September 23, 2020

Molly Wright Steenson - Architecture, computation, interaction design, AI and ethics

My guest in this episode is Molly Wright Steenson, a designer, author, professor, and international speaker whose work focuses on the intersection of design, architecture, and artificial intelligence.

In this wide-ranging discussion, Molly explains how the history of computational technologies, architecture, pattern language and AI combined to define the fields of Agile, interaction design, UX, AI and pretty much the rest of today’s digital world.


Transcript

N.B. This transcript is mostly AI-generated. There may be a few errors.

Andy Polaine: [00:00:00] Hi, and welcome to Power of Ten, a podcast about design operating at many levels, zooming out from thoughtful detail through to organizational transformation and on to changes in society and the world. My name is Andy Polaine. I’m a service design and innovation consultant, educator and writer.

My guest today is Molly Wright Steenson a designer, author, professor and international speaker whose work focuses on the intersection of design architecture and artificial intelligence, a topic more relevant than ever today. She is the Senior Associate Dean for Research in the College of Fine Arts, the K & L Gates Associate Professor of Ethics and Computational Technologies and an Associate Professor in the School of Design all at Carnegie Mellon university.

And as if that wasn’t enough, she’s the author of “Architectural Intelligence: how designers and architects created the digital landscape” and co editor of “Bauhaus Futures”, both published by MIT press. Even better, molly owns the best domain name ever, which will reveal at the end of the show. Molly, welcome to Power of Ten.

Molly Wright Steenson: [00:01:01] Hi, Andy. It’s good to be here.

Andy Polaine: [00:01:04] I’m looking at you sitting in front of lots and lots of books. So I’m assuming you do quite a lot of work from home anyway, right?

Molly Wright Steenson: [00:01:10] Well, I’m, I’m a writer as well as a professor. So, but I have to say, I just succeeded at getting 15 boxes of books to my office at school. And just as I finished the last box books, Coronavirus hit. So now I had to bring back a couple more bags here and we’re all living, living in distributed times, but I think that that’s the worst, the worst complaint I have right now. My dog very much likes the invention of stay at home and shelter in place.

Andy Polaine: [00:01:43] I’m pretty sure it’s some kind of, conspiracy by pets too, to just get there owners to stay at home. So, now you’ve talked about yourself as an old-school web geek. And “since 1994, I’ve worked in many capacities as a UX designer and strategist design researcher, writer, and geek.”

And you said you built the first news delivering website, Reuters, in 1995, which, and managed the “second most hit page on the internet”

That’s such a, that’s such a telling kind of expression these days, even for the fact you said the most hit page. So tell me a little bit about that history.

Molly Wright Steenson: [00:02:18] Okay. So it’s a long, strange history and I promise you that it all comes together in the present moment. I was a kid who had an Apple IIE computer at home. I’m now 48 years old. So that kind of gives you an idea of, of where I fall. I’m a child of the 70s and 80s.

Andy Polaine: [00:02:33] Do you still have it?

Molly Wright Steenson: [00:02:36] Somewhere and all my silly if then statements in my programs.

And I, when I was in college at university, I majored in German, but I also did a lot of writing for the various newspapers at the University of Wisconsin, where I went to college. And one of those papers was The Onion I wrote for the AV club. And, a number of us who are connected with these various projects got involved with the first web project at the time University of Wisconsin. It was called Online Wisconsin. It was an online newspaper in 1994. And, I guess October of 1994, I had an epiphany that, “Oh my God, this web thing was going to be huge.” And I wanted to do it for a living and I needed to move to San Francisco or New York and fast before it was too late.

All of us who worked on that project went on to really interesting jobs. Pam became the production editor at Wired. Roger was on the founding team of CNET. I landed at a subsidiary of Reuters in New York as a manager of online services, which was funny, cause my previous job was working in a record store. So, so I had this management position right out of, right out of the university and It was just really amazing because you could make things happen.

And from there I moved to California. I was Howard Rheingold’s first employee at his startup Electric Minds. And then,I left electric minds before launch and worked at Netscape with Hugh Dubberly, where I was… Yes, indeed, indeed. And this figures into where we go with the architecture book.

I should also mention that my undergraduate honors thesis was about, the Bauhaus and its connection to sustainability some way or another, through landscape architecture. And so I was no stranger to architectural history and, in my first day at Netscape, Hugh Dubberly suggested in 1996, He suggested that we use pattern languages, this idea from Christopher Alexander in how we think of redesigning the Netscape website. So I thought that was what’s interesting here. I’m hearing about architecture and we’re working on something with the web in 1996. And you know, I think a year later, late, no later that year, I heard the phrase information architecture for my first time when I was working at a company called CKS Interactive. And we looked at Wurman’s information architects book for inspiration. So fast forward through a career that takes me through, you know, e-business consulting through being one of the first content strategists and UX leads or customer experience architects 20 years ago. and to being a professor at the Interaction Design Institute in Italy, which, you might know…

Andy Polaine: [00:05:26] I know from yeah, from Ben and Lavrans obviously.

Molly Wright Steenson: [00:05:29] Yeah, exactly. So Ben and Lavrans were, graduates of the Royal College of Art. The first, service designers and I was part of a team teaching service design at, the Interaction Design Institute Ivrea. And they taught on our course with us. They were very, very important visitors and project leaders.

After I taught at for two years I came back to the U.S. And I knew I didn’t want to go back and work on web 2.0 stuff. being at Ivrea introduced me to classes like, building says interface, which was taught by Stefano Mirti and if you’re in Italy, all designers are trained as architects and we lived in this, you know, just absolutely amazing structure called Taponia by the architects Gabetti e Isola. and I knew I wanted to study how technology fit into the world around us. And so I turned down a job at Yahoo and I went to Yale and, I did a masters in architectural history, and then I used, the many people from that program went on to the PhD program at Princeton. And that’s where I started researching the history of AI and cybernetics and architecture.

Andy Polaine: [00:06:37] So, what kind of, about what time was that?

Molly Wright Steenson: [00:06:40] I was at Ivrea from the very beginning of 2003 to the very end of 2004. So two years in there and, And then I came back to the U.S. In 2005 and started my masters and PhD. So the, the architecture, academic part of my career began 15 years

Andy Polaine: [00:06:58] And the AI a bit, then you were pretty early in - not in AI in general - but you were pretty early in what we’re now kind of thinking of that, the sudden rise of, of interest in AI machine learning right now, you were, you were sort of involved in thinking about that early.

Molly Wright Steenson: [00:07:14] And I think I may have been one of the first people to draw the connection between Christopher Alexander, AI and, the work that we also do in UX and interaction design.

Andy Polaine: [00:07:25] Right. So let’s go there because in architectural Architectural Intelligence, I, you know, I, I know my HCI and interaction design history pretty well, but I somehow missed that Nicholas Negroponte was an architect. I don’t know how I managed to miss that, but that, that bit had sort of escaped me when I, in my sort of readings over the years.

But you, you run through a kind of list of, there’s this conference where, you know, people like Doug Engelbart, where, and you talk about Terry Winograd later on. For any of the, younger listeners that are out there, these are all names that you should know about. Yeah, no, this is seriously this is actually a bit of a, so before I’m going to come back, I’m going to have a little bit of a gripe. Which is, in any other field you kind of, one of the things you learn is about your history. If you go and become a fine artist you learn about, you know, the Old Masters, you learn about kind of the whole history of art and so forth. And the same is true in, in actual product design, industrial design and product, physical product design, and so forth. You learn about all that stuff. And one of my biggest kind of gripes about people who are working in, in loosely sort of digital these days and in interaction design and UX is a lot of them actually don’t know that history very much.

And part of the reason is because most of it’s disappeared from the internet. And so a lot of the work, I mean, you’ve probably had this too talking about your AppleIIo before and your, if then statements,

Molly Wright Steenson: [00:08:49] Or my, which is a whole set of things, the web zine that I ran that has disappeared from the Internet.

Andy Polaine: [00:08:54] So, so all of that stuff is, you know, I mean all of my work. I keep a really old Apple laptop around, so I can actually occasionally, you know, play some of the work I did for a good kind of decade or more of my career is no longer accessible. And so there’s this kind of cliff of history that, is constantly kind of moving until the web happened. And then, you know, now it’s just becomes a kind of, the opposite problem where there’s so much stuff out there you have to find.

So that’s my rant aside. Tell me how, the architectural stuff comes together. Cause you, you, you start to say, well, you know, there’s architects who are, who are actually really influential, In the thinking around the internet and it was a really interesting, fascinating connection. So how does that all work?

Molly Wright Steenson: [00:09:38] A number of ways. So, for me, this journey started, remember how I told you about Hugh Dubberly mentioning Christopher Alexander back in 1996. So what 20, 24 years ago now? Everybody in interaction design. And for that matter software engineering software development seems to love Christopher Alexander architects, incidentally, don’t like him very much, which is interesting and strange, but I started following that. I wanted to, I wanted to understand what it was about Christopher Alexander, that I wanted to understand what was in his work. And so, I read Notes on the Synthesis of Form for the first time in 2007.

This was written by him in 19… I think originally in 1961, and then it was published in 1964, the published version of his dissertation at Harvard. And the book is really interesting for its footnotes. So like you could almost read the back half of i, and, and see all the things that he’s trying to put together as people do in dissertations.

Right. And, I noticed at the end of the first chapter, the very end of the first chapter, he said something he, he wrote, “we must face the fact that we are on the brink of times when man may be able to magnify his intellectual and inventive capability just as in the 19th century, he used machines to magnify his physical capacity again, as then, our innocence is lost. And again, of course the innocence once lost cannot be regained. The loss demands attention, not denial.”

So he wrote that and the footnote for that statement, says “the possibility of amplifying intelligence has already been hint hinted at by Ross Ashby design for an intelligence, amplifier Claude Shannon and John McCarthy. See also Marvin Minsky ‘Steps Toward Artificial Intelligence.'” And I thought, well, that’s really strange because they’re… in the early to mid 2000s, there was a lot of writing happening in 20th century, architectural history about cybernetics. In fact, cybernetics on any number of things. I took a class at Yale on cybernetics and literature, you know, systems theory and literature and, in particular, because of the work of the British architect, Cedric Price, who had died in 2003 had done a lot of work with cybernetics. This was coming up a lot for me at that time, but artificial intelligence was not. And I thought, well, that’s fascinating.

And so after that I started reading Marvin Minsky, Claude Shannon… I am not a math person. So, you know, I I’m at a loss when it comes to most things with advanced equations, with John McCarthy with Ross Ashby, all the people that he notes there. And I started tracing back through their work. And as a result of that, I also began to find my way to Nicholas Negroponte and the Architecture Machine Group and MIT and Nicholas Negroponte in the late 1960s.

Well, let’s see, he graduated from MIT with his bachelor’s in 65, 66, he got his masters five days later, he started teaching at MIT, taking over Steven Coons’s course. And what that meant was also that he began working. So Steven Coons was in mechanical engineering, but he also began working with the AI lab.

He was very close friends with Marvin Minsky till Minsky’s death. and with Seymour Papert and was very closely engaged with the, AI lab at MIT, basically the architecture machine group was building and experimenting with, Interfaces for artificial intelligence that at that point in time, and, um, indeed, Christopher Alexander’s and architect, Nicholas Negroponte is an architect, Richard Saul Wurman is an architect, and all these people were working with information cybernetics or artificial intelligence in some way or another as material for architecture.

And that had really profound influences for the kind of work that you and I do today.

Andy Polaine: [00:13:45] Yes, in what way? I’m going to deliberately…

Molly Wright Steenson: [00:13:51] Our fields don’t exist without it.

Andy Polaine: [00:13:53] So I’m deliberately sort of asking. I mean, I have some pretty good understanding, but explain, explain why that’s had such an influence on us in, in, in it, there is a bit of a leap from, you know, that group of people being interested in this and then where we are now. And there’s a kind of, like I said, there’s a bit of a cliff of history that is not so well known.

Molly Wright Steenson: [00:14:15] Oh, yeah. Well, I’ve very much agree with you on the history front. and I hope we could, we could circle back to that because you know, when, when we say things like artificial intelligence or AI and act like this is a brand new shining term. It’s a term that’s almost old enough to have social security in the United States. It was coined in 1955 by John McCarthy and machine learning was popularized as a term starting in 1952. So these are, these are terms that are eligible for retirement, you know, and we’re, we’re treating them as though they’re very new. I think that can be harmful. if we don’t acknowledge that there are lessons to be learned from a long history of things that haven’t worked in AI before and why they haven’t worked probably more importantly. but in the case of, the U S it’s about the connection to architecture and Christopher Alexander is the architect that a lot of computer scientists have known about, but they, you know, without knowing about other architects, they seem to gravitate toward him.

So people like Alan Cooper who is of course a computer scientist. At the, at the end of it all, or Ward Cunningham or Kent Beck used to read Christopher Alexander in their school libraries or university bookstores. Cause his books are really expensive and yeah, would actually read them in the university bookstore bit by bit.

And when he and ward Cunningham were working on Smalltalk in the late 1980s, I think 1987/88, they realized that they could use, They could use pattern languages as a way to codify the programming languages they were developing.

Andy Polaine: [00:16:03] So let’s just pause there. So pattern language, is a pretty crucial one. Christopher Alexander wrote A Pattern Language. Can you do a sort of summary of what Christopher Alexander was talking about, it’s not just him? Is there’s a couple of other authors as well.

Molly Wright Steenson: [00:16:17] Many other authors.

Andy Polaine: [00:16:19] yeah. when they’re talking about what a pattern language is.

Molly Wright Steenson: [00:16:24] So Christopher Alexander, joined the faculty at UC Berkeley in 19… Oh, I forget exactly what year, but in the 1960s. And he and his colleagues founded what’s called the Center for Environmental Structure and they started trying to find a way that they could sort of codify knowledge about design at every scale. So I liked the fact that you’re talking about Powers of Ten being design at every scale. And this is what they were trying to do from a, the smallest details of where you put things in a house to how you structure a nation state. And, they published a book called A Pattern Language in 1977 and then Alexander at the same time published kind of the philosophy of it, which was called the Timeless Way of Building.

And you could think of a pattern language as an operating system for architecture and what he and his colleagues, Sara Ishikawa, and Marie Silverstein said they defined patterns as a… It describes a problem which occurs over and over again in our environment, and then describes the core of the solution to that problem in such a way that you can use the solution a million times over without ever using it the same way twice.

So, so each pattern has kind of a grammar. It’s got a number, it places it in a hierarchy, there’s a title, there’s a photo, there’s a problem statement. There’s some more information describing the pattern sort of sketches along the way, showing how it works out. And, and how, how, how to explain how it connects to other patterns in the hierarchy.

So, you know, it’s almost kind of hyperlinked in that sense. and so he patterns where they, excuse me, they see patterns as networked, right? It’s a network of patterns and this idea was really, really influential for, and impactful for computer scientists. So I mentioned Ward Cunningham and Kent Beck, and they brought that idea over to object oriented programming languages.

By the early 1990s, about 1994, the book Design Patterns in Software was published. And now there’s something like 1400 books on Amazon that refer to design patterns in some capacity. That’s not an exaggeration, but there are two other ways that that picks up, one is In extreme programming and Agile.

So Alexander’s theory was direct implementation or the, excuse me, Agile software, methods and development methods are a direct implementation of Alexander’s ideas about putting down a set of patterns that aren’t definite, but that begin to allow you to sort of. design from the center out and provided a flexibility in the design process that Kent Beck says is a re a rearrangement of the politics of design, basically.

And the other thing that it influenced was the Wiki, Ward Cunningham apply to Alexander’s idea patterns, and, you know, a conversation that has a center, but no edge to the idea of the Wiki that we all use and things like Wikipedia. So direct connection. And he also really influenced Terry Winograd for how we think of software and the idea of user experience designing for being inside of a space of software, like understanding a user, inhabiting a system. And that’s a direct interpretation of Christopher Alexander as well.

Andy Polaine: [00:20:01] But it’s also like pattern language also seems to be a… it’s almost a sort of genesis of open source as well. As far as I understand it, there’s a, sort of democratizing force behind the idea of it that you’re giving this set of patterns and tools that people should be able to use .

There’s a bit, I’m sort of reading a quote of where it says that “at the core is the idea that people should design their homes, streets and community. And this idea comes from the observation that most of the wonderful places in the world were not made by architects, but by the people.” And it’s very, not only sort of human-centred, it’s very much of a, sort of this idea of you give people the toolbox and the tools and, they will then start to create the things that work for them.

Molly Wright Steenson: [00:20:45] Architects hate Christopher Alexander.

Andy Polaine: [00:20:48] For that reason?

Molly Wright Steenson: [00:20:49] Well, they hate him for all kinds of reasons. And I can’t entirely figure out why. I think they protest too much. I mean, for all of the people in the world that are, are, kind of jerks as human beings, for all of the egos that are in architecture, I don’t know that art that Christopher Alexander is public enemy number one.

Andy Polaine: [00:21:08] Do you think it’s partly because he, he sort of dethrones the architect as kind of auteur of the kind of whole thing in Pattern Language, do you think that’s one of the reasons why?

Molly Wright Steenson: [00:21:18] It could be, you know, the, probably the most interesting part is, Maybe it’s a sort of cliqueishness. So the architect, Peter Eisenman, who is, who is training in many of the same places as Alexander and interested in math, in some of the same ways that Alexander was, his, there are people who’ve drawn a lot of parallels that connect Eisenman and Alexander. Although Eisenman says the reason he did a PhD was because he hated Christopher Alexander so much. I mean, these are men who are now almost 90 years old. So, you know, it gives you an idea.

Andy Polaine: [00:21:55] So much hate, but it’s led to some amazing things. I’m conscious of time. So let’s move to Doug Engelbart. There’s there’s loads of other people who are involved, but doug Engelbart well the inventor of the mouse, inventor of so many things. In fact, I’ll try and find, a link to that this presentation that’s pretty amazing…

Molly Wright Steenson: [00:22:14] oh, yeah, the mother of all demos.

Andy Polaine: [00:22:16] The Mother of All Demos. I’ll, I’ll put that in the show notes, so people who haven’t seen it they should, they should have a look at it and they should also listen to, MarkPesce, talking about this, cause it’s pretty interesting.

So, perhaps you can describe what it’s, can you in a couple of sentences describe kind of what Doug was sort of getting at because, and he was using architectural scenarios as the kind of demos for this. Cause I want to go into the clerk thing.

Molly Wright Steenson: [00:22:40] Yeah. Yeah. So this is, this is way back, even before sketchpad was invented, or, well, I shouldn’t say invented, but…

Andy Polaine: [00:22:47] This is like ‘68. We’re talking ‘67/‘68.

Molly Wright Steenson: [00:22:50] I’m talking about earlier than that I’m talking about the ideas that proceed the mother of all demos. it was 1962 that Douglas Engelbart, published the scenario in a paper that in a report he produced called Augmenting Human Intellect.

So let’s pause for a second and note that a lot of times today we talk about augmenting humans and augmenting artificial intelligence versus automating. This goes back to at least 1962, if not earlier. So again, a very old idea. but Augmenting Human Intellect was basically his proposal for what would eventually become this amazing mother of all demos, this kind of online system that would, connect different user interfaces in different pathways or information.

But his use case was architectural and he described a, you know, basically a computer aided design system that. that had some element of BIM of, building information modeling built into it.

Andy Polaine: [00:23:49] So at the same time, I mean, he’s kind of in the mother of all demos, he’s, he’s kind of basic describing what we’re all suddenly doing now, where he’s he describes the web, he describes the internet. He describes a lot of the kind of interface, but he also describes people working together and working together at a distance.

So he’s, he’s, he’s gone. I mean, it is remarkable. I mean, it is really worth seeing that, watching that demo because you have this moment of, Oh my God. You know, he’s, he’s kind of in one sort of demonstrated kind of everything that we take for granted now, but then you’re talking about let’s move on to the clerk because we need to get into the AI a bit.

So tell us about this idea of, he talks about the clerk. So this kind of helper, that kind of helps you do stuff, right?

Molly Wright Steenson: [00:24:33] Yep. I’m looking to see if I can. I just published something in this book about Englebart actually, and about the mother of all demos, but yeah. You know that he would have this, Here’s here’s how he describes it. So, Englebart writes in 1962, he said, “this is the virtual architect. He sits at a working station that has a visual display, some three feet on the side. This is his working surface and is controlled by a computer has clerk, which. with which he can communicate by means of the small keyboard and other various devices. He is designing a building. He has already dreamed up several basically outs and structural forms and is trying them out on the screen.”

So why is Douglas Engelbart not an architect using architecture in 1962 to talk about what a computer system can do?

Andy Polaine: [00:25:21] That was my question!

Molly Wright Steenson: [00:25:22] What’s up with that? Well, there are a couple of reasons. One is, I think at one is opportunistic, right? Architects were a really good use case for the development of early computing. Symbolic logic, representation, inputs, and outputs, because you need all of those things to do design well. Right? You know, I’ve got to laugh because right now we’re in a situation where architecture and design schools are on Zoom, because working from home and we’re still grappling with making everything happen on a computer. And yet so much design does, right?

Andy Polaine: [00:26:02] Yeah.

Molly Wright Steenson: [00:26:03] The other thing though, is that architects build worlds. And I think that system builders, computer system builders, AI researchers, interface designers, early information, architects have all envisioned themselves as we’re building worlds too. And so that’s why it’s so attractive to look at architecture has your use case.

Andy Polaine: [00:26:26] But, I mean, equally you could have some of those things you just said about architecture also would have been true of say industrial design or, aircraft or automotive design and so forth. Why, why architecture? Was it just the kind of people he knew he was hanging out with back then? Or was it deeper than that?

Molly Wright Steenson: [00:26:42] I don’t think that actually, from a perspective of how the technology develops in those fields, that they’re separate. Mechanical engineering and aircraft design are where augmented reality comes from. And, you know, the, the complex engineering that you see today in Frank Gehry’s work, or Greg Lynn’s work comes from an engagement with, mechanical engineering and different kinds of modeling that take place. So I don’t think those are discrete kind of chunks of, of professions, the way that maybe universities see those things. they’re actually much more fluid, but it is way, way more sexy to talk about the building of a world than it is about the, the working of a b spline or, you know, how you make an airplane wing.

Andy Polaine: [00:27:33] Well, I don’t, it depends, depends on what floats your boat, but I can see, I can see how that, yeah, I can see that the attraction. There is also a thing though, in there about this idea of, they’re talking about this idea of, you know, architecture is also about the communication of information across networks. And so therefore architects have to kind of consider that, you know, the theory of networks and you just talked about academic institutions. One of the things that ironically stops, academic institutions being more interdisciplinary is, is the architecture, right?

You’ve got these people in this department, you’ve got these people in this department. So you’ve got these intellectual fields that are related. related to each other, separated physically. And of course famously, you know, MI T threw the, up in the air, some of it anyway.

Molly Wright Steenson: [00:28:16] Many or my own, my own university does too. I believe we have the top ranked machine learning program in the country, if not the world. it might be worth talking just for a second about what it’s like to do innovation within a university construct. Right. That, that, I think that the depth is necessary, that one gets into being specialized say in industrial design versus architecture.

But on the other hand, you don’t build really, really amazing groundbreaking things with people who are exactly like you, you need difference. And in, in this history that I I’ve written about, and that I’ve spent so many years researching about AI and the development of computational architecture, it was having really interesting problems to solve that made it attractive to design computing systems for architecture, to have architects involved in designing for AI.

Andy Polaine: [00:29:11] But despite all of the egos in architecture that you’ve talked about before, you also need a sort of psychological, safe space and some professional humility to properly work with people from other disciplines, right? You, you have to be interested in the way they see the world, not trying to fit the way they see the world into your existing box. Right? And, I think, you know, I, I’ve worked in academia for quite a number of years too, and I think that’s one of the things that’s one of the things I kind of always missed a bit from the commercial world is I think there is quite a lot more of that despite there’s a lot of jostling in academia, too.

That’s for sure. So going back, so Englebart has got his, this idea of this clerk that is a kind of a helper. And if effectively augmenting your, your intellect or sort of doing tasks for you, leaders on it. So that that’s the sort of, you know, I was gonna say that’s like the, kind of the Siri of, of his world, but it’s, it’s, it’s more than that.

And, lead us on to how you go from there into that’s the beginnings of kind of AI and humans working together, at least the idea of it, lead us to where you are now with AI and ethics.

So, I, okay. There’s, there’s another architect to be sure to, raise and that’s the work of Cedric Price You know, the book is essentially four chapters about white men who are architects who worked in technology.

I did notice that.

Molly Wright Steenson: [00:30:41] Yeah, well, I made a pretty, a pretty long mention at the beginning about, about the fact that there were women and, yeah, again, if I were going to write a book like this again, I don’t know that I’d write it. There are a lot of problems about the MIT media lab, and I think about the innovation culture and the culture of fundraising, At that university in particular that we all know about in the wake of Jeffrey Epstein, the sex trafficker, Joi Ito’s decisions, and some things that Nicholas Negroponte have said that make me not really want to pursue that line of research anymore or write about MIT anymore. But, you know, that’s, that’s a thing. I want to point out that Cedric Price plays a big role. And, if you, if you pick up Architectural Intelligence, it’s chapter five, he’s the secret patron saint of interaction designers. He is so much fun.

Andy Polaine: [00:31:34] It’s so secret that no one at the hardly anyone in interaction design might have heard of him. Do you think?

Molly Wright Steenson: [00:31:40] We tend to only, well, you’ve already said that we’re not very good at history, but, and we’ve already acknowledged that people aren’t good at crossing, crossing disciplinary boundaries. So this is my job. No Cedric Price was actually a very well known, British figure. And if, if we had been around in the 1960s, at least to be engaged with culture at that time. I mean, his wife, excuse me, life partner was Eleanor Bron. Yeah. Who is the, she was in the Help movies, but she was in the Beatles movies. She was in Help. And she was, yeah. Is the inspiration for the song Eleanor Rigby. He, yeah, he wasn’t known for buildings. He built, but he was known for saying things like “technology technology is the answer, but what was the question?”

Andy Polaine: [00:32:28] That’s right. You talk about quite a lot of stuff he came up with never got built, right?

Molly Wright Steenson: [00:32:34] Yeah. Yeah. It wasn’t about the building. it was about the thinking about the building. So he, he changes how we think about things. And one of the things, you know, to bridge into kind of this, this idea of how I think about AI and ethics, he’s sort of a way to, to make that bridge. I think that you need to consider that architecture and design and the work that we do in all of these different disciplines is the kind of instantiation of an ethical decision made at some point, right?

We make a decision, we make choices and trade offs. We. Have our biases, we do certain things and then they meet the road, right where the rubber meets the road, where they meet the road, thanks to something we do. And it makes me understand the role of designers and architects as being very vital from an ethics, perspective from an ethics standpoint.

And two years ago, I was awarded, an honorary professorship in computation and ethics here at Carnegie Mellon. So there are two of us who hold these honorary chairs. And I used it as an opportunity to look into what, what that means. And I, I kind of joked that there was the great ethics boom of 2018 and 2019.

Andy Polaine: [00:34:00] But now it’s now it’s gone away?

Molly Wright Steenson: [00:34:02] Well, it’s, it’s on the decline. I’m actually, I’m doing a study right now, with the, group called dSharp, which is our digital humanities group here at Carnegie Mellon working with Scott Weingarten, Matt Lincoln and Nick Martellero to see. How the terms, AI ethics and design intersect using a corpus of texts from Medium.

So I’m trying to see where AI is an ascending, a term, an ascending phrase we use when we start talking about ethics and, and where we don’t. And the interesting thing is if you just look at articles about AI, they’re about chat bots and tech, and you know, different ways to do things. But if you start looking at ethics, what starts figuring in are the words, technologist and designer.

So when we start considering the role of who we are, and not just the tech that we designed for is when we, as a field start talking about ethics. Maybe that’s really obvious. And maybe it’s really different because I think that what it might mean is something different than human centered design, like user-centered design. It’s almost like considering ourselves and our own morals and ethics within that design process.

Andy Polaine: [00:35:17] And why do you think, I mean, do you not think that’s been, we’ve considered those things in the past as designers?

Molly Wright Steenson: [00:35:24] Sure. And I also think that we…

Andy Polaine: [00:35:25] You sound so unconvinced by that..

Molly Wright Steenson: [00:35:27] Well, I mean, there are a lot of examples that, you know, are kind of the apocryphal examples, like, soap dispensers that won’t dispense soap to a dark skinned hand, or a, you know, a webcam that will follow a white face and not a dark face. That’s fine. Like I’m sure the people who designed those are perfectly fine human beings, but they didn’t think.

Andy Polaine: [00:35:48] Yeah. So ethics, is a deep and complex and a long and… out of all those things got the longest history of kind of thinking about it. It often is equated as kind of, “oh, we should do the right thing from the perspective of me as the designe and my moral judgments of things,” but it’s much, much more complex than that.

What’s your, how do you kind of, introduce this idea when you’re teaching, ethics and AI what’s the, what are some of the core ideas that you think people should, examine and get their heads around?

Molly Wright Steenson: [00:36:25] Well, that’s a really good question, probably, you know, a whole other, a whole other show. There’s the entire field of applied ethics and I’m not an ethicist. In fact, I consider myself an accidental ethicist. I think a lot of us consider ourselves accidental ethicists, right? We realize there are enormous implications for our work.

But if you look at something like the Markkula Center and that’s M A R K K U L A center for applied ethics. It might be really interesting for, for your listeners because they they’re based at Santa Clara university in California and they have curricula, problem sets, you know, ways to kind of think through, different approaches to ethics in say cybersecurity or software engineering. but. In my own classroom. So I teach the first semester master of design students, the seminar about interaction design and service design concepts, and actually my approach to service and interaction design concepts is to look at contemporary technology and to look at questions of bias and ethics and, business models between behind tech companies, what AI it does, what some history is of where these different ideas come from and get them to look critically at things that are far bigger than them, you know, It’s it’s complicated because you think that most of our students are going to graduate and probably work in tech, right?

So they’re going to be working on these huge systems. And, I think that it’s important to understand the workings of bias and representation and, the effects of, the, the racial and gender and sexuality effects of some of these technologies.

Andy Polaine: [00:38:17] Yeah. I guess the point I was getting to with this sometimes very simplistic view of ethics, which is, you know, “I need to just kind of do the right thing” as if there is kind of one right thing to be done. And yet my sort of, again, accidental understanding of ethics is much more, one of continuous inquiry. And in fact, all of the, all the hypothetical’s of ethics, you know, in that kind of Socratic dialogue way of talking about ethics, which is to give us an idea. And what would you do in this scenario? There’s always in those hypotheticals, there’s always then, and then you get given an addict piece of information, you know, like, you, you see someone walking down the street and they drop a $50 note, Do you pick it up and kind of give it back to them or not, and then you, I know they’re the right thing is to pick it up.

Okay. What if that person is, you know, billionaire robber baron, that’s a Donald Trump, -who’s probably a millionaire robber baron - would you still give it back to them? Oh, no, I don’t well I don’t know. You know, there’s all of these kinds of things where you kind of keep adding layers onto it and changing the situation and obviously at people’s ethics change all the time, because the thing about ethics is… it depends. Right? It depends on context. And so one of the things that I thought was an interesting sort of parallel, and I wondered if this is the one that you’re kind of drawing, which to do with scale, right?

So when we’re working on this stuff at scale, and, you know, classically, you know, 1% of Facebook users is several million people. So it’s very easy to say, well, that’s only 1% of our users, but it’s actually very many million people. And as we’re seeing in the middle of the Coronavirus, you know, small numbers turn out to be an awful lot of people There’s a parallel there with architecture too .

Obviously there’s lots of problems in say industrial design, where something has been created the automobile, let’s say it goes out in the world in sort of multiples of scale and cause all the climate change and sustainability problems architecture also has this kind of, imposition on human space and human interactions, in a, in a very big way. Is there a kind of parallel to be had between those two senses of scale and thinking about working at scale? I mean, architecture is also incredibly permanent, right?

Molly Wright Steenson: [00:40:23] Well, it’s the hardest problem about architecture is that buildings are one of the worst things for the environment. concrete is I forget the number. I have it

Andy Polaine: [00:40:34] Oh yeah. It’s one of the worst contributors, right?

Molly Wright Steenson: [00:40:36] 40% to, Yeah of landfill is, you know, of our dumps is, is construction material. Right now we’re in a position where our cities are empty. We are all in homes and our economic system has plummeted. what we understand the city will be, will be different. After this, we’re going to look differently when we’re able to go out again and it’s going to be a virus that’s done that. Now cities have done that to us before, right? Like you think of cholera, right?

There’s sir, John there’s there’s the SirJohn Snow story.

Andy Polaine: [00:41:10] Yeah. That everyone knows. Well, should know. I’ll put in a link to it in the show notes in case people don’t know.

Molly Wright Steenson: [00:41:16] But, you know, cities were redesigned, from their medieval models in part, because of sanitation, in fact, in large part because of sanitation and, you know, again, it’s, it’s sanitation because of disease.

Andy Polaine: [00:41:30] Yeah and networks, right? Because cities have evolved. I live very near Strasbourg in France. And so it’s a, it’s a city that was on, you know, real kind of hub on the trade routes, across Europe, certainly from sort of North to South of Europe and also where Gutenberg famously invented the printing press.

So there’s, there’s, there’s this kind of amazing history that goes backwards and forwards between Germany and France, throughout history too. But you kind of see how, you know, the cities evolve and change based on, based on comment commerce and the economics of it.

Molly Wright Steenson: [00:42:03] Well, yeah, based on commerce, but also I think the idea of networks and, and water have, interesting connections. So one of my areas, something that I also researched is 19th century infrastructure - information infrastructure. So I look at postal services, pneumatic tubes, and some of the earliest..

Andy Polaine: [00:42:20] Oh I used to love those pneumatic tubes.

Molly Wright Steenson: [00:42:22] Oh don’t even get me started gosh.

Andy Polaine: [00:42:24] So when I was a kid that there was a, it was a department store in the town that I lived in and they still had those things. So, so again, we’re showing our age here, but they used to, there’s still a few places that have the map on there. And so for people that don’t know, you have a little, I would say like a kind of tumbler glass size kind of pod and that the cashier would put a… how would it work? You’d put the, they will put some, some kind of information inside those. They’d put it in some, sometimes the tubes were kind of brass. The ones now are kind of plastic and it would be sucked up. It was like a vacuum tube. It would be sucked up and ended up sort of somewhere else where someone do something with it and know it gets sent back down in these and they are fascinating.

They’re so cool.

Molly Wright Steenson: [00:43:10] I have them, do I have them in here? Yes, I do. We do

Andy Polaine: [00:43:14] Oh, wait, I’m not ready. I’m not recording the video, but there we go. Oh, I’d have to have to get you to take a picture of them because it’s, it’s great.

Molly Wright Steenson: [00:43:20] This is yeah. They’re magic. in fact…

Andy Polaine: [00:43:24] I have to give away, you just sniffed that because it’s obviously got it some kind of amazing brass smell or something.

Molly Wright Steenson: [00:43:29] Exactly.

It smells…

Andy Polaine: [00:43:30] It smells of history.

Molly Wright Steenson: [00:43:31] Yeah. Yeah, it does. It does. That’s a nice number one on it. This is one that would have been used sometime from the 1920s forward inside of a building, but yeah, they’ve I did a, an interview with Roman Mars’s podcast years ago, 99% invisible about pneumatic tubes.

Andy Polaine: [00:43:50] I’ll look for it. And I’ll put it in the show notes.

Molly Wright Steenson: [00:43:52] To the point of what we’re, to the point of what we’re talking about, it, you, if you look at early communications, infrastructure, communication, and finance go together and, and affect how our infrastructures work and affect how our cities work and our cities are slow to change.

I’m I’m talking to you from a house of the future and it was built in 1890, right? You know it’s, it is a house of the future because it has information technology in it. And I’m, I’m able to talk to you in Germany, on this podcast because, because of information technologies that are kind of ubiquitous, right?

Andy Polaine: [00:44:35] So look, we’re, we’re coming up to time, well actually we’re sort of in, well over time, but it’s been fascinating. So let me, let’s bring it home, to, to people who are in the design profession or maybe, clients who are kind of, got client teams working on AI, they’re working on some kind of assisted technology they’re working on new products and services. What would be sort of a take home message, do you think, to those people?

Molly Wright Steenson: [00:45:01] I want to say that. You know, beyond our really big, wide ranging discussion here about the technology and history. And, there are things that designers do that make us design better technology and in better ways for things that have AI or that are AIs, right. Designers are really good at reframing questions. And framing questions in a bigger kind of way. It’s not a matter of this feature, this one little thing, but rather the bigger picture. I like the story of, that Shelley Evanson used to tell of you can, I can ask you to design a vase or I can ask you to come up with a beautiful way to integrate flowers and plants into my daily life. Design reframes the question.

So that’s one thing I think that designers are able to see implications that, that sometimes just engineering focused people can’t and, I think that designers can understand that we have a whole long design process that isn’t just operationalize and optimize, but rather that starts with research and concepting and iteratively prototyping before we come up with this implementation and tweak the implementation phase. And I have strong concerns about only, about tech companies only incentivizing shipping and deployment,

Andy Polaine: [00:46:19] Deliver, deliver, deliver. Yeah.

Molly Wright Steenson: [00:46:21] Yeah, where this earlier part, you know, again, back to Hugh Dubberly where we started, he has this compendium compendium of like 130 pages of design processes, the Dubberly Design Office. We have lots of different ways that we can design. And I think that it’s people like us that can remind that of our teams.

Andy Polaine: [00:46:40] That seems like a good place to, to end. The Power of Ten, as you know, is named after this Eames film, the famous Eames film, where they kind of zoom out. And, it came about because I was talking with someone, about, you know, this idea of in service design in particular of zooming backwards and forwards between how a small detail can make a large difference and at the same time making a shift at there at the top end of that zoom in society. And as we’ve seen with Coronavirus, right. It’s suddenly kind of changes everything else and it kind of ripples down into all the details. So the final question is what one small thing that is either well-designed and overlooked or needs to be redesigned and perhaps is overlooked, has, or would have an outsized effect on the world?

Molly Wright Steenson: [00:47:26] It’s hard not to think of the Corona virus situation and the fact that something as simple as hand washing, you know, you, you can hoard toilet paper and hand sanitizer, but if you can wash your hands, certain things are better. But I think the other thing is… I think we’re about to find it. And I don’t know precisely what it is, but we’re dealing with a point in time right now that’s unprecedented where we all have to stay home. And as one friend of mine, pointed out, people in her community are moving from grief into anxiety. And now, you know, seven to 13 days in for most people in North America we’re really starting to hit, a fearful time. And it’s gonna feel like this for a long time, especially if you’re in New York people, I think in, in China and Italy know this. So maybe there’s something that we had met now. That’s very small and very well defined that makes us feel less alone and it’s not Zoom.

Andy Polaine: [00:48:32] Well, it’d be interesting to see what comes out of this.

Molly Wright Steenson: [00:48:36] And, you know, it makes me think back to that website that I worked on in 1994 and the first online communities I was a part of and the BBSs before the web. The reason I had the epiphany than I needed to do this internet thing was that it was going to bring together everybody instantly all over the place.

And that was the promise. And so I hope that maybe that’s what we’re getting back to is some sense of connection through it.

Andy Polaine: [00:48:59] After pretty toxic era lacking in ethics, huh? So where can people find you online? We come to the reward for those of you who sat through and listened to us, talking for the last almost hour.

Molly Wright Steenson: [00:49:14] Oh my gosh. It’s what happens when you register a domain name and you’re 24 years old. You can find me at girlwonder.com. girlwonder.com is my personal site. My book is Architectural Intelligence, how designers and architects created the digital landscape and then Bauhaus Futures, is a look ahead from the legacy of the Bauhaus. And we ask if the Bauhaus were around today, what would keep it up at night?

Andy Polaine: [00:49:41] Right, quite a lot, I should think. So and what about, social media? You’re on LinkedIn. You’re on Twitter.

Molly Wright Steenson: [00:49:47] I’m on, I’m on the social medias. I’m on Twitter. I’m Maxie Molly, M A X I M O L L Y. And I’m also a Professor and Associate Dean at Carnegie Mellon university.

Andy Polaine: [00:49:59] I’ll put links to it all in there in the show notes. How come you’re not girlwonder on Twitter? Was that, was already gone or had you sort of grown out of that to by then?

Molly Wright Steenson: [00:50:07] I think it may have been already gone. Every once in a while someone tries to, to grab it from me. The funny thing is in the, from ‘97 to ‘99, we ran a pop culture, feminist webzine called Maxi. So my name is Maxi Molly from this kind of 22 year old vestige. It could go. It could go to college and get it’s own job now. It’s such an old name.

Andy Polaine: [00:50:31] It’s so funny. My… the best Twitter handle, I know somebody who’s from Kevin Cheng, who he wrote a good book called. See what I mean? And he managed to get to K just the letter K as his Twitter handle. It’s pretty impressive. Molly, thank you so much. It’s been very fascinating. Thank you for being my guest on Power of Ten.

And stay safe.

Molly Wright Steenson: [00:50:50] Take care, Andy. Thank you. You too.

Andy Polaine: As I’m sure you’re aware, you’ve been listening to Power of Ten. My name is Andy Polaine. You can find me at @apolaine on Twitter, or polaine.com where you can find more episodes and sign up for my newsletter Doctor’s Note. If you like the show, please take a moment to give it a rating on iTunes. It really helps others find us. And as always get in touch. If you have any comments, feedback or suggestions for guests. All the links are in the show notes. Thanks for listening and see you next time.