99% Invisible - 465- Shirley Cards

Episode Date: November 10, 2021

Even if we think of the camera as a neutral technology, it is not. In the vast spectrum of human colors, photographic tools and practices tend to prioritize the lighter end of that range. One example of this bias was Kodak's Shirley Card, a reference photo used to calibrate photo printing machines.  For decades all of the models on the Shirley Cards were white. This meant that photographs of people with darker skin tones were often not printed correctly. But that's just one example of the limited dynamic range of photography purposefully excluding people with darker skin.Shirley Cards

This is 99% invisible. I'm Roman Mars. Today we're going to start back in the 1970s at a Catholic school in South Los Angeles. It is picture day and all of the kids are filing into the school auditorium in their uniforms. They wearing powder blue sweaters, white shirts, and dark coloroys. And they would be a photographer, they're usually a white guy. This is Ibario Dexperelo, who attended the school. And they would just march all the students in, and you would sit on this stool,

and you have your look at the camera, and you would make a couple of pops to the flies, and then you were done. After a few weeks, the students would get their photos back in a big envelope towards the end of the school day. And everybody wanted to share what everyone else's photographs looked like. They wanted to see what do your pictures look like. And the kids who had darker skin tone were always just, there was always a kind of negative reaction to that. Because the kids with lighter skin

looked basically like themselves, but the kids with darker skin looked way darker than they did in real life. A lot of the detail in their faces was just lost in the shadows. And the kids, kids could be cruel. You know, they all, you look so dark, it looks so black. That was the first sense that there was something different about how people look in photographs. That experience stuck with him, but Porello didn't realize why this was happening until later in his life. When Porello got a little older, he learned how to use a camera. Reporter Kai McNamey.

And the process of creating an image out of nothing just seemed like magic to him. But even as Perillo grew to love photography, he kept running up against the problem he first noticed back on school picture day. Capturing images of people with darker skin tones posed certain technical challenges. Sometimes he felt as if he had to work against the limitations actually built into the camera and the film. And that's because even if we think of the camera as a neutral technology, it is not. In the vast spectrum of human colors, photographic tools and practices tend to prioritize the lighter

end of that range. And that bias has been there since the very beginning. It is not an overstatement to say that the history of photography can be divided into two eras, BK and AK, before Kodak and after Kodak. In 1888, George Eastman introduced a camera that revolutionized photography, making it accessible to thousands of amateurs. Kodak simplified the camera

and offered to process film for the consumer. One of the company's early advertising slogans was you press the button, we do the rest. Kodak quickly grew into the biggest player in the industry. The company became almost synonymous with photography itself, like the Kleenex of cameras. Kodak wanted to be the company that recorded family histories. This is Lorna Roth, a recently retired communications professor

at Concordia University. By the mid-20th century, most people were shooting their family pictures with Kodak film. most people were shooting their family pictures with Kodak film. ["The World of the World"] ["One Little Girl, One Precious Childhood, save for years to come, in pictures. You can do it too. All it takes is a camera, Kodak film, and thoughtfulness."

To meet the booming demand for photo processing and printing services, one of our photo labs popped up all over the country. And Kodak supplied a lot of these labs with printers. Here's where you can start to see the enormous power the company had to set the standards for film and film development. Each Kodak printer needed to be calibrated and standardized before photos were printed on it. And so the printers came with something called a Shirley Card, which was a color reference card created by Kodak in the 1950s. The original one had some color swatches and a picture of a white woman named Shirley Page, who worked as a Kodak employee at the time. She was used as a basis for measuring and calibrating the skin tones in the photographing printers.

The way it worked was that Kodak would ship printed Shirley cards to photo labs across the country, along with the film negatives needed to print the same image. Lab technicians processed the film, printed it, and ended up with multiple test prints. This allowed them to compare the Shirley cards printed at Kodak with the Shirley cards printed in the lab. If something didn't look right with the colors, they'd adjust. Once the printers were set, they'd start feeding film into the machines, cranking out hundreds of rolls a day.

As time went on, Kodak began including other women on the cards, not just Shirley Page. And they, I think, just decided to name all of the women who were going to be in that position. They were all going to be in the end, surely because it was easier that way. They were all going to be in the end surely because it was easier that way. But even as Kodak expanded the number of women

represented on the cards, all the surely's continued to be white. Whiteness was invisible, whiteness was the norm. On some of these images of the surely cards, it was written the normal. Because all the surely's were white, it meant that the printers were set for white skin. And Shirley was basically used to calibrate every printer, every time, regardless of the color of the people in the actual photographs being printed.

To get accurate prints or a person with darker skin or even someone in a different lighting environment, you might have to adjust the printer settings. But that just wasn't happening at most one hour photo labs. They weren't about customized service. They were about being fast, standardized, and relatively cheap. Because it was about getting that stuff out because you had a line of people out the door,

dropping film off and waiting to pick it up. This is Everard Williams Jr. He teaches photography at the Art Center College of Design in Pasadena. Back in the 80s, he worked at a 1-hour photo lab. And he says that lots of different kinds of people used to get their photos printed there. The demographic was everybody. It wasn't like it was say, oh, it's old white men that were wealthy that had enough money to buy a fancy camera and shoot some film. You would

get everybody because making pictures is important to everybody because it helps kind of preserve the family legacy memory of something that's important. But the way that photography prioritizes white skin goes beyond the role of Shirley Cards. If that were the only problem, it would be relatively easy to fix. Instead, issues can start much earlier in the process,

including the lighting and camera settings used to capture the picture. Take, for example, Ibario Nexperrello's Disaster School Picture Day. That day, the photographer probably showed up and set up the lighting the same way he'd done hundreds of times before. Yeah, they probably just got in front of it, took a light meter reading, where the kid was supposed to be, gave him a neutral value for that exposure and that's what they went for for the rest of the day. That was it. The photographer made a call because that's what photographers have to do.

Taking pictures is all about trade-offs and making decisions about which colors you want to prioritize. In this case, the photographer prioritized the lighter skin kids during the setup, which meant that the kids with darker skin tones just didn't look as good. There was no nuance to, you know, making it look good or adjusting it for the nuances of skin tone. It was like, you got an exposure,

you take the picture, move on to the next kid. But even if the photographer had finessed the lighting in camera settings with a range of skin tones in mind, there would still be another problem. The film itself was optimized for white skin. If you imagine light as a gradient from black to white, there's only a small chunk of that gradient that's visible to a camera. Dynamic range describes the width of that chunk that a camera sees. The wider the range, the more tones the film is able

to capture, the more accurate the image. But historically, consumer film, like the kind made by Kodak, was pretty limited. If you have experience making pictures of people with darker skin tones, you know how to compensate for all these limitations in the technology. You know you need to change the shutter speed, open up the lens or use a flash. Film may have challenges recording shadow detail, just the physics and the chemistry and how film functions, but there are ways to compensate

for that in terms of how you might light something or how you might expose something. So as a black photographer, photographing largely black people, we're very conscious of that. Photographers like Gordon Parks made amazing and indelible images of black life using the tools available to them. But that's because they cared enough to figure out how to do it well. Many photographers just haven't, resulting in the kinds of images that Porello remembers from picture day. And it might seem like a small thing, school picture day. But multiply that experience hundreds of thousands of times, not just with professional portraits,

but also with home photography, all the birthdays and family gatherings and events where pictures are taken. And think about the impact of not being able to see yourself quite literally in those pictures. I can't help but think that not seeing your face and what you look like and what your family looks like and what your friends look like, being lifted up as an example of beauty, really distorted these kids' impressions of what is attractive and what's not, what's good and what's bad. The poor way they are rendered when they are seen, that cumulatively has a really negative effect. Over time, the problems with Kodak's film became clear.

During the 1950s and 60s, as schools in the US began to integrate, black and white kids were increasingly photographed side by side in earbooks. In parents of black kids started noticing the way that their kids didn't look right in those images. But for a long time, Kodak didn't make any changes to the chemistry of their film. When I think about Kodak, I think of Kodak being a company centered in Rochester, New York. I don't think of it as a place that has a lot of black people. That was especially true back in the 1950s.

And I don't think of Kodak being a employer at high levels of decision making that there are a lot of black people present, so that some of those things may have been avoided because someone's been like, and you know, there's a whole spectrum of shades that we need to make sure you can consider. On top of that, there was this attitude among chemists and researchers at Kodak that photography was just science. They didn't think the film itself could be changed to solve any of these problems, they thought. This is not a matter of choices. We can't change film because that's the way it is.

This dictates the way it should be. But Kodak, as a company, had proved it could solve big problems when it wanted to. In the 60s, they helped map the entire surface of the moon for the Apollo 11 landing. And around the same time, Kodak actually did start making some changes to their commercial film as well. But not because they were listening to the complaints from people of color. The story would probably be shorter and end better if they had.

Instead, Kodak changed their film because they were going to lose the business of two big professional clients, a chocolate company and a furniture company. Because dark wood-grained furniture and the lighter version couldn't be differentiated, they all look the same. And milk chocolate, as well as dark chocolate, looked the same. The problem again was the film's dynamic range, the same concept that affects how you photograph skin tones. The chemistry had been very monochromatic in terms of like it would produce a brown, but the brown was the same. It was

like one shade of brown as opposed to like a hundred shades of brown or whatever, you know. According to Earl Cage, the former manager of Kodak Research Studios, the company had never even considered Cage, the former manager of Kodak Research Studios, the company had never even considered how expanding the dynamic range of their film would also improve how dark skin tones are rendered. He noted, quote, it is fascinating that this has never been said before, because it was never black flesh that was addressed as a serious problem that I knew of at the time. Unquote. By the 1980s, Kodak made adjustments to their film emotions and eventually introduced a product called Gold Max. Gold Max was leaps ahead of earlier Kodak films when it came

to color representation and the company advertised it that way, but without actually acknowledging the bias that had been baked into the film before. It was a consumer film that it claimed could photograph a dark horse in low light, which to me was an encoded way of saying we can use this film to create beautiful images of people with darker skins in low light. As Kodak started becoming more aware of racial bias in color imaging, they also introduced a multi-racial Shirley card. It would have a black person, an Asian person, and a white person. Alongside these changes at Kodak, digital photography started to grow. Home photographers transitioned to digital cameras whose dynamic range turned out to be much

better at capturing darker shades. This made photographing a wide range of people easier than it was in the days of film. Digital photography also killed the one-hour photo lab, making the Shirley Cards obsolete. Now home photographers could take and edit photos without the bias built into Kodak's early film and printing processes. But even though the technology of today is way better than the technology of the 1960s and 70s, there is still a lot of cultural inertia preventing dark skin from being photographed in compelling ways.

Photography and cinematography isn't just about the film used or the digital sensor in your camera or the techniques used to process your images. It's also about lighting and staging and all these other elements. In other words, photography is still all about choices and in many ways people still stumble when creating images of darker skin. A common mistake is to overcompensate and make everything too bright. If you look at sitcoms from the 90s, for example, Black Red shows like the Fresh Prince of Bel Air have an overlit, washed out look. That's because the producers are dumping way too much light on the set. Compare that to another show from the era like full house and you'll

see that the approach to lighting dark skin is heavy handed. I think that the super bright look is a result of people just being like, well, I just like we can't see people of color unless it's really bright because they're dark. So we have to add more light.

This is Ava Brakowski, a cinematographer who has worked on TV shows like insecure. You know, I don't even know what to say to that because it's not very nuanced. It's not very nuanced, but neither was anything Brakowski learned in film school about lighting dark skin. Brakowski says their lighting workshops never used a single person of color as a standard model.

And when their instructors did talk about lighting people with darker skin tones, it was always talked about like it was a real headache. Like, oh, that's going to be tough. You know, I'm got sick of seeing pretty white girls in front of the camera all the time. And you're like, oh, if you can make this pretty white girl look like a pretty white girl, then you have succeeded and you're a good cinematographer. And it's just not that's not what happens. That's not what's interesting. With Justice Gordon Parks figured out how to work within the limitations of photography,

there have been pioneering cinematographers who have done the same in TV and film. There's Ernest Dickerson, for example, who shot a lot of Spike Lee's movies, and now there's a whole new generation of cinematographers who are bringing this kind of dexterity to their work, like Kira Kelly and Bradford Young, who shot Soma. So when Brackowski started working on insecure, they studied the work of Dickerson and others to come up with a visual language that worked for the show. Instead of signing lights directly on the actors, creating the overly bright, washed out,

fresh, prince of Belair look, Rukovsky used reflective lighting, which bounces light off of white fabric to create a more luminous glow on the actors. And it works. The show is gorgeous. It's full of beautifully lit scenes. Yeah, this is the first time

that there's nothing in the way of us getting together. So isn't that worth exploring? No. Say, I just got a great course on a beach. Yes. What I would hope is that there's a beauty to the image. And if you shoot people of color, those same way you shoot white people,

you're not gonna find that beauty. You're not gonna find that specificity. Today, it's much less the technology we're working against when it comes to accurate representation of images. It's the users of the technology and the institutions around them that shape the images we see. And of course, there are still problems. Every other month, it feels like there's a new dust up on Twitter as people rail against some terrible magazine cover shot by someone who didn't pay enough attention to skin tone. But even if Codex early promise was, you press a button, we do the rest, the technology

will never achieve point-and-click perfection. Because no technology is ever neutral. There will always be choices and trade-offs and aesthetic judgments. The camera is an amazing tool. But the work of creating a beautiful image, on that part, is up to us. Coming up after the break, we talk about another infamous image and the technological legacy of a Swedish model named Elena after this. So we're back with Delaney Hall, hey Delaney. Hey Roman.

So what do you have for us? So I was the editor on the story we just heard. And as I was working with Kai McNeigh, the reporter on it, I kept thinking about this other story, which I almost pitched our show a few years ago. It turns out it connects to the Shirley Card story in a really uncanny way, and so I wanted to share it with you. Cool, hit me. So the story is about another infamous image that helped create a technology that we use

pretty much every day. And that technology is the JPEG. It's a format for compressing image files. And so, yeah, so practically every photo you've ever taken, every website you've ever visited owes some debt to this image, which is of a Swedish woman named Lena Forsen. Okay. So Lena Forsen. Okay, so Lena Forsen has like the Shirley Page of computer images.

Yeah, yeah, she is the Shirley Page analog in this story. And the first thing I should say is that there have been a number of great stories about Lena and her role in tech history. Notably there was one in Wired by a writer named Linda Kinsler, who actually interviewed Lena and learned about her life story. And the way it goes is that back in 1972, when Lena was 21, she appeared as misnovember in Playboy magazine. It was a nude photo. She was wearing a feathered hat and boots and stockings and a pink boa. Kind of a classic 1972 centerfold image. Yes, exactly. It is very

70s. So a few months after that image was published, a copy of it showed up at the University of Southern California's Signal and Image Processing Institute, which was doing pioneering work at the time in the field of image processing. There was this team of researchers there, and they were developing an algorithm that could make big image files smaller and more manageable, easier to share. And as it turns out, they were looking for a new photograph to use in their tests of this new compression algorithm. And they decided that Lena's centerfold picture would be a good candidate.

I bet they did. I mean, other than the fact that, you know, these dudes wanted a picture of a naked woman to work with all the time. Was there any sort of rationale as to why this image over anything else? I mean, there were some technical arguments and favor of it, you know, it was an image that featured a complex range of colors and textures. It had a human face. It had the texture of feathers.

Like there were things that made it a good test image. I see, I see. But you could imagine picking a picture with a complex range of colors and textures or someone else with a boa around their neck and not using image from Playboy. Yes, yes. You could have found any number of other pictures that were not a naked woman,

but that is not what happened. What they did is they tore off the bottom of the image and then they converted Lena's top half kind of from her shoulders up into a digital image, which they then used in their compression testing. Soon, that image of Lena, looking over her bare shoulder, became an industry standard. It was replicated millions of times as the format that we now know of as the JPEG came into being. Oh, and so, I mean, you stated that it had a lot of, you know, textures and colors and stuff, but is there just a reason this line of image got used over and over again besides

just kind of habit? Well, yeah, I mean, it's hard to imagine now, but back then in the 1970s, scanners were rare. Digitized images were hard to come by. And so the library of images at USC quickly became the standard for imaging scientists around the country. And Lena was one of the main images in that library. There was an image of colorful bell peppers

that got used a lot too, and like a mandrel's face, but Lena was one of the main ones. And then beyond that, the researchers who build and test algorithms actually like to use images that have been used before, because then you can immediately make comparisons to previous iterations of your algorithm or to other algorithms that are out there.

Right, you keep the image the same, so you can create a standard to compare different algorithms against that does make sense. Right. Right. And so because of that, because of those factors, the Lena image became one of the most common images in the field. She was one of the first pictures uploaded to Arpanet, which was the precursor to the internet. She's in thousands of papers. You'll still see her image at conferences today, and she just became iconic. Like some computer scientist just call her the Lena. She just ubiquitous. She can, she's a chair. You can refer to her by her first name, and that's just, you know exactly who you're talking about. Yeah, I mean, but at some point or at any point, was there pushback of this standard image being

a center of fault from Playboy? Yes, absolutely. So over the years, women in the field of computer science, you know, who are still vastly outnumbered by men have talked about how alienating it is to see Lena, for example, projected up on the screen during lectures in college or high school. They've described their mostly male classmates laughing or awgling her, and that's weird for them, as you can imagine.

Yeah, I can't. I mean, it would make me kind of uncomfortable in a classroom to tell you can imagine. Yeah, I can't. I mean, it would make me kind of uncomfortable in a classroom to tell you the truth. And people have pushed back, they have tried to change this. So in particular, there have been some women professors who have protested. There was a math professor at UCLA who, along with a colleague, acquired rights to a headshot of Fabio. And they used that in their imaging research instead of Lena. That ties back to another story we did about romance covers.

But I mean, so you mentioned that a reporter from wire had talked to Lena. I mean, she posed for Playboy, but didn't know this image would be, like co-opted for computer science and JPEG algorithms. I mean, what, you know, how were she of all this and what happened to her? So after her centerfold photo was published, Lena did not do any more pictures for Playboy. She was living in the U.S. at the time.

This is still back in the 1970s. And she got divorced from her then husband. She got a new boyfriend. And together they move to Rochester, New York, which you might remember from the story we just heard is the headquarters of Kodak. That's an interesting connection. Yeah, okay, but the connection is deeper than that, because while in Rochester, Lena got a job working as a Kodak model, and she became one of the companies Shirley's. That's crazy. Isn't it? I could not believe it when I first read about it because working on the Shirley Card

story got me thinking about Lena and I like went back and started reading about her again. And then there's this chapter in her history that just connects exactly to the Shirley card story. She was on those color testing cards that Kodak sent out to printing labs across the country. Oh, that's amazing. So not only was she the digital standard to calibrate things, she became one of the original analog standards to calibrate images too. That's a, what an amazing coincidence. Isn't that wild? Yeah, I love it. And so this whole time when she was in Rochester and computer sciences developing, was she aware of that the whole time that her image was being used in this way for computer imagery or was this something that came like way later? Yeah, I am not sure exactly when she became aware, but I know that over time she has become aware. So she has attended technical conferences, in fact, that wired article I

mentioned described a scene from 2015 when Lena was a special guest at an image processing industry conference in Quebec. And there were photos of the event that show her like stepping on stage through a shimmering projection of her iconic photo. Yeah. Okay. Well, I'm glad that she's like that she owns it in a fun way. If that's what she's into it, yeah, that's great.

Yeah, and you know, the author of that wired article, Linda Kinsler, managed to track Lena down a few years back because while it was clear that Lena knew of the image and had sort of attended some conferences, there wasn't a lot of information about how she felt about it. And so in that article, Lena actually expressed a alarm at the thought that she could be playing a part in herding or discouraging young women, so much so that she actually participated in a recent documentary that came out a couple years ago called Luzine Lena. And that film documents the history of the Lena image,

and then it also calls on researchers to abandon the image and find something new. I'm Lena. I retired from modeling a long time ago. It's time I retired from take two. tired from take two. I imagine other than the sort of sexism inherent in the image that coming up with something new would also be who've, you know, just computer imagery in general just because, you know, a single picture of a white woman, you know, it's just this problematic in terms of creating a kind of narrow standard as it is on the Shirley card.

Right. Exactly. You could imagine a test image with a more diverse array of people on it where you really get a sense that the algorithm is working well across skin tones, for example, in the way that the Shirley Card didn't do at least originally. So yeah, I think there's various arguments for changing the standard in this field. Wow. Well, it's super fascinating, and I can't believe there's actually a Shirley Card connection that she became the digital standard for digital images and analog images as well. That's just stunning to me. It's such an interesting collision of stories. Yeah, well, thank you so much.

It was a great coat. I love it. Okay, thanks, Roman. And we'll include links to that wired article and the Luzin Lennah documentary on our website in case people want to learn more. And that website is 99pi.org. 99% of what was produced this week by Kai McNamey, edited by our Executive Producer,

Delaney Hall, mixed in tech production by Amida Ganatra, fact checking by Francis Card Jr. Music by our director of Sound, Swan Rial. Kurt Colftid is the digital director, special thanks this week to the LA Center of Photography, Tom Allerman and Serita McFadden. The rest of 99 PI team includes Vivian Le, Boshamadon, Christopher Johnson, and Fitzgerald, Chris Baroube, Joe Rosenberg, Sophia Klatsker, and me Roman Mars. We are part of the Stitcher & Serious XM Podcast family, now headquartered six blocks north

in the Pandora building, and beautiful, uptown, Oakland, California. You can find the show and join discussions about the show on Facebook. You can tweet at me at Roman Mars and the show at 99PI org. On Instagram and Reddit too. You can find links to other Stitcher shows I love, as well as every past episode of 99PI at 99PI dot org. .org. Person, woman, man, camera, TV, stichre, Sirius XM.