula, etc). The particular file at left contains data from the Sloan Digital Sky Survey, which uses an extremely high-resolution telescope, digital camera, and pair of spectrographs to create an extensive review of the sky.

We use the programming language Perl to translate 3d data files into something that Maya can understand. Maya is the 3d software package we use. The Perl script at left translates the speck file above.

The result from this translation is a file written in MEL, or Maya Embedded Language. Each line in this MEL script corresponds to a line in the speck file. For each line, Maya will create a polygon with the same attributes of the corresponding object in the

speck file.

Once the MEL script has been run, the Maya GUI (graphical user interface, screenshot at left) can now be used to animate a camera to travel through these “stars.” The camera movement is the crux of this particular animation: though SDSS surveys the sky in 2d (two-dimensional) strips (which you can see in the upper left corner of this screenshot) moving the camera into these seemingly 2d strips shows how SDSS’s spectrographs can assign spacial attributes to

2d data.

After the view from the animated camera is rendered from Maya, color, glow, and text can be added in After Effects. At left is a screenshot from the final HD animation.

The Electromagnetic Spectrum: translating non-visible wavelengths into the visible range.

With many of the phenomena discussed existing outside of the realm of visible wavelengths, we've begun to develop an extensive key to interpret where on the EM (electromagnetic) spectrum these phenomena lie.

Of the fields the Science Bulletins covers, astronomy tends to use the whole EM range most widely and frequently—there is just no other way to collect data so vast and far. The Astro Bulletin (left and above) is where the EM key was first implemented.

A certain spectral range (usually non-visible) is used to capture each image displayed in the Astro News/Snapshot. That range is highlighted in the EM key.

The key is maintained in three parts, the first being the spectra labels ranging from long radio waves to short gamma rays. These labels are made in Adobe Illustrator.

The next part is the wave shape, made in Adobe Illustrator and Photoshop.

The wave shape masks a visible light “rainbow.” This rainbow is reposition within the wave shape mask, depending on which spectra are shown in the key. The shown spectra changes depending on what range was used to take the image under discussion.

For instance, to show that an image was taken with a very narrow part of the ultraviolet spectrum, we use a different set of spectra labels. Instead of the broad EM spectrum, we only show the range from visible to ultraviolet.

The EM spectrum key has been modified for the Bio Viz to interpret false color satellite data highlighting landcover characteristics like forests, farms, and cities.

Most Earth-observing satellite data is collected over the infrared and visible spectrums. By first applying 3 datasets outside of the visible range respectively to the red, green, and blue ranges of visible light then compositing the three sets, a false color image is made. Different false color combinations highlight landcover characteristics more clearly than the true color combination, where the data taken with visible light is shown with visible light.

Trade routes: translating the Earth’s scale.

One way to make bio-geographical data more intuitively understandable is to place it all on a sphere to simulate Earth. The image at left shows the world’s major trade routes, rivers, used in an Invasive Species feature story.

Readability/Visibility is an element affected by the resolution of the medium. The HD canvas of 1920x1080 may be larger than standard- or DV-definition canvasses, but it is still a hard industry standard dictating the need to use 3d camera zooms to convey high-resolution, small-area data.

Many people use visual aids like globes from a young age, and now applications like Google Earth have made accessible the concept of multiple datasets applied to the globe. Our job is to take those datasets and make their appearance and movement more seamless than what is currently seen in a real-time environment like Google Earth.

Tiling the Earth: Translating between the globe’s local and distal scales.

Most Bio and Earth stories involve events that take place in a specified geographical space, and for that reason it is important to illustrate the geography. We are using the BlueMarble Next Generation (BMNG) dataset, which is rendered by NASA's Earth Observatory with a maximum resolution of 500m per pixel.

Bathymetric and topographic data are also available. The BMNG dataset can be processed to align different projections (image earth events – bio news). The projections most often used are the Apianus (oval) and cylindrical projections.

The BMNG dataset tiled in After Effect comprises of 200 tiles in 500m resolution. This gives us the flexibility to zoom in seamlessly from 20,000m per pixel (40,000/1920) to the maximum resolution of 500m per pixel.

Some should not to see the map of the world as a topographical abstract drawing signifying location and morphology only, but also as a collection of real data.

The screenshots in this section show how tiling is used to create a seamless zoom into the ancient Angkor Wat site in Cambodia.

Earthquake Plot: Compressing years into seconds.

Providing temporal information is often challenging because it can be distracting to the visual flow of the story. In the case of the Earthquake Plot that deals in large amounts of temporal information, computer programming/scripting

becomes necessary.

Certain datasets are too massive and would take far too many work hours to be plotted manually. One example is the earthquake data maintained by the US Geological Service (user interface at right). About 8000 earthquakes (most of them minor) occur each day, so plotting multiple years of earthquake data is very labor-intensive.

Scripting shifts that labor from the animator to the computer, freeing the animator up to concentrate on design aspects. The animator uses the user interface above to be given this list of “Earthquake Search Results” (at left). In this case, the results show all the world’s earthquakes for the last three years.

Each line of the Earthquake Search Results gives the latitude, longitude, magnitude, depth, and time of one earthquake. A java script (at left) takes each line of the Earthquake Search Results and translates it into an Adobe After Effects script.

When the script is run, each earthquake is placed on an Apian projection in After Effects (at left). When three years of earthquakes, about six million in total, are compressed into one space over a short span of time, the shape of the tectonic plates emerges.

A bathymetric image of the Earth is laid under the Earthquake plot, with data annotations and a time bar laid on top. With the computer now scripted to take care of the actual timing and positioning of the plot, the designer can tweak elements like shape and color of the plot itself, as well as font and positioning of the title bar.

3d development of the Human Bulletin.

Working with the Human Bulletin, Science Bulletins’ newest content category developed in parallel to the overhauled Hall of Human Origins, allowed for a fresh approach to designing moving imagery for the museum. At left and below: cell designs in development (left) and in final output (below).

Moving imagery deals with more than the two-dimensional canvas surface. That extra dimension is, of course, time. Statistically, a visitor spends an average of 7 minutes in each museum hall, leading to presumably less than two minutes on each particular artifact or object.

The time the visitor spends on an exhibition provides an evaluating factor on the structure and durations of the media pieces presented. Our approach with this project is to keep text, data, and illustration more isolated by color, position, and form. At left: a 3d model of the Human Bulletin’s “Avatar,” use in the loop’s introduction.

Some of the content in this Bulletin comes full circle to the earliest motion pictures at AMNH and their anthropological content, though audiences may be more media-savvy and perhaps even oversaturated. A final version of the Human Bulletin “Avatar” is at left.

One major difference of this Bulletin is the dual focus on paloeontologic and genetic discovery. We hope to soon introduce molecular data into the Human Bulletins as a way to sharpen the way we use moving imagery and animation to explain current science research.

Human Bulletin Color Development.

The "Human" category gave us the opportunity to develop a new Bulletin completely from scratch, and color theory has played a particularly important role. This is for both utilitarian and design reasons. (At left, the Munsell color system developed in the early 1900’s. Its Hue/Chroma/Value axes were the precursors to the Hue/Saturation/Brightness systems we see in computer applications today.)

Not as much visual data is currently available for this category. In other Bulletin categories, a color scheme is determined by how the data's information is mapped to the visible spectrum. At left is the color picker used in most animation applications, with the HSB (Hue Saturation Brightness) levels based on the Munsell system.

Without much data, most input sources are photographs. We created a series of animations and "wallpapers" to support these photographs. At left: animation of DNA being “sequenced.”

We chose of palette of generally cool hues, low saturation, and high brightness to best display dark-color text and warm, saturated annotations and highlights.

This process has led us to rethink how we handle data in other content categories. Generally highly saturated and bright across all hues, we've begun to experiment with toning down HSB to highlight salient areas only.

Color similarities between the object and the text can help the viewer identify the annotated elements that some times exist in distribution all over the image and not always annotated in their full.

Other ways work on the annotations and illustration on the image initialize all the three dimension of color. For example, a full range of brightness helps achieve maximum contrast. Other solutions include the use of complimentary color to achieve color-contrast.

Sometimes, especially when the image is initializing most of the spectrum, (cases like are usually false color satellite data - that transpose non visible colors or make more vivid visible colors), the task of annotating the image becomes more complicated.

Conclusion

The main challenge of working with this kind of visualization is to convey an idea as clearly as possible. The information used is extruded from scientific research. The scientific method of data collection and qualification provides the axis for the design process as well.

In addition to the problems that might occur in the process of computing/rendering the information; presenting the complete set of information can confuse the viewer and prove the visualization unsuccessful. Therefore, the a major part of the entire design process is to identify what information from this data set is useful to the point of the visualization.

We wrote this paper to explore the logic behind our design choices and to set it in the broad history of moving imagery in the museum. Even after this extensive exercise, we are still unsure as to whether our conscious design decisions over the past three years have had any impact on making the content more intelligible and/or appealing. In our day-to-day experience, we don’t see a major change in AMNH’s audiences for Science Bulletins.

Perhaps one reason for this uncertainty is that our society is conditioned—by media and educational systems—to deconstruct work set in the cinema/theatre more skillfully than that in museum. The holistic, linear narrative is more conventionally satisfying and easier to analyze than the fragmented narrative that we have to construct ourselves. As Julian Schnabel says in the 20 November 2007 New York Times, more people understand movies than understand painting. Even as designers trained in a cinematic paradigm, we are not satisfied with our own work unless it signifies “one complete thought.”

But perhaps there is something rich to be found in that space between the designer’s linear thought process and the visitor’s fragmented narrative construction. Some of Science Bulletins’ most enthusiastic spectators are those that watch at Johns Hopkins’ University’s Earth and Planetary Sciences Lecture Hall. Undergrads, faculty and staff drift by the Bulletins on the way to class in the mornings and treat it like a “daily dose” of NPR-like news. Sometimes they even gather a bunch of chairs and “settle in” for an evening beer session. This significantly points to what may be a promising future for the movie in the museum—to dedicate lounge-like spaces to this media and invite experts, non-experts, and those in between to settle in (perhaps with food or drink) for a dose of scientific information. The lounge setting may be particularly useful for science and natural history museums, because much of the media the present is text- and/or fact-heavy and cannot be quickly absorbed from a walk-by or even a 1-2 minute watching session (as it may be in, say, a fine arts museum).

However, there are two major challenges to the potential of the lounge being realized. The first is the aesthetic separation between children and adults. As natural history museums have become more media saturated, they have begun to look more like the sets of kids’ television shows. Their advertising has also become more targeted towards children. Attracting children to science is an admirable motive, particularly in a country where the study of science is falling sharply, but the primary- and neon-colored, exclamation pointed aesthetic trends in natural history exhibition design often alienates or overwhelms these children, not to mention the teenagers and adults that accompany them. Children do not need to be pandered to, aesthetically or conceptually—in fact, many a child’s holy grail is to access those things termed “adult.”

The second challenge is the separation of educational spaces (mostly exhibit halls) from break-taking space (like lunch rooms, museum cafes, restrooms, and lobbies). In many museums, break-taking space is treated as something separate, bland, and ignorable. However, it is these spaces where museum audiences are most captive—whether it’s the family with hungry kids, the couple who needs a coffee, or the lone person who needs to use the bathroom. There is much potential to use these time-hijacked spaces for time-based media and to make the “edu-tainment” experience into something threading through every corner of the museum, not just certain parts of it. Even when exhibition halls are turned into lounges, like at AMNH’s First Friday Jazz Nights, large media screens are deactivated. This might be a waste of the medium’s potential.

There may be a more actualized future for the moving image in the natural history museum, one where visitors are more attracted to actually watch exhibition screens. Perhaps the solution lies in moving the screens more to the edges or out of the exhibition altogether. Judging from the screens set near many exhibitions’ entrances and exits, there is already some movement towards this approach. Perhaps the solution lies in designing even shorter content that “backgrounds” itself more. Whatever happens, moving imagery, particularly HD media, continues to gain traction in the museum. It is to a museum’s benefit to consider the way its visitors really use it.

[ Back ]