Daniel Rosenberg
Boardroom Battles & Semantic Interaction Design
In this episode of Brave UX, Daniel Rosenberg speaks frankly about executive design leadership, why he’s trying to take IxD back to first principles, and how the sun is setting on design as we know it.
Highlights include:
- Which Jedi would Don Norman say is most like you?
- How did your first presentation to Larry Ellison at Oracle go?
- Why is design education being overly intuitive a problem?
- What is Semantic IxD and why is it important?
- How do you remove enemies while maintaining moral authority?
Who is Daniel Rosenberg?
Daniel is the Chief Design Officer at r-CDO and during his 40+ year career, he has forged a near unparalleled path as a world-leading and highly respected UX practitioner, global design executive, consultant, author and educator.
Recently, Daniel has been working at the intersection of AI and interaction design, advising a number exciting startups, including WellDoc, CanSurround, Workato and Kleeen Software.
Before starting r-CDO, Daniel was the top UX executive at Oracle where he served as VP of UI Design, and then at SAP, where he was SVP of Global User Experience.
Prior to joining Oracle, while he was at Ashton-Tate and then Borland, Daniel invented several design patterns that are still popular in product design today, including tabbed UI.
Daniel’s latest book, UX Magic, was published in 2020; a groundbreaking book that lays out a complete system for Semantic Interaction Design.
Transcript
- Brendan Jarvis:
- Hello and welcome to another episode of Brave UX. I'm Brendan Jarvis, Managing Founder of The Space InBetween, the home of New Zealand's only specialist evaluative UX research practice and world-class UX lab, enabling brave teams across the globe to de-risk product design and equally brave leaders to shape and scale design culture. Here on Brave UX though, it's my job to help you to put the pieces of the product puzzle together. I do that by unpacking the stories, learnings, and expert advice of world-class UX, design and product management professionals. My guest today is Daniel Rosenberg. Daniel is the Chief Design Officer at rCDO, the UX strategy and product design consulting firm he started in 2012. Through rCDO, Daniel mentors CEOs, CPOs, development VPs and UX leaders, helping them to optimize the alignment between their UX investment and their company's business plans. Most recently, Daniel has been working at the intersection of AI and interaction design, advising a number of exciting startups including WellDoc, CanSurround, Workato and Kleeen Software.
- During his 40 plus year career, Daniel has forged a near unparalleled path as a world leading and highly respected UX practitioner, global design, executive consultant, author and educator. Before starting rCDO, Daniel was the top UX executive at Oracle where he served as VP of UI Design and then at SAP where he was SVP of Global User Experience. For 18 years, across both companies and all of their product lines, Daniel led their global UX efforts. Prior to joining Oracle, while he was at Ashton-Tate and then Borlamd, Daniel invented several design patterns that are still, yes, still popular in product design today, including tabbed UI. And yes, he does have the patent to prove it. Daniel is also helping to level up the skills of future UXers through his work as an Adjunct Professor at San Jose State University, where he teaches postgraduate classes and interaction design. A truly generous contributor to the field,
- Daniel is also an advisory board member of the Interaction Design Foundation and for UC Irvine's Masters in Interaction Design. Along with some people you may know, Don Norman, Jakob Nielsen, Gordon Bell, and Brand Foram, he is a Founding Member of the editorial advisory board for the ACM Networker Magazine. Daniel is also the author of dozens of publications on UX and has been a contributing author to three foundational books, including the very first edition of The Handbook of Human Computer Interaction, published in 1988, where he wrote the chapter on rapid prototyping. Perhaps his most notable contribution to the field has been co-authoring Human Factors in Product Design with William Kushman, published in 1991. It was the first book of its kind to address the application of human factors, principles to consumer products. In 2019 Daniel's tireless contributions to the field were recognized when he was awarded the ACM SIGCHI Lifetime Practice Award.
- However, if Daniel has his way, it's his latest book, UX Magic, that he hopes will be his greatest contribution, at least I suspect. Published in 2020, UX Magic is a groundbreaking book that lays out a complete system for semantic interaction design. More on that soon. A sought after speaker, Daniel has recently delivered keynotes at Salesforce's Design Days, for the Design Management Institute, the Human Factors in Ergonomic Society, Rosenfeld Media, the CHI Conference, and the IxDA's Interaction Conference. From what I can tell, Daniel is truly a titan of the field and it's an absolute pleasure to have him here with me on Brave UX today. Daniel, welcome to the show.
- Daniel Rosenberg:
- Thank you for inviting me.
- Brendan Jarvis:
- Thank you for very patiently sitting through your introduction as
- Daniel Rosenberg:
- One detailed note for you. Mm-hmm. [affirmative] Networker Magazine doesn't exist anymore.
- Brendan Jarvis:
- Oh, it it's met its end has it?
- Daniel Rosenberg:
- What you said is true. Don and I started it, it was just on a call with Don this morning about something else but.com era thing and it lasted about four or five years and then got nuked. It immediately disappeared when the.com era ended.
- Brendan Jarvis:
- Right? Was it the last thing on people's to-do list while the world was burning around them?
- Daniel Rosenberg:
- Yeah, it was fun, but it was not a long lived. I think we ran for about five years.
- Brendan Jarvis:
- Well you, you've certainly run for a lot more than five years. So Daniel, I've listened to a number of your previous talks preparing for today, and I feel this is a safe place to start cuz you've mentioned Jedi at least twice in those talks. I just wanted to very briefly get your thoughts on the renaissance of Star Wars
- Daniel Rosenberg:
- As a movie series?
- Brendan Jarvis:
- What Disney's been doing with it?
- Daniel Rosenberg:
- Honestly, not that impressed. No, but it works as a metaphor because it was so universally accepted that people get the idea of being able to practice it at a really, really superior level that both in terms of skill and both in terms of the kindness and the intent, the opportunities to do good.
- Brendan Jarvis:
- And if I was sitting here with Don Norman, who you just mentioned, and he had to tell me which Jedi best represented you, who would he say
- Daniel Rosenberg:
- Yoda.
- Brendan Jarvis:
- Yoda. I love it. And why would he say Yoda?
- Daniel Rosenberg:
- I'm not really sure, but that became the label actually in signature at SAP for a while because of the email signature, because of the amount of theology and philosophy involved. And when I would make this the Jedi reference and then since I'm Jewish, they would say, well, you should just be Rabbi Rosenberg [laugh]. And we've settled on Rev Yoda, and so I had that as my email signature for seven years.
- Brendan Jarvis:
- So it's got a bit of history.
- Daniel Rosenberg:
- Try to take yourself too seriously.
- Brendan Jarvis:
- Yeah, that's an important lesson that I think more people should internalize. We'd have more fun as a world for sure. Now, you mentioned that you were Jewish and that is something that I also mean clearly by your last name but also something that you've touched on briefly in some of the content that I've watched before today. Now I understand that you grew up in an Orthodox Jewish family in America's Bible belt, which unfortunately sounds like a recipe for trouble quite frankly. What brought your family to the Bible bell?
- Daniel Rosenberg:
- My father became a professor at the University of Nebraska.
- Brendan Jarvis:
- And what was he? A professor of
- Daniel Rosenberg:
- Agricultural meteorology. My father wrote the book, microclimate in the 1960s, had a long, very successful scientific career.
- Brendan Jarvis:
- Maybe I'm reading something in here that isn't here, but it sounds like he was quite a strong role model for you.
- Daniel Rosenberg:
- I think that's true in terms of accomplishment.
- Brendan Jarvis:
- Yeah, I understand you were kicked off the high school football team for missing practice on Rosh.
- Daniel Rosenberg:
- That is correct. But you've done really deep research.
- Brendan Jarvis:
- Well, I mean this is something like I said, you mentioned in passing, but I mean,
- Daniel Rosenberg:
- Did I say, I really said that on a recorded session
- Brendan Jarvis:
- You did, and I pick up on these sorts of things because obviously someone have listened to your intro, will know that you've achieved too the highest levels. And this is memory of being kicked off. That football team in high school seems to be something that has stayed with you. And I was curious about these challenges that people face all of us face, but this is clearly one of the challenges you have faced with an anti-Semitic approach that you've run into being in those environments. What role did living and growing up in the Bible belt and experiencing those anti-Semitic beha sematic behaviors play in drawing you to the West coast to computers, to Silicon Valley and out of that middle part of America?
- Daniel Rosenberg:
- It definitely played a role. I'd say it was accidental in some ways. In graduating from high school, I definitely wanted to leave Nebraska. I drew a 500 mile radius and said I would go to any college that accepted me that was outside of that circle. And of the colleges that accepted me, I chose to go to Tufts University in Boston to study civil engineering. This was before computers had screens basically for all intents and purposes. And I serendipitously discovered that Tufts had a human factors engine engineering program, one of the first in the country and took some classes in that. And it obviously put me on the vector in this direction and with very good mentors, very special people for whom I'm still grateful. Professor John Krefeld retired 20 years ago and still alive. And Professor Percy Holmes Hill, the who passed away quite a while ago. But that got me started and it wasn't about software originally, it was just human factors.
- Brendan Jarvis:
- Talk to me about that distinction because there are a lot of people in user experience now, probably people from my age down to people that are in their, I suppose early twenties. So I'm in my late thirties down to people in their early twenties who don't necessarily completely grasp that distinction between the two.
- Daniel Rosenberg:
- Well, the origin of human factors engineering was mostly military and particularly cockpit design, but World War II to keep pilots from crashing planes by moving levers forward when they should have moved them backwards. And so the original focus was more on physical ergonomics, control system design. And in those days airplanes didn't have CRTs in them or flat panel displays of any sort. So it was more, I would say physical design oriented though software was just becoming creeping in with early four bit microprocessors and things like that into various devices. So it was necessary to learn to program a little bit, but there wasn't no such thing as user experience as a profession until a bit later.
- Brendan Jarvis:
- So you really arrived at what seems like a sweet spot in time where you were able to parlay your interests in engineering into human factors just as software was becoming, I suppose a medium of choice. So a growing industry.
- Daniel Rosenberg:
- Yeah, and again, it was somewhat serendipitous and I'd mentioned my father was a professor at the University of Nebraska and he had a fairly well-equipped scientific laboratory, which included the first data general mini computer made put into a university setting. And so as a teenager I did learn to program. I didn't think it was very interesting after a while. And so I said I wanted to be a civil engineer and build bridges, but knowing how to program turned out to be very, very useful at that inflection point. And in fact, one of the first professional instances was I left graduate school and I was working for a company in Al Oldtown, Alexandria, Virginia, that was a human factors engineering firm. And they were starting to do commercial work and Xerox was a client of theirs. And so they put me on assignment in Rochester, New York and Xerox. And it was just at the time, Xerox was putting the first c r t into a copier. The human factors team was embedded in the industrial design department and the focus was clearly getting all the metal and the plastic formed and card control panels. But there was the C R T in the middle of this thing and everybody was looking around saying, does anybody know what to do with this thing?
- And I kind of raised my hand and I said, yeah, I can do that. Do what's on the screen. You do the physical stuff.
- Brendan Jarvis:
- Of course I'm imagining you didn't actually have any huge depth of experience at that point to have put your hand up for that. And correct me if I'm wrong, but you are someone who has a fairly clear track record of firsts, you know have contributed, as I mentioned, your intro to the first handbook of hci. You're attributed as the person who first brought rapid prototyping to the masses. You were employee number one at Oracle in UX. Who gave you the confidence or what gave you the confidence or the right to do what you've done?
- Daniel Rosenberg:
- I assume the confidence is probably genetically inherited, more maybe from my mother than my father. But it was only really pursuing things that were interesting and often in these cases, nobody else raised their hand and said, I'd like to try that. But that's really the origin of any new field is kind of trial and error and adventure
- Brendan Jarvis:
- And it's that spirit, which I suppose gives us some of the stories of the things that people like you and others of your elk like Don Norman and Jacob Nelson have done. Is that spirit still alive and well? In your observations of your students and of the firms that you've been consulting to?
- Daniel Rosenberg:
- I think that I've been fortunate to work with clients, many of whom are doing things that are very cutting edge. I think with regard to the user experience on much of it, it's necessary but not sufficient. So if you mentioned, well Doc, for example, getting the UI correct for a product like WellDoc Diabetes Solution, which maybe we'll talk about later, but really the first instance of digital medicine, it was important, but the IP was in the expert system and in the diabetes management technology. And so the UXs, I would say my students want to work in areas that are in innovative and things like that, but they don't have the expectation so much that they're inventing the field and so much has been done and so much can be borrowed from what's been done in the past. So it's different. It's a mature profession as opposed to one that was being invented 40 years ago.
- Brendan Jarvis:
- Yeah, I suppose you're, you're right. I mean the passage of time has passed and we are at a different place. You mentioned WellDoc and let's just briefly have a chat about that and we can potentially come back to it in more detail later. But just for people that aren't familiar, what is WellDoc and why is it noteworthy?
- Daniel Rosenberg:
- WellDoc is a startup in the Washington DC area and over a decade ago invented a product called BlueStar, a solution for treating high risk type two diabetics, and it's now available for type one. And it was based on some very innovative methodologies and it was actually started before there were smartphones. The first version of the code ran on a like text type of phone. And this particular solution, which deals with everything from medication management, it's capable of insulin dosing which is why it needs clearance from the FDA for the RX version, not the over-the-counter version, mindfulness education, exercise, diet, the whole thing outperforms in clinical trials, Metformin, which is the leading pharmaceutical. And so we have here a digital partner, like a coach, diabetes coach in your phone that was showing on average about a two point reduction in A1C for high risk type two diabetics when metformin shows about 1.8. And what was unique about it was prescribed by doctors reimbursed by pharmaceutical companies and adjudicated as a pharmaceutical, it legally has a drug code when insurance companies pay for it. So it's just basically a digital instance of medicine and of course has no drug interactions cuz you're not putting anything in somebody's body. It was quite a pioneer. And I was just covered, I think in the Wall Street Journal last week or two weeks ago. It was a cover story. Yeah,
- Brendan Jarvis:
- Yeah. I mean it's a phenomenal example of, I wouldn't say it's just user experience though, this is more than that. It seems to me at least it's wrapping in behavioral science. There's many layers and interaction design, there's many layers to
- Daniel Rosenberg:
- Many, many layers. Many layers and many groundbreaking firsts. The first was everybody said the FDA would never approve this and they did and then they said doctors would never prescribe this. And then doctors started prescribing it and they said the insurance companies would never reimburse for it. And then they started to, and then all the venture capitalists showed up. But there were these hurdles of it's never going to happen. None of that had any much to do with the user experience. It was innovation and business model, innovation and technology.
- Brendan Jarvis:
- Are you familiar with, I think it's the meme of the lone dancer on the side of the hill? No. Have you seen that clip? Just to give some context to that. So it's this person at a music festival dancing like a crazy person and they're on their own until that first follower comes and starts dancing next to them, at which point everybody around gets up and starts dancing. Has is WellDoc still that lone dancer on the side of the hill at the festival or are there people coming to join the dance with them?
- Daniel Rosenberg:
- Oh, not alone at all. Mm-hmm. In fact, the focus of the Wall Street Journal article was that digital therapeutics has emerged as a very specific category of health solutions and there's quite a lot of competition. There are early innovators and people who are more established or are less established, but it's different cuz you're not a data logger, you're really treating, and in many cases in digital therapeutics you're diagnosing. And that's why of course the FDA has to get involved.
- Brendan Jarvis:
- My wife's a clinician in eyes, she's a training ophthalmologist and she sees a lot of people with diabetes as a result of poor management of their disease. So something like this is truly a, from what I can tell, striking it the root cause of the problem, which isn't actually putting the pill in your mouth that's actually changing your attitude and behavior to how you see your health and how you see your disease.
- Daniel Rosenberg:
- That's
- Brendan Jarvis:
- True. So Daniel, we've spoken a little bit about firsts. I wanna speak a little bit about seconds now. I understand that you've worked for 11 CEOs in your career and that you've said in the past that Larry Ellison of Oracle was your second favorite, which is something that I doubt that Larry would like to hear, but it's the truth according to you. Who was your favorite c e o and why?
- Daniel Rosenberg:
- Philippe Kahn at Borland. Philipp was a renaissance man who's a genius. He definitely had the crazy Silicon Valley c e o personality but he was personally extremely innovative and set the culture for a lot of the gooey innovations that came out of Borland. And so that was very, very high on his priority list.
- Brendan Jarvis:
- And of course one of those gooey innovations was, as I mentioned in your intro tab ui, which you have a patent with your name on it, which I believe expired 20 years ago or so. So just putting into context just how long ago that was, that culture, that culture that you had at Ashton Tate, which was then bought by Boland, if you could describe it in any terms you like in three words or if you could liken it to something else what would you say about it?
- Daniel Rosenberg:
- I would say it was honors physics class in high school. All the nerds were there and they were all really smart and the teacher was basically the chief nerd herder and it was a joyful place to work in that sense. And so it was a very, very unique culture. Microsoft succeeded in stomping it into the ground, but while it was there it was the company to beat.
- Brendan Jarvis:
- Well let's talk about perhaps situations that aren't so joyful. You went on from Boland to Oracle and you had a fantastic career there really again, a number one sort of leading the charge there for user experience. I wanna come back to Larry and your experience with Larry. If you could just take us back to your first presentation to him and the board in that boardroom and tell us what that experience was like and how that went.
- Daniel Rosenberg:
- It went well though. I kind of thought I was going to be fired. There's a lot of stories about Larry and he is a larger than life character, but he actually listens and he knows what he doesn't know. So I had done my survey of the land, so to speak, having been there for a month or two. And I came into the boardroom with the support of the person who hired me, so who was my boss when I was there for a long time. And I said, look, this is what we have to do. This is what usability is, this is where you are, these are the systemic changes and the investments that need to be made. And I started laying all of this out and I would admit that I was probably nervous and going a little fast. And I got to about the fifth slide of a fairly short presentation and Larry looked at me and he said, Dan, stop. And I figured, okay, this is
- Brendan Jarvis:
- Not going
- Daniel Rosenberg:
- Well, what I was hoping for. And then he looks at the rest of the room, which was mostly, he was essentially the head of engineering. So these were all the engineering executives in the company and he looks at them and he says, I don't understand exactly what you're saying, but I know this is very important. And since this is all new to me, I'm sure it's very new to everybody else in the room. I want you to go back to slide one and start very slowly and take us through this. Don't assume we understand any of the foundation here. That was my first experience with Larry and I would say within six months he signed off on a 10,000 square foot usability lab. We got an entire floor plus on one of the towers that was under construction.
- Brendan Jarvis:
- There is an emerging perception or potentially an attitude in user experience these days that the job of fighting that usability battle has been won. And that labs and usability testing is somewhat secondary to formative or generative research. I believe you disagree with that. I believe you still see there's an important role in there for things like lab in a company that values user experience. But let me know if I'm putting words in your mouth.
- Daniel Rosenberg:
- I think there needs to be a balance today. I mean in the early days the focus was very much on measurement and there was a lot to correct and getting unassisted task completion rates to reasonable levels before products were released because the cost of not doing that was very, very high with some support organizations and otherwise, obviously discovery research is at this point the bulk of, I would say the user research practice and it's certainly extremely important. I do think that one of the reasons that there's such a strong emphasis on discovery user research is, and I know you cater in your podcast to product managers, is because product management is so weak universally and they're so often incomplete or simply non-existent requirements that designers are faced with that the design community has taken it upon itself to fill that hole.
- Brendan Jarvis:
- I was going to ask you that. Whose responsibility is it to define the requirements? Is it design or should it be product management? There is tension there.
- Daniel Rosenberg:
- I think it should be product management in my opinion. While design is a very mid maturity kind of spot right now, there's still a lot, I think they could become more mature. Product management is quite random. Some companies have very good capability. Other companies, you get subject matter experts, but they don't really have the MBA level skills. In many cases you get engineers who are tired of programming and that's not really helpful cuz they don't bring any research skills and they only bring kind of their perspective of what the old way of doing things was. So certainly discovery research is critical, but to go back to your question, I spend the most of my consulting in medical products and the idea that you're going to AB test how many people die doesn't work. So you really do need very tight, precise laboratory testing on prototypes and simulations.
- Brendan Jarvis:
- I mean, it's not an acceptable outcome for a patient to die as a result of poor design.
- Daniel Rosenberg:
- No, it's an, it's absolutely to be avoided at any cost and that's very different than the notion of clips, clickstream analytics, being able to tell you everything in a consumer product. You can't just ab test your way.
- Brendan Jarvis:
- I do want to come to UX Magic and I promise you we will get there soon because it is a phenomenal invention. I know you're standing on the shoulders of some other giants here as well, but I do want to give it appropriate airtime. But before we do that, I wanna come back to another life or death, albeit figuratively, this time situation, which is the boardroom. And there's a lot of torque and design circles and there has been for a long time about getting the seat at the table and you are someone who has occupied the seat, the top design seat at the table in two very significant Silicon Valley big tech companies. Now I've heard you say in the past, Daniel that, and I'll quote you now, being in a boardroom every week is a blood sport. So literally at the end of the quote, this is literally life or death in terms of people's careers type environment.
- I imagine that make that thought for people that haven't yet got to got that seat or perhaps design leaders that are in that seat, that thought of getting into that boardroom every week is a little bit uncomfortable, still makes people a little fearful perhaps. How did you come to, or how did you overcome that discomfort assuming you did face some discomfort? How did you get comfortable going toe to toe with the SVPs of engineering or the C E O or whatever it was to fight the corner of doing what was best for design and for the users of the products?
- Daniel Rosenberg:
- Well, the first thing to really understand is your obligation, let's say if you're the chief design officer is actually to the shareholders of the company. If you go in and make the argument about perfect usability, that's infinitely expensive. So there has to be a business reason that you're doing things. And as long as you really do two things, I would say stay focused on what's the business value. Happy customers make you money, good reviews of your products make you money. And good user experience if you have a direct sales force, helps the sales force make money for everybody. And so you have to understand where you sit in the business. And the other thing you really need to do is we're trying to give advice to the next generation is you have to be neutral, you have to be Switzerland and be very fact-based. And one of the things that did work in the early days at Oracle, which I think was really one of the first large UX teams of over a hundred people back pre.com era was having the usability lab and having the data.
- We were doing a lot of testing, testing was required and the data speaks for itself. And you definitely have competing lines of business with competing large ego SVPs and general managers and the data speaks for itself. So it's not about you think their product's a problem, product date, task completion rate, new users unassisted, 82%, product B 40%, which one is ready to ship the data tells you it's not the opinion of me or any other leader in the company. One thing that I thought was actually kind of a funny Larry Ellison story was we got a new executive and at that time we did have weekly reporting of usability test results in his staff meeting. And this new person had no idea of what we were doing. It was all new. And Larry basically said, you see these numbers And it was like rank score testing results and what the goals were for release said this is, just, think of it like the IQ of the product. If you see these numbers, if the product isn't over this level, it's too stupid to ship, right? It's that simple, right? This is how we manage risk and that's what it was about. And so that would be my advice, but you are not going to get away from politics at that level of any organization. Doesn't matter whether it's Oracle or the Catholic church.
- Brendan Jarvis:
- You've talked about this as well. There are so many insights that you gave in your SIGCHI acceptance speech, and this is actually one that I wanted to zero in on here, which is that politics that exists and becomes more intense the closer to the top of the organization that you get. And I'll quote you again now, you've said when you get to the level of conflict in enterprise leadership, you are basically in the business of removing enemies and maintaining your moral authority when you do so. Now, I understand that some of your thinking about how to navigate the political waters of the organization have been informed by Don Coone and while you were at Oracle, there was a situation that involved a colleague, which maybe you were touching on just a little bit earlier Larry Allison and two versions of the same product that illustrate this particular point particularly well.
- Daniel Rosenberg:
- Yeah, there was a situation where there were two competing divisions of the company trying to innovate a product idea. I would almost say incubate it. And it was really pretty much the same product. And of course we had a central UX organization that was supporting the whole company. And so you had a very good idea, you know, saw everything I think, and a lot of my talks of me have said, it's the all roads lead to Rome kind of thing. And we ended up doing a couple things. One was basically the same UX design for both, but due to implementation issues, one was really performing better and one, the VP of the one that was not performing well, I guess made some kind of aggressive move and basically told the C E O that the other thing didn't even exist. And so I got a call from Larry who said, which was the most typical conversation was what's going on because of that you see everything crossing in UX. And he said, does this thing exist? And I said, yeah, you want to come try it. It's in the usability lab right now. It exists, we're running subjects on it. But again, that's a very neutral position. This is just fact, right? We have the code, we're testing it with live users. That tells you everything you need to know about how ready it is. And you sort out the politics, you're the CEO O,
- Brendan Jarvis:
- So you didn't wash your hands of it, but you maintained that Switzerland position and you didn't throw your colleague under the bus, but you also were completely honest with the C E O without putting your own spin on it,
- Daniel Rosenberg:
- Right? Yeah. But I think that's also the case and most of the software stuff, I mean Philippe Khan had a wonderful expression about this. He used to say code talks, BS walks. So show me the product, use it, we'll try it, okay, but hands on, no arm waving in the boardroom. Give me the keyboard, gimme the mouse, and let's see what the coat does.
- Brendan Jarvis:
- What actually happened
- Daniel Rosenberg:
- As a really long time ago, I think the teams actually got merged as a result. And obviously one of the two executives ended up with the responsibility. I don't actually recall which
- Brendan Jarvis:
- One. I do recall something that you did say about this, which was given the way that you handled it, it actually ended up working out that once you left S A P, I know that we're talking about Oracle here, but in 2012 when you left S A P and you started rCDO that this other person in that Larry situation, that story actually ended up referring you your first piece of business.
- Daniel Rosenberg:
- I think that's true. None of it was personal. Mm-hmm. Right. And I think the fact is that you have to be, I mean, as I said, it's a dangerous political place, but you have to be trustworthy, particularly because if you're the UX leader, you don't own the factory, you're a service provider, you don't have the means of production.
- Brendan Jarvis:
- There's one more story or insight before we come to UX Magic that while we're on the sub subject of UX leadership that I think is really important to share with people today. And you're a really quotable person, Daniel, so I'm going to quote you again. You said trial and error in design is very informative. Trial and error in management is suicidal. You can lose a lot of political capital and friends, like I said, I thought this was such an important insight. It almost seemed to me like you were suggesting that design leaders need to completely switch the mindset, get rid of some of that design dogma when they step into that management or leadership position. Is that what you were getting at there?
- Daniel Rosenberg:
- I think what I was trying to get at is that when you do trial and error in management, you're affecting people's lives and it causes a lot of collateral damage. And one needs to be very intentional and thoughtful about change management in general in every way, whether it's an organization change or it's a process change or a method change that these things have consequences and you just don't wanna try stuff and see if it sticks. And when you see companies that reorg every 90 days or 180 days, that's because they're basically in trial and error mode and they don't really have a vision and they don't have an operating management structure that's sustainable
- Brendan Jarvis:
- And probably not a lot of joy for the people that are working there.
- Daniel Rosenberg:
- No, it's whiplash.
- Brendan Jarvis:
- Well, let's talk about UX magic. Now. I realize we're going to go from a 30,000 foot view of management and strategy down into interaction design here, but this is your new book more than a book though. It's a entire way of rethinking interaction design. It's a system, I have my understanding of it, but you are the creator. So what is this? What is UX Magic? What is Semantic Interaction Design? How does it bring together all the things that you've learned over your 40 plus years and why now? Why is it important now? Now that's a lot of questions. So maybe we just start with what is it and we go from there.
- Daniel Rosenberg:
- Okay, so Symantec Interaction design is the term I use for what I would consider to be the kind of end-to-end methodology, but I don't wanna take credit for inventing the foundation at all. As you've probably seen in some of my talks though, original work in this area goes back into the seventies. The gist of it is the following that people do problem solving using natural language. When we look at any kind of graphical user interface, our brains are decoding it into objects and actions. And it's for digital natives, it's kind of explaining water to a fish, but they see actions are in menus, objects are things we click on in the canvas area, and objects have a set of attributes. And we know from spoken languages as well as a large body of research in the usability of programming languages and learnability programming languages, that language grammar affects cognitive load.
- We know Mandarin Chinese is the hardest language to learn and master, and we know the Semitic languages like Hebrew and Arabic are the most regular, symmetric, fewest exceptions. And English is kind of in the middle, but not particularly good. So you can start by designing what has been called the conceptual model. The first book on the topic of conceptual models that's relevant was Jeff Johnson and Austin Henderson's book, I believe in 2012, that laid out this notion of designing the grammar before you design the product. Jeff and Austin's book stops at sort of that level. What I did is try to create a pyramid out of it where we talk, where you design that foundation. The thing I added to the foundation was the concept of prioritization. Because a task in human factors parlance is some object plus one or more actions. And not of all tasks are equal importance, right?
- Either to the user or to the business. And then added on top of that, how do you map semantics to visual design patterns at everything from the component to the widget, like level widget being maybe tabs or a date picker to a whole wizard. And then to screen archetypes and introduce the notion of grids and grammar where you're reserving locations on the screen for specific semantics, not grids as gutters and spacing. And then how to take semantics into flow design of experiences, which is again, is not something I invented that really can be credited to Jesse James Garrett again 30 years ago. And then I go further into applying how it works in game theory if you're trying to incentivize users to choose particular objects or actions or flows over one another. So that's kind of the gist of it. And the goal is lowest cognitive load, shortest flows, fewest screens, and you can kind of figure out what the cognitive load will be without even drawing a screen. Once you figure out the semantic res design.
- Brendan Jarvis:
- It sounds like you are trying to bring us back to, the only way I can frame this is to first principles by going back to grammar, going back to this very basic building block that we seem to rush past these days, that you want us to come back to those first principles.
- Daniel Rosenberg:
- That's correct. I think what I am advocating and as people are adopting the methodology is designing what you're going to design before you start drawing screens. It's too easy to start sketching and in fact there's generations of designers who were taught just start sketching and it's just, it's too expensive. So when we run the tutorials through Rosen Blue, it is 10 x that's the marketing moniker and not semantic interaction design. It's just design 10 times faster and get 10 x better results in terms of the lowest cognitive load. Again, it's a somewhat economically motivated approach. Even running large teams at the era I was at sap, I think we got up to 250 or 260 people in UX team, but there're 12,000 developers in the company. So you need to be fast and you need to, and I wanna say not iterate, iteration is always necessary, but start from a much more intellectually solid and cognitive cognitive science foundation that's more solid and you just do better.
- Brendan Jarvis:
- And one, from what I could gather, one of the basic concepts in this in Symantec IX D is the density of the grammar on which the mm-hmm [affirmative] interaction is built. So I know we are limited by this medium because we're not in a visual medium. Some people may be watching this on YouTube, but it's still very much without presentation slides or anything like that. But how would you, if it's possible, paint a picture and words for people to understand the difference between a sparsely sparsely populated object action grammar and a densely, a dense object action grammar and why you actually want a dense object action grammar.
- Daniel Rosenberg:
- Sure. So I'll try and paint the picture using example that I use in the book from word processing. Today we think of the objects in using a word processor as characters, words, paragraphs, documents, et cetera. And we know that there are consistent actions like cut, copy, paste, that apply to all of them. And so if you were to look at that matrix of objects and actions, it's pretty tight. They're very few exceptions. The early word processors, let's say even pre pre gooey had tons of speed keys and function key combinations. And you'd have unique key strokes for cutting a character and removing a word and deleting a paragraph. And so if you draw the graph of all the objects and all the actions just to do something as simple as writing and a word processor, the matrix would be very big and there'd be very few combinations that were consistent.
- And when you have this sparse, highly irregular, lots of inconsistency and lots of let's just say dead space in terms of this action only applies to one very niche thing, cognitive load grows as the number or is approximated it be fair to say, as the number of objects times the number of actions. So cognitive load is going to grow exponentially as you add functionality. If the grammar is very tight and very regular and as few exceptions as an omit and foy as possible, it will grow linearly. We'll give a number of actions times the number of objects. And so there's a huge difference in when you tighten up the grammar and make it the fewest number of concepts that can most consistently reflect the functionality.
- Brendan Jarvis:
- And this is at the core of your claim the 10 x claim. And I know you've got a couple of case studies where it's actually greater than 10 x in terms of the reduction in cognitive load between existing product and then re-engineered product. But this is it, isn't it This ability to predict out what the cognitive load will be as we crunch down our grammar
- Daniel Rosenberg:
- And the result is easiest to observe. And there are case studies in the book where in one medical EMR system, 800 screens down to 50. Okay. I think we can all agree that if you can do the same functionality in 50 screens and it's all graphical user interface that you did previously in 800, usability is better, learnability is better, cognitive load is lower. While we can measure cognitive load in a usability lab, you don't even have to do the experiment. It's pretty obvious. So been working on another product for a client that I think would be upset if I mentioned who they were, but it's again a do-over and it's very sophisticated scientific and it's kind of 500 screens to 15 and everybody's saying, oh my God, why didn't we do it that way the first time? Because it's a lot less to code,
- Brendan Jarvis:
- I imagine it'd be a lot less to maintain as well. So it's not just the, oh yeah, user benefits here. It's actually, there's significant cost savings in supporting and maintaining that code
- Daniel Rosenberg:
- Over the life cycles tremendously. Again, tremendously advantageous for everybody. It's a win-win, right? The scientists get a cleaner, simpler UI for a very complicated thing. Company doesn't even need to write a user manual. The support team doesn't need to run webinars for weeks to train people. It's a lot of it's, it's all
- Brendan Jarvis:
- Good. What does it do to headcount?
- Daniel Rosenberg:
- I don't think it does anything to headcount because at least when it comes to the design community at this point in time, demand exceeds supply. There's always more to do. Maybe in the future that would be, but I think the bigger threat to headcount, and I assume you're going to go in this direction and at some point in the interview it was automation and artificial intelligence doing generative design.
- Brendan Jarvis:
- I do want to go there cause
- Daniel Rosenberg:
- That could change the employment scene.
- Brendan Jarvis:
- Yeah. And you've been working at that inflection point which is really rich territory for us to explore soon. I just wanna come back to the UX magic, the book itself, cuz there were a couple of things that I noticed there looking at the cover. So one is the title UX Magic, the other is the graphic of the Rabbit and the hat, which is, it's probably a fairly universal graphic that comes to mind for people when they think about pulling a rabbit out of a hat. It has that kind of connotation. You have expressed perhaps sarcastically in the past somewhat of a dislike for that magic moment and design where the designers, the creative people go off to the coffee shop and sketch out what it's going to be on the back of a napkin. Were you poking fun at that sacred cow or that sort of belief that's held in design, that creativity exists in that way with the book title and the cover? Or am I just reading too much into that? Yes.
- Daniel Rosenberg:
- No, you read it the way it was intended. It's basically a pun, right? It's the theory from a long time ago instilled the operative theory for a lot of the profession and a lot of executives not UX executives per se, but in general is there's this magical moment when somehow the creatives get the requirements, whether they intuit them or there's good product management or whatever, and then they just go off and start drawing. And genius, everybody's a da Vinci, but it's not magic. That's really the point of the book. There is a science and there's a methodology that will get you much, much closer in much shorter period of time. And then you can start from that foundation and yes, you'll have to iterate and you'll have to sketch, but you'll be in a much stronger position intellectually,
- Brendan Jarvis:
- Dare I say, this is almost engineering proceeding design
- Daniel Rosenberg:
- If you consider it cognitive engineering. Yes.
- Brendan Jarvis:
- And you've said in relation to your role as an educator, which I believe part of the inspiration for this was actually you couldn't find something that you could impart to your students how this should be done. So you actually created it. But you've said, and I'll quote you again now today, the practice of interaction design is overly intuitive and it doesn't need to be. That is almost a goal of design, isn't it? To create something that is completely intuitive and people don't have to think terribly much about it, but you're actually running up against, not necessarily that because I know that's the outcome of design and not necessarily the process of it, but what is wrong with our education and design being overly intuitive these days?
- Daniel Rosenberg:
- Well, I think the major issue is there's not no one pedagogy for design education. In fact, this is one of the problems that the future of design education project is trying to solve, which is why I was on this call with Don this morning cuz he's sort of the leader of this activity. So you have a very broad range of user experience being the focus of art schools, colleges of art and design where it's really an mfa. You have probably the far a reach from that at the same academic level as a human factors engineering program, which is very quantitative in its basis. But then you have all these boot camps and then you have center Center or whatever Jared Spool is calling his educational thing these days. And there isn't a standard pedagogy and there's no licensing, there's nothing that says this is what it takes to be a designer.
- And so what I ended up doing with the book was these were techniques that my teams used for years and decades all along. And I was just really surprised I didn't find a comprehensive book. There is Jeff and Austin's book, which I immediately started using as a textbook but it wasn't the full picture. And so that was kind of the motivation. But I don't know what design education is, it just ranges from very established design schools at the Royal College and Delft, the leading universities of the US to an online bootcamp where you go from being a real estate agent to a full stack UX person in 10 weeks.
- Brendan Jarvis:
- Mm-hmm. Something wrong with that picture, I gather.
- Daniel Rosenberg:
- Well, again, if you take a 10 week approach, it's going to be very tools focused. It's not going to be scientific. If you go to, again, take the human factors pedagogy, everything is about cognitive load and avoiding errors and not crashing the airplanes.
- Brendan Jarvis:
- How confident are you that through the efforts of what Don's doing and what you're involved with and others, how confident are you that this is a solvable problem?
- Daniel Rosenberg:
- I think you'd have to define what you mean by solvable. Can we make improvements? Yes. I think the only quote solution is a controversial one, which would be licensing. So the same way that a civil engineer needs a license to practice and build a bridge or a mechanical engineer or a doctor or a lawyer, or particularly an architect, it's most similar to architecture. They need a license to practice. And that's kind of the only way you're going to enforce a base level of consistency in education. But the UX design world has had an allergic reaction to that idea for decades.
- Brendan Jarvis:
- And from what I hear that's tied up with this notion of design is something that everyone should be able to do and therefore to gate keep it like other professions would be the antithesis of what it was designed to do
- Daniel Rosenberg:
- Or what it is to be a human.
- Brendan Jarvis:
- It seems to me that gets us fairly fluffy that I don't know how that helps the field. I'm not quite sure how we increase the credibility of a field that doesn't seem to have any well-defined boundaries.
- Daniel Rosenberg:
- That's true. But being able to have a somewhat more universal curriculum that would include certain foundational material, which all needs to be adapted to local culture and the goals of a particular academic program, but they're not all identical. That would help up level the field. But then again, that deals specifically with academic programs and the future of design education does have a track on non-academic programs. And I guess I play in both worlds. It's a little contradictory, but being on the board of the Interaction Design Foundation, which is perhaps the best, at least I would give it a plug online resource, and one of the lowest cost resources in terms of education related to interaction design versus being an a faculty member and an established program. There's room for both and they're not mutually exclusive. I see our students looking for online materials and courses that are beyond the curriculum that we have at our university and we're reasonably well funded.
- Brendan Jarvis:
- And I suppose is an unfair expectation that anybody who engages in education of any form could place on it, which would be that it would give you every answer that you could ever wish for and correct. It really is just skill for every foundation, every skill, right? Yeah. But what you are talking about sounds like we just need to get more intentional about what the foundations of design are and get some broad agreement about what those are so that we can actually educate the next generation of designers in a way that we can be more confident that what they bring to the table is what is needed to drive design and products and businesses forward.
- Daniel Rosenberg:
- Yes, that is true. And there are a number of things that are more universal that are not sufficiently dealt with in education in terms of ethics and sustainability, regenerative design, things like that. So as designers, we have contributed to a lot of damage. And I think this is best captured in Nathan shed Roth's book design is the problem and there's a lot that needs to be done better. And so even those core things, and the same would be true with computer science, for example, having a technology ethics class is a mandatory class is a relatively new thing, but it's highly, it is in the ACM and the I e E curriculums standard curriculum recommendations, but you can't even find a standard curriculum recommendation for UX.
- Brendan Jarvis:
- Mm-hmm. And just before I bring us down to the close, cause I'm just mindful of time, I couldn't help but notice in your career, Daniel, that you have made a, you've had a couple of shifts, you've shifted out of what I would consider to be almost a sole focus on hands-on practice of design into management. And then you've shifted again from that into well leadership at the global level and then again out of that into, and then maybe this isn't fair, but into education and consulting. So kind of left that big tech world behind from the inside, at least playing an inside role in it to sort of more advising it from the outside. But you seem with UX magic and listening to the passion at which you talk about the future of design and what you're trying to achieve with UX Magic and semantic I X D that you've never really let go of what I would consider design practice. Are you a polymath in that regard that you've managed to both lead design while maintaining your eye on practice and hands in practice? Or has this been a tension for you through your career?
- Daniel Rosenberg:
- No, it hasn't been a tension. And most of my consulting practice is
- I work as for several companies as the UX architect. I know the name of my consulting company, rCDO, rent a Chief Design Officer. I don't do very much consulting. That's strategic. I mean, there's some, but a lot of it's very tactical. Mostly do-overs design of a new product maybe in an existing category, but usually shift when there's a shift from on-premise to cloud and everything has to be recoded anyway and it needs to be done and it needs to be done fast. So to me, doing the design work is always very joyful. You go into management when you can't scale yourself, you just can't do everything. But I was always extremely hands-on in my executive days, having competent leaders under me to participate in all the design reviews and occasionally do my own individual contributor work. So maybe that's my one advice to anybody for a long career is you need to be a player coach. If you're going to be a UX leader, you need to be able to get on the court and play ball, and then you need to know when to get off the court and be the coach. And then sometimes you need to know how to be the general manager, but the joyful part is never management, the joyful part's always designed creating something
- Brendan Jarvis:
- After 40 years in the field, you've seen, you've done and you continue to do a lot. And most of your work, as we've been speaking about most recently, has been hands-on and it's been at this intersection of AI powered products and interaction design, and there's one product that you've worked on that I believe could take a set of quite loose requirements or specified through the gooey by anybody and translate that into a fully functioning product that didn't need any input from an engineer or a designer to actually make that happen. Is the sun setting on how we currently see our role as designers and what we currently do?
- Daniel Rosenberg:
- Yes. I think it has begun to set for the simple cases. So I think you're referring in particular to the case study I did on clean software where the goal is to describe the requirements and the semantic model, do that level of intellectual work, feed it into an ai, and the AI can generate experience, design options. And actually, again, that's only part of the IP because the rest of the IP actually writes code and generates a working system and provisions it to the cloud provider of your choice that will do a bunch of things that will get rid of a lot of the really easy and often highly manual work. I guess if I would theorize what it would do, I think it would, the AI can easily compete with the bootcamp graduates already in certain domains that are fairly repetitive. And at the same time, it gives the opportunity for the design community to have a deeper role to become more product designers and not UX designers and become problem solvers at a bigger level because a lot of the grunt work is taken away.
- Brendan Jarvis:
- You've spoken about the importance of defining terms in the past. Can you just for our listeners, and for me also, could you just define what you see the distinction between being a UX designer and a product designer? Like why do we need to be less UX designer, less of a UX designer and move more to becoming product designers? What are those two things?
- Daniel Rosenberg:
- I think when most people think about UX design, they think about what's on the glass, or if they're doing conversational design, what comes out of the speaker, the smart speaker product design is really going to the root, including all the design discovery research and figuring out what is the problem to be solved, what is the systemic improvement that you're going to make through the creation of whatever new product or service is going to be delivered. And so it's a much more big picture view than the complaint that let's say a lot of UX people have of just being mock monkeys and just being said, okay, make this screen, make that screen deal with this micro interaction and move up to the big picture level.
- Brendan Jarvis:
- Sounds like we might find ourselves competing more and more with MBAs.
- Daniel Rosenberg:
- Well, that's not a new thing. There are MBA programs and design strategy. So in the boardroom, if your goal is to be a chief design officer, you're absolutely competing with MBAs from some very prestigious universities that can claim they have specialization in design strategy, even though they've never designed anything of the scale of a birdhouse or a doghouse.
- Brendan Jarvis:
- That could possibly be a podcast, another podcast in itself. That topic. Dan, I am mindful of time. Thank you for investing this time with me and for sharing some of your stories, your insights, and clearly well earned and hard earned wisdom. I greatly appreciate, as I'm sure our listeners do as well, your significant and continuing contributions to this field.
- Daniel Rosenberg:
- Well, thank you. Thank you for inviting me. I hope your readers, followers, viewers find it all interesting.
- Brendan Jarvis:
- I have no doubt that they will, and it's been my pleasure. Daniel, if people want to find out more about you, more about the book UX Magic and your work at rCDO, what's the best way for them to do that?
- Daniel Rosenberg:
- The best way to go to the rcdoux.com website and you'll feed me lots of links and references and also how to contact me and
- Brendan Jarvis:
- People have questions. Great, thanks Daniel. I'll make sure that I link to that website, your website, and the show notes to everyone that's tuned in. It's been great to have you here as well. Everything we've covered, including the rCDO website, will be in the show notes and we can find Dan Daniel on LinkedIn and all the other things that we've spoken about will be chaptered in the show notes on YouTube. If you enjoyed the show and you want to hear more great conversations like this with world-class leaders in UX, design and product management, don't forget to leave a review on the podcast, subscribe, and if you feel that other people in your sphere would get value from these conversations about design and product and UX depth, then please share the podcast with them. If you wanna reach out to me, you can find me on LinkedIn as well. Just search for Brendan Jarvis, or you can find a link to my profile at the bottom of the show notes as well. You can also find me at thespaceinbetween.co.nz. That's thespaceinbetween.co.nz, and until next time, keep being brave.