Discovering the Future of Color – Part One
GAMUT Podcast Episode #69 from Idealliance
(Full transcript below)
In this episode of the GAMUT Podcast from Idealliance, Don Carli, co-founder and President of Nima Hunter eloquently discusses the future of color management by first describing the beginning of innovative solutions like stochastic screening and expanded gamut technology. Don also shares the key findings of the research study, “Discovering a New World of Color and Appearance Management Solutions.”
Don Carli has been a management consultant and senior advisor to advertisers, publishers, and Fortune 1000 brands, including Adobe, Agfa, Dupont, Hewlett Packard, IBM, Kodak, Reed Business Information, Time Incorporated, The Economist, and Xerox. Don is also a Senior Research Fellow with the nonprofit Institute for Sustainable Communication (ISC) where he directs The Sustainable Advertising Partnership and other programs addressing advertising, marketing, corporate responsibility, sustainability, and enterprise communication.
This episode of the GAMUT Podcast is sponsored by:
Reliance on human decision-making for planning and imposition is becoming unsustainable. Tilia Labs is leveraging AI technology to drive a revolution in preproduction, helping printers estimate faster, plan smarter, and manufacture profitably.
Full Episode Transcript
Jeff Collins (00:03):
Here’s the question, how can we serve innovative voices, smart ideas, and the latest technology to improve brand identity, product consistency, and profitability, and the print and packaging supply chain. Welcome to the Idealliance GAMUT podcast. I am your host, Jeff Collins. Idealliance is a non profit association and we serve the global supply chain for brands, print and packaging with 12 offices located around the world. If you’re interested in becoming a member of Idealliance, you can join us by visiting our website: www.Idealliance.org. On today’s GAMUT podcast, we are speaking with Don Carli from Nima Hunter Incorporated. Nima Hunter helps business leaders and marketing executives deploy and develop robust strategies for business transformation and sustainable growth. And Don has been a management consultant and senior advisor to advertisers, publishers and fortune 1000 brands, including Adobe Agfa, DuPont, HP, IBM, Kodak, Reed Business Information, Time Inc, and on and on.
Jeff Collins (01:17):
He is also a senior research fellow with the nonprofit Institute for sustainable communication or ISC. Today we have Don on to talk about a study called “Discovering a new world of color and appearance management solutions.” This study was conducted by Premier, the former research unit of APTech, and they commissioned Nima Hunter to provide its members with credible baseline metrics and projectable consensus forecasts of vendor product opinions, expert opinions, prepress customer opinions, and end-user opinions. One of the opinions is related to the adoption and the use of new color and appearance management software, as well as new hardware.
Jeff Collins (02:05):
From 2017 to 2022, the scope of the study was limited to markets in the US and commercial print and plant packaging and wide format inkjet for color management solutions, ICC profiles, as well as prepress file preparation. So we’re very excited to hear what Don has to say about the future of color management and color measurement for the print industry, as well as outside of the print industry. Don, thanks for joining us on the GAMUT podcast. This will be part one of a two-part series with you. We have a lot of ground to cover talking about the study, doing a deep dive and looking forward to the future of color instrumentation or color measurement devices and technology like ICC max. But I want to share with our listeners something that I find fascinating. And that’s your background in history and the print industry working with solutions like Hi-Fi color, or we know is extended gamut printing as well as stochastic and FM screening.
Jeff Collins (03:08):
So going back to 1991, Don Carli – I’m reading from a blog from John Seymour, his blog, John, the math guy, if guys ever want to read something interesting, check out John’s blog, but made the prediction that about high fidelity methodologies and that it represents an opportunity to account for 50 to 20% of the $150 billion worldwide color printing market by the end of the decade. And you actually coined the phrase. Now, when you told me you coined the phrase, I was a little skeptical because I talked to someone else that, you know, Don Hutchinson and Don was talking about Hi-Fi color back when I first met him in 2006 in the work that he did at DuPont, and then come to find out you coined the phrase Hi-Fi printing. Talk to me about 1991. I mean, that was a totally different landscape. And when we were looking back then, I mean, we had solutions like cytotechs that were digitizing, you know, scan, film, things like that Photoshop really hadn’t taken off yet, although it was there postscript. And so talk to me about that research and why you believed that high-fi printing was a thing of the future.
Don Carli (04:38):
Sure. Well, to begin with my background is a little bit unusual in that I actually understand color and printing kind of from a first principles perspective. In college, I did study art as well as English literature and upon graduation, I took an unusual career course and apprenticed to a master printer who at the time was working on limited edition prints using the full spectrum of printing processes, available to the art world for very famous artists like Joseph Albers, the former chair of the art department at Yale university and the author of a number of books on color.
Don Carli (05:34):
So having worked with Joseph Albers and other artists I was never limited during that period, the printing processes of commercial printing, we employed lithography etching and engraving screen printing woodcuts. We made our own paper. We formulated our own inks because our inks had to have light fastness for a hundred years because we were creating original artworks using 15, 20, or more colors on multiple printing processes with not just color, but surface effects in the early 1970s. Having come from that background and working with artists who had something in mind and worked with a collaborator in the studio, Tyler Graphics, where I was an apprentice, and then ultimately became a master printer. Our job was to give life to the ideas that lived in an artist’s head. So at the time in 1991, people were still talking about WYSIWYG color. What you see is what you get, right?
Don Carli (06:50):
And that was actually somewhat of a response to the compromises that started to be made when desktop color emerged. I had always operated on a different frame, which was what I called wiziwig color. What you see is what you had in mind. And coming from that background, I was responsible not just for creating these original additions for artists who would come to me and say, in the case of the artist, Robert Motherwell, he would say, Don, I want to read with the redness of red. You didn’t give me a Pantone color. And he said, I want to read with the redness of red, a phenomenological red. And my that’s not very objective, is it? Well, it was totally subjective, but my role was to understand the available technology and to manifest that for the artist until they looked at it. And kind of in the case of the greatest artists in the world, who I had the privilege of working with, when they see it, they know it they’re unequivocal and it would be, yes, that’s it in working with Joseph Albers once he said, Don, I want a gray that’s like the color of grape leaves in arbor with the Mediterranean Twilight was a dusting of copper sulfate on them.
Don Carli (08:14):
That was my color specification. And for over a month I spent mixing grays for Joseph Albers, just grays. What type of reprinting I was using the screenings at the time we were producing a series of prints for Albers called gray interactions. And Albers was one of the first artists who really looked at how color interacts with color, an important issue. We know this in the graphic arts, when you put one color next to another color, Albers would say, you get a third color – for free. And it’s at the interface of these colors that we see effects: simultaneous contrast, for example, is a color interaction effect. So my background comes from an understanding of the art of color and color as explored by some of the world’s great artists in that field like Joseph Albers and Robert Motherwell, Frank Stella, Roy Lichtenstein, etc.
Don Carli (09:19):
But I also was responsible for creating the reproductions of those prints that we produced in four-color process. So then I had to understand how to make the necessary compromises in the photography and in the color separation processes, which in those days were still being done in many cases with indirect screen and cameras, not with scanners, right. Ultimately I left the world of fine art printmaking and I became the Technical Director of a commercial printing company in New York City. That was about a $15 million sheet fed printer that specialized in annual reports and extremely PI and graphic design production for some of the finest graphic design firms in the world, Push Pin Studios and so on. So I had to then understand how to translate an incredible color gamut with all kinds of surface effects into the best possible reproduction using four-color process. But the printing company that I joined didn’t buy four color presses. They were the first company in us to buy six color presses. So now with my background, I began to explore how we might adapt those commercial printing presses to take advantage of some of the things I understood from the world of fine art printmaking, where we did regularly produce prints with more than four colors. And after I left Tyler Graphics I then became the technical director of another studio that produced 15 and 20 color reproductions of Norman Rockwell paintings.
Jeff Collins (11:08):
And just to pause and clarify for a moment when we think of using supplemental colors to expand gamut that’s typically what we use it for to capture extra Pantone colors. At least that’s the way it is today, but you’re talking about Norman Rockwell. And my frame of reference is a Saturday evening post. And why would you use supplemental colors for something like that? It doesn’t seem to really have those outer range gamut colors that we would use, you know, six and seven color process to reproduce.
Don Carli (11:42):
So gamuts are only one aspect of it and this is an important issue. We were also looking for the nuances of internal reflections and surface. And in many cases transitions where we know this from printing, tones and tri tones that if you have a gamut or not a gamut, but a dynamic range, it many cases cannot be mapped. Using eight bits of gray, you may need 16 bits of gray. Well, if your printing process only allows you in one deposition of, of ink to achieve a certain dynamic range, you can turn it into two plates and increase the dynamic range effectively spreading it over a larger number of bits. So we did the same kind of thing. And in many cases you might have flesh tones, or you might have nuanced specific object colors that we were really trying to reproduce, not just in terms of their gamut, but also in terms of metabolism, right?
Don Carli (12:52):
We were looking for non-mid America color matching in many cases or we were looking for subtle transitions that didn’t quantize, that didn’t create breaks or compression of the dynamic range in ways that caused, seen failure caused the picture to fail, to convince you, to suspend your willing disbelief. So my background is unusual in that I’ve created original additions for artists using first principles using offset with photography, stone, lithography, etching, engraving, screen printing, embossing to create these incredible objects, even making your own paper in many cases. And they’re in museums around the world to then working in the reproduction of those for the galleries and the museums that were going to, and the auction houses that we’re going to offer those for sale and had to present the best possible representation of that original artwork as they could. And then to the world of commercial printing, where we were producing just annual reports and commercial printing,
Jeff Collins (14:00):
I don’t want to stop your stream of conscious. This is amazing. I did want to stop there with something that just jumps out at me. And you were talking about mixing the different screenings. Okay. Were you using a wider range of colors, a larger palette, and when you applied essentially the same principles to, let’s say a six color press or seven color press early on where you’re combining, let’s say an orange and a violet on top of each other, or maybe two other supplemental colors? And did you, could you effectively predict based on your experience with mixing screenings, did you get the same results using the lithium tanks?
Don Carli (14:52):
Well, you know, again, it’s a matter of whether the inks are transparent or whether they’re opaque or translucent, obviously. Ink is ink and it’s going to absorb and, or reflect light. It’s going to interact with light and it’s surface and it’s an interiors, and it has relationships with the substrate as well in terms of the substrates surface characteristics and absorption so on. But yeah, for me, the most important thing was to realize that there were always going to be designers, artists, creatives, who had something in mind that the four color process gamut and dynamic range, limitations and detail limitations were going to disappoint. There was always going to be a market four approaches to placing colorants and surface treatments on substrates that we’re going to have an audience that was willing to pay more for something that did not disappoint them, that rendered what they had in mind.
Don Carli (15:57):
The important aspect for many of those people was that the rendering had to look the same under different lighting conditions and different viewing conditions. Metabolism was a concern for them to solve the problem. Absolutely we did then use analog methods, the revelation having, you know, moved from the world of fine art printing to the world of commercial printing for me was I went, luckily, went to work for a company that was owned by the Sonoma Corporation in Finland, which at the time in the early eighties was a $500 million company based in Helsinki that had the largest installation of cytotechs systems and health scanners and SII, editorial coyote terminals in the world. It was one of the most advanced and still is one of the most advanced communications printing publishing companies in the world. And I became the Head of Research and Technology and business planning for that company’s subsidiary in New York and our position at the time was to bring all of that advanced technology that was already in place in the early eighties, in Finland to New York.
Don Carli (17:18):
And so my job upon arriving in New York was to implement a full scale transformation of all of the prepress and printing infrastructure in that company. It was a $15 million capital improvement program. And that was my, my role was to plan it and implement it. And that was just at the time when electronic prepress was reaching its apex of capability with Crossfield. And cytotechs where we had, you know, scanners a quarter of a million dollars and prepress systems cost a millions, and you could sell an hour of retouching time, quote unquote retouching time on a Crossfield or cytotechs workstation for $500 an hour to an ad agency. But we were also just at the cusp of the desktop publishing revolution. We were just at the cusp of the advent of PCs from Apple and IBM. And if you, if you think of it, we also were in the face of one of the most significant economic downturns in history in 1989.
Don Carli (18:26):
And so at that, at that inflection point, there were a number of innovations that the lineups scan 300image center the Barney scan, color, desktop scanner and Apple computers that started to introduce applications that could manipulate pages like all this page maker and then eventually Photoshop. There were conferences coming up saying, well, we’re going to now adopt desktop color and color is going to be done on the desktop and it’s going to be good enough. What you see is what you’ll get. It’ll be good enough, but it’ll be inexpensive and democratic. And my counter to that, having run not only studios that produced incredibly expensive, limited additions that would sell for $20-30,000 a piece per print to then running a commercial printing company, producing exquisite annual reports and posters for some of the finest design firms and ad agencies in the world was that there was always going to be a market.
Don Carli (19:36):
And I estimated it was 15 to 20% of the market that would pay more for a visually differentiated product that rendered what they had in mind to faithfully. Now, the challenge was to turn that from being an art form, which is what I knew into a science and engineering discipline to do that we would need algorithmic means of converting red, green, and blue data into not just cyan, magenta, yellow, black data. But what I knew from my background as a master printer into pallets that might consist of six or seven colors, right? Same as each yellow, green, orange, violet, and black to begin with based on the Cooper’s color model. These models did exist and they were all employed by artists and fine art print makers like myself, but didn’t have any basis in the world of software. And again, computational capability was limited at the time.
Don Carli (20:37):
So my response to those who said all of the high-end processes that we’ve employed are going to be lost. I said, well, perhaps some of the technology will, but there’s new technology. There’s new computational capability coming online. Moore’s law is going to bring us the ability to calculate incredibly complex algorithms more and more cost-effectively more quickly than we can imagine. And so I simply said, instead of just throwing your hands up in despair and saying all color’s going to become mediocre, I said, some color will become mediocre just as some typographic became mediocre when desktop publishing brought fonts to the masses, but there will always, I said, be a market for processes that can be engineered to render intent, to render the visual intent of a creator in a faithful manner. And by that I meant not just the color gamut, but also the nuances of managing and representing the dynamic range of an image as well as the detail and structure of an image.
Don Carli (21:50):
And so at that point, having drawing on my background, I was at a conference where one of the consultants in the industry at the time was complaining about how it was all going to change and go, you know, go to mediocre color effectively. And I said, well, it doesn’t have to be. And he said, so explain it to me. And I gave my explanation of my background and he said, so could you explain how this would work? And we could convince companies to support this prototype, how it would work. And this was in the face of the 1989 market crash. And I said, well, sure. So we wound up getting support from about 20 companies in the industry, Adobe and Kodak and paper companies and printing companies like Anderson, lithograph and LA. And we prototyped how one could print using a process and software images that were scanned as RGB, but then converted into cyan, magenta, yellow, green, orange, violet, and black, right. And to deal with things like color angle. I said I would call that high fidelity color. Right, but I would also say we need to deal with some of the screening issues. So I coined another term at the time, which made some people laugh – stochastic screening,
Jeff Collins (23:17):
Sorry, stochastic screening is what you said, right.
Don Carli (23:20):
Well sarcastic screening is what Frank Romano called it, but I did in fact, call it stochastic.
Jeff Collins (23:27):
Don Carli (23:30):
From the Greek Word, stochastics, meaning to take aim. My argument was at the time we had the technology in the military industrial complex to calculate the ballistics of multiple objects, warheads through turbulent atmospheres that could be targeted to land within a few meters of their intended spot. And I said, how is that any different than determining where you’re going to land a laser spot to create a dot or now inkjet drop or at the time inkjet as well. But the point was, these are ballistic problems and we can solve them when computational capabilities become more robust than they are. And they will. So at the time it was forward-looking, it was speculative. It was intended to give some hope to an industry that you could develop a premium solution to specific problems. But also interestingly, it offered tremendous opportunity at the low end of the market, where you had limited resolution and where the combination of stochastic screening and hi fidelity color could actually significantly improve flexography because you could create higher apparent resolution plastic screening.
Don Carli (24:53):
You could reduce the introduction of various forms of Moray, Subject Moray, and Object Moray, and you could also reduce metamerism properly employed. It was not necessarily adopted as I had intended it in every case. And a lot of the complaints that people might have had were often based on an incomplete understanding of what had been articulated originally, but that’s the way things go. People take we’ll take things and run with them. It’s interesting that I would often say, well, if you print it 300 line screen with a conventional screen, there’s little advantage of stochastic of using stochastic screening, if you don’t have to use it on all of the colors. You could just take one palette like the yellow palette and use stochastic screening for that to relieve you of the angle constraints and then continue on with conventional screening for the rest.
Don Carli (25:51):
So that’s where it started as my background as a master printer. And then as the technical director of a very advanced printing company that owned picture agencies and ad agencies produced magazines, newspapers owned a cable television network, had a venture capital arm. All of these things were kind of an input to my formulation of color separation and color rendering and reproduction approaches that at the time were not common. And I also spent a lot of time at conferences like Siebold and the tone color conferences, trying to convince the creators of the first ICC standards that they shouldn’t limit themselves to four color process, right. That they should consider more than four primaries profiles and manage. I also said they shouldn’t limit themselves to eight bits because eight bits is not enough. I did do a number of studies in the early nineties for at the time, the NPS, and then later premiere on color management and color proofing.
Don Carli (27:11):
My background aside from being a master printer is in market research. So by and large, what I’ve been doing for the past 30 years is qualitative and quantitative research. I spent about a decade at Xerox running business intelligence and market research for the graphic arts industry business unit at Xerox. In that role, I had resources at my disposal and, you know, the ability to conduct very extensive, qualitative and quantitative research to size markets, to segment markets, to identify and characterize the needs of markets, to articulate the benefits that were going to be presented by the brand to satisfy those needs and to help by using techniques like discrete conjoint analysis and so on the product teams to prioritize the features of products that they would configure to address those benefits in order of priority. So I’ve got an unusual background in that. Unlike most market research professionals, I’ve actually run a production operation manufacturing operation. I’ve been responsible for business planning and calculating that present value and internal rate of return on multimillion dollar investments. I have this other dimension of my background in working with artists and gallerists and museums, auction houses in the Reaper to production and reproduction of exquisite works of art. That kind of proceeded this most recent project
Jeff Collins (28:55):
Don, before we begin talking about that recent project in this study that I mentioned in the intro, I’d like to take a moment to recognize our newest sponsor Tilia Labs. If your company or print operation is struggling to automate imposition and planning, and you’re tired of the building imposition templates for every single job Tilia Labs provides artificially intelligent planning and imposition solutions at your fingertips with their award-winning software, Tilia Phoenix, you can quickly run cost base analysis to plan and impose your customer’s orders with Tilia Phoenix. You can easily integrate with MIS and workflow solutions using industry standard JDF XML, or state-of-the-art restful APIs for increased automation. Whether you’re printing books, brochures, cartons, labels, or signs, Phoenix is imposition AI can nest dynamically gang and calculate the most cost efficient production plan without ever creating a single template. To learn how Tilia Labs is modernizing the print and packaging software with smart AI driven technology, just visit Tilia labs at www.Tilialabs.com that’s TiliaLabs.com.
Jeff Collins (30:20):
Through the end of the year Tilia labs is offering a free 30 day trial of Tilia Phoenix. And you can request your free trial today by visiting TiliaLabs.com/gamut that’s TiliaLabs.com/gamut for a free 30 day trial of Tilia Phoenix. Don, to pick up where we left off, you were discussing a variety of different solutions that solve problems and limitations for not only print manufacturers, but also the customer using stochastic screening, as well as supplemental inks. And of course, we currently have ICC color management, and that also has its limitations. And really that set the heart of the study that you recently conducted.
Don Carli (31:06):
There were certain problems that those technological prescriptions could solve. And there were certain problems that the technical prescriptions of ICC colorimetric color management up to version four could and can solve, but there are other characteristics of human, visual response and human perception that the ICC version for colorimetric approaches will never be able to address structurally and capable of addressing. The fundamental issue is described as it looked the same before, but it doesn’t look the same now. My socks don’t match my pants now that I’ve walked outside, they’ve matched when I was at home. So that problem metamerism is one of the fundamental problems. Metamerism is not a well understood set of phenomenon. It’s not one thing it’s in fact, many things it’s caused by many things, the root and the root description of it is two things that match under one set of conditions don’t match under another set of conditions.
Don Carli (32:25):
The conditions could be the viewing angle. It could be the light source. It could be the observer, it could be motion. It could be any factor that when introduced the perceptual frame changes the perception and ultimately perception is reality for the observer. So to the degree, the observer is the buyer. And to the degree, the observers expectation is not met. That the thing should look a certain way, that the person who buys a couch online and sees the color being what they believe is going to match their drapes upon arriving at home, find out well, when the lights are on in my home, it looks okay, but when the windows are open and the shades are drawn, it doesn’t, I’m not happy. I wanted it to look the same. These are the problems that current color management using colorimetric approaches can’t solve, never will.
Jeff Collins (33:43):
And so the solution is spectral, correct?
Don Carli (33:47):
Well, it’s a combination of spectral and specular measurement.
Jeff Collins (33:54):
Don for myself and our listeners, what are the differences between those two terms, specular and spectrum?
Don Carli (34:01):
We know what gloss is when something is glossy light interacts with the surface differently than when it’s dull. And the measurement of those reflections using is done using a device called a goniometer or a gloss meter, more commonly. So we’ve had the ability to measure gloss and in many packaging and printing companies today, they have gloss meters that are used to measure gloss, correct? That’s specular data. The data that defines the way in which light bounces at, at given angles from a surface, or is diffused by a surface. Spectral data on the other hand is a measurement of the wavelengths of light and light energy. Today all of the color management approaches that are dominant while they may use a spectrophotometer, they ultimately convert the spectral data into some tristimulus value, red, green, blue, L a B L U V a three component tri stimulus value. And all of the calculations are done in that three value space.
Don Carli (35:25):
Unfortunately, the representation of where reproduction of a color can have many possible solutions. And because of that, it matches in that three color definition, but it may not match if the lighting composition, the composition of spectral energy in the lighting changes or the viewing angle changes, or if the substrate changes. So a whole array of factors can, or the absorption characteristics of the inks are different. It will match color metrically, but it will not match spectrally. And we’ve known this in packaging for years, if you were a printer and you went to an into a company saying, I want to match color, you could define it as either a spectral match, a non-metameric match or a metameric match. And the nonmetameric match would not change its appearance under different lighting conditions, the metameric match would.
Jeff Collins (36:36):
And what impact does that differentiation between let’s just say the ink formula impact the price of the raw material.
Don Carli (36:45):
It can impact it, obviously. An example would be, you know, in many cases, a printer might take waste ink from the previous job and have it used to blend a gray or a black blended into the ink. Now that could change the metaphoric index of that ink understood. Right? So we often make trade-offs trade-offs are okay. And for much of the world, that’s concerned with four color process, existing ICC V for color management works, especially when employed in combination with color process, with process control mechanisms, like the G7 protocols, it works! It’s been adopted by a significant proportion of the commercial printing industry to good effect. It took a long time. It took a decade or more, but nonetheless it has been adopted by about half of the universe. That’s the good news. The bad news is there’s still a fairly significant number of applications for which it will be insufficient to solve the problems that they confront, that it won’t be able to address the challenges faced by applications in packaging and labels, industrial printing, textile printing, and other arenas where metabolism, for example, is a major concern or where appearance, characteristics that are not specifically about color, but about attributes like fluorescents or pearlescents, or what is called gonio apparents that’s when the color of something changes based on the viewing.
Don Carli (38:48):
Yeah. So the purpose of the study was, first of all, the ICC and the association for print technology OEM vendors and print service providers wanted a baseline. They wanted to understand what some of the trends and issues and challenges and opportunities were likely to be for both the vendors and the print service providers related to this fundamentally new approach to managing color and appearance in a standards-based way. We’ve had proprietary approaches to managing appearance for some time. For example, Pantone has its total appearance capture system, which is a proprietary approach to managing appearance, but industry has for a long time benefited and relied upon standards-based approaches that provide for interoperability that provide for lower switching costs. When one hopes to move from one platform to another also provides flexibility and configuring solutions that might employ component technologies from multiple vendors, as opposed to being locked in to a single proprietary solution.
Don Carli (40:19):
So with a background in supporting de jure standards like ISO and ancy the association wanted to understand what the outlook was for the adoption of a fundamentally new standard based on ISO. An ISO standard for color and appearance management, which is termed ICC Max. But the actual standard is ISO 20677. So it is a standard. It has been, it is an official ISO standard, which, which means vendors can now adopt it and create products based on the standard. There was only at the time of the study, one product that embodied the standard, that was the Onyx rip one of the Onyx rips, but you can’t create a marketplace with just one product. You need multiple products, you need complimentary products for the different aspects of the value chain. And so the study was really to help the vendors on the one hand, understand what the potential market was, where there were pain points and frustrations with current ICC solutions that presented an opportunity for new products to be developed based on the new standard, but also to understand from the PSP side where there were applications of those technologies. Whether it was packaging labels, industrial printing, etc, that they could pursue to create new revenue, new opportunities for their businesses as print service providers by developing solutions applications that addressed the needs of brands more effectively.
Don Carli (42:07):
So it was undertaken to create a baseline. Now, whenever you face a arena, a market where you don’t have prior data, you can’t really project from the past. You can’t unfortunately turn to the market initially and ask them about something. They have no awareness of.
Don Carli (42:39):
So we started this by doing a review of the literature of what had taken place under up to ICC version four. And then we recruited a panel of experts using a methodology that was developed during the cold war, by the Rand Corporation called the Delphi Method. And that group of 50 experts who were color scientists and color management professionals, including yourself, were asked to make forecasts based on their experience. Now this approach of using experts has a lot of background in other fields. I’ve been using it in my work for the past 20 years. And when I was at Xerox, I had the resources to do this regularly. So we asked the experts to make forecasts and to make and have the experts arrive at a consensus of opinions. And it’s not about right answer, wrong answer.
Don Carli (43:46):
It’s ultimately all answers – it’s opinion research. So any answer is a good answer. The question is what was the consensus of experts, right? And then we use that set of consensus opinions, as well as the explanations for why they held them or disagreed with them as the basis for a quantitative study that used a combination of machine learning and natural language processing to not only ask closed-end questions, which are framed to minimize bias no double-barreled questions, you know, no leading questions, but, but also after someone answers the closed end question, the platform that the application asks them to explain why they made that statement in their own words, in real time, that response is fed to a natural language processing engine, which parses it semantically and looks at the Corpus of statements that have been made by others and presents back to the person who made that statement.
Don Carli (44:56):
Another statement that could be conceived as similar and asks them do these two things mean the same thing to you? Yes or no. I remember if, so that’s a process called disambiguation. Now, typically in your kind of the desktop color equivalent, the desktop color equivalent in research is, is using tools, common tools I won’t name a brand name, but you know, there are some, some things that hang from trees at times and eat bananas who will be used to make surveys. And they will ask an open-ended question and you fill in the blank, those open-ended questions then have to be reviewed by a human being, one human being, typically the survey administrator, and that individual has to make decisions about which of the statements are similar or dissimilar. And then they have to create a summary of what those statements say. Those open-ended statements also have no correlation with the closed end responses that were given in this approach. It’s a little bit different because we’re asking for the entire audience, a sample of the audience to disambiguate the answer. So it’s not done by one expert. It’s done by the whole crowd. And it takes advantage of a phenomenon called the symmetry of ignorance.
Don Carli (46:26):
And in emerging areas, we have an asymmetry of knowledge. Some people know more than others, and everybody has a different way of seeing the same thing, but everybody is no one person has all the answers. So we have asymmetry of knowledge and a symmetry of ignorance. And by creating these kinds of Delphi studies and tools using crowdsourcing of information with the right controls over bias, you can create a composite view of reality that is more accurate and more representative of the population than any particular view. And you can arrive at either a consensus or at leastwhat would be called acceptable compromises as to what the future is likely to be and why it’s likely to be. So that was the role of this study. It was trying to get a consensus of experts and then of the market in general. We did a survey of over several hundred print service providers, creative professionals, and brand owners to see if what the experts said the future was going to hold was what they believed the future would hold. So we calibrated the experts with the actual market, and we use the hypotheses of the experts as the basis for closed end questions that were presented to the market as a whole.
Jeff Collins (47:50):
And how close were the two audiences aligned the brands compared to the experts?
Don Carli (47:57):
Yeah, you know, as a general rule, I’ve been doing this for 20 some odd years, experts do not accurately represent the market. They tend to lead the market. They tend to be optimistic for a variety of reasons. And so there’s somewhat of a sobering effect when you present the audience with what experts say is going to happen in the end that the audience says not quite so much, but at least we had a reasonable measure of that. And we were able to get not only the levels of satisfaction that users in the market have today, but how likely they believe the, these current solutions will need their needs going forward. What we found was, you know, 65% of the respondents in the quantitative study, and that was about 200 print service providers and brand owners and users of color management said they were either very satisfied, somewhat satisfied with today’s color management solutions. Interesting. That’s, you know, only 6% of the respondents were either very or extremely dissatisfied.
Don Carli (49:17):
Well, you know, what’s interesting if you look at it, we had approximately 18% of the market said they were dissatisfied to some degree. And that kind of maps back to the rule of thumb 80%/20% rule that I used in my Hi-Fi color study in 1980 and 1991, rather, which was there’s 80%, somewhat satisfied or neither satisfied or dissatisfied. And there’s always going to be this long tail, this 20% that are dissatisfied to some degree with what’s available and want more we’re different. So now the question is, will it always be that 20% or could it become 50% or 60%? And the reality is it varies based on the application domain. So when we asked those questions in the aggregate it’s 20%, but if you started asking that question specifically to people who were doing textile printing, the distribution is somewhat different because in textiles, the color observer standard and lighting standard is not the 52 degree D 65, 10 degree.
Jeff Collins (50:40):
So would you say there’s more dissatisfaction on the textiles side?
Don Carli (50:45):
Frustration, pain, yes. And cost.
Jeff Collins (50:48):
And, and is that due to the current technology being utilized in the digital realm of textiles or the analog realm of textiles?
Don Carli (50:55):
Yes. I get back to the root definition of what a standard observer is in the textile world. The prevailing standard observer is a D 65, 10 degree observer, not a D 52 degree observer. And that might be somewhat technical and trivial to some people, but it matters tremendously because you cannot use current ICC V4 to effect the transformations you want to maintain color fidelity from those two observer constraints, you can do it, but it requires a hack. You can’t use the current standard. You have to hack around it. It’s, you know, there are some companies that have figured out proprietary ways to do it in what they call happy hacks, but happy hacks are not a standard. So the real issue was do we go forward with standards based solutions that can solve the problems of packaging labels, industrial printing, 3D printing, but also the growing use of 3d modeling and rendering using physically based rendering in e-commerce in order to do that, we need to move beyond colorimetric color matching to spectral and specular color matching using appearance models,
Jeff Collins (52:28):
Ladies and gentlemen, that concludes today’s GAMUT podcast. And this is again part one of our two part discussion. So make sure you stay tuned for the second half of Don’s discussion on ICC Macs and the future of color management. And if you’re interested in the study that we’re talking about today, you can get that from AP tech or the association for print technologies, just visit their website. The study is called discovering a new world of color and appearance management solutions. Thanks for listening to the gamut podcast. If you have ideas, suggestions, or would like to join us or even sponsor future podcasts, email firstname.lastname@example.org. Take care and have a productive day.