2014 Main Conference Talk
Like fish who take water for granted, organizations often overlook how language is a critical part of their infrastructure. We often think of content, metadata, names, or links, as stuff we put into the environment. But that way of thinking underestimates just how much language is actually part of our environment. What happens when we don’t take language seriously as a primary medium? Organizational confusion and incoherence: from planning to making, and from databases to interfaces.
Understanding how language is infrastructure is even more important now, with the explosion of mobile, cross-channel, and blended environments, across entire service experiences and touchpoints. The language we use as infrastructure is the information we use for architecture.
Some of what we will cover:
- How language works as the primary interface between people and complex systems.
- Why labels are more than just names we put on things, and how they change (or even create) the environments they name.
- How taxonomy is more prevalent than we tend to think, and how it plays an essential role in all kinds of environments and services–online and off.
- Stories and everyday examples of how language creates the “maps we live in.”
About the speaker(s)
Andrew Hinton is Senior Manager of Product Design at Honeywell Connected Enterprise, in Atlanta. For more than twenty years, he has led teams and organizations in creating great places, products, and services for humans. Andrew is also the author of Understanding Context from O’Reilly Media, and keeps a home on the web at andrewhinton.com.
Andrew Hinton: Language is infrastructure. I am going to start with an actual, practical example, it’s a simple example, but it’s one that’s not that different from something, I’ve encountered before. Let’s say, in this hypothetical situation, you’re doing user experience work for retailers in e-commerce site. The project you’ve been assigned, as what sounds like a clearly defined goal, “Make the shopping cart easier to use. People are complaining about it, can you make it better?”
“OK, great.” We all know what the cart is, right? It’s been around our websites for a long time, but is it really that clear cut? If you look at, for example, just the word “cart” What does that mean? It’s the kind of question that would seem to invite eye rolls and derision in a lot of IT meetings or planning meetings, or project kick offs.
You’re already sitting at a project kick off and the jerk piping up and going, “Hang on, hang on, hang on. What do you mean by cart?” People are all like, “Can you just not be all weird?”
Andrew: “We have work to do.” That exchange describes about half of my career, but here’s the deal guys. Joke’s on them, where’s the cart here? Here’s Target. There’s a little icon up there, a little cart, a little number on it. Look at all this other shit.
That’s just some of it, right? You’ve got the cart actually showing up in many places on the site, various forms. It’s enmeshed with all sorts of other functions like authentication, store choice, product availability, shipping information. The cart actually changes what it is, depending on what you are in a retail site for various reasons.
Some of them having to do with just how the infrastructure is created, the data and systems infrastructure, the services infrastructure. Some of it having to do with the steps and the things that users have to, customers have to deal with at certain stages in their retail work that they’re doing on the site.
Some of these things involve whole different sets of business rules depending on where you are in the process. Now, this might sound like a stretch but I’ve run into this very kind of a thing before. As a consultant, being assigned to lead a team for cart and checkout.
Guess what? Their cart and checkout team started at the cart page, right? Then went through checkout. But that meant that we couldn’t make decisions about all of the other instantiations of cart other than that one that’s at the bottom right. Now, a collaboration is fine and good, but the problem is that the way the organization was structured and the way the work was structured made it very hard to collaborate with other teams who were dealing with what cart instantiated itself as in other parts of the experience.
Not even going to mention the mobile team, which was like…they’re special. They’re off in some other area doing mobile things.
I can’t be the only one who’s encountered something like this, where you’re just trying to get to the bottom of what the nature of the problem is, what you’re actually trying to do.
The other issue being that you feel everybody around you thinks it’s so simple, but you’re looking at it and going, “Wait. There’s all of this complexity going on under these rubrics, these labels you are throwing around.”
The card isn’t just the card. It’s the signifier we use as a shorthand for a hugely complex cluster of business rules, infrastructure, not to mention interactions, behaviors, and all of these are involved or made of language.
The only way to work this out is to have a conversation about it, to unpack the parts, gain some consensus or direction on what the mission is. This is a simplified example, but I know I’m not the only one who’s had this scenario before, or something like it.
More often than not, there’s just not time or willingness to stop making staff to have a conversation like that.
Design is obsessed with the tangible for good reason. Everything else feels secondary and peripheral to these objects that we can see or touch. There’s so much to think about with things like icons, GUIs, or Internet of Things widgets and gadgets.
It’s all very interesting and necessary, but here’s the thing. Too often we don’t talk about what’s underneath all of that, what ties all that stuff together. It’s the context that makes all these physical and virtual objects valuable or meaningful and the stuff that connects them and defines them.
A phrase I hear a lot is, “That’s just semantics.” You know what? Sometimes it’s deserved because sometimes, mea culpa, I’ve done it, it’s a lot of fun for some of us to get into these philosophical semantic conversations-arguments.
The thing is that all of that noise, it seems like noise, it’s actually really hard work. It feels playful because sometimes it’s fun, but it’s actually people as an organism figuring out what things mean, what things are so that they can do better, so they can grow.
This is the problem right here. It’s the stupid little word “just.” Say, “That’s semantics,” not, “That’s just semantics,” because semantics isn’t small. It’s not inessential. It’s not deviation of purpose, but that word, “just,” diminishes it.
When we hear that, it’s basically someone saying, “I don’t want to deal with that or have patience for that.” It’s on us though to do a better job of having those semantic discussions productively and to do a better job of that, to include people people and those in a better way. But still, one reason why we have so much trouble with this is because it’s like a fish and it’s relationship with water. We’re so immersed in language, it’s such an native medium for human beings, that we see right through it and we take it for granted. But it is a medium. It does have properties that we should understand.
I do a lot of talks. More often than not, my talks try to cram 20 different theses into one talk. Today it’s one thesis, and it’s this, so you’re going to hear me talk about this 20 different ways. Because here’s the thing, I think as a community, and it finally occurred to me the reason why information architecture has had such a hard time saying what it is that we do, or what our job is. Because it’s invisible to us, but our job is to make it visible.
Really, my talk, as I said, just has the one message. That’s that language is a real medium for making, and semantics isn’t fluff, it’s not nonessential. It’s really crucially essential and central. It’s not a disembodied pair of abstraction that we optionally add to the objects, places and systems that we design. It’s instead both the body and soul of those objects, systems and places.
In particular, language is what we use what we design the structure of information environments. That is, when we do information architecture. Language makes structures that are real and critical to our lives, just as these walls, floors, roads and bridges. I’m going to talk about this in several loose categories, and we’re going to start with organizations and how semantics function is a part of the environmental organizations. Some maps aren’t representations of physical landscapes, but they’re new environments in and of themselves. This was made around a month ago, and I guess it’s the first real ore chart.
It’s beautiful, it’s from the 1850s for Erie Railroad, and it’s got a beautiful Victorian thing before the mechanistic metaphor is totally ate management. I urge you to go find it. There are huge scans of it out there. But either way, ore charts are representations of social relationships, processes, collaborative structures. This railroad company spread out geographically over space and time zones. They needed an architecture the virtually organized all of their relationships, their work. An augmented reality architecture for existing as a corporate entity.
When we create requirements or user stories in organizations, we’re putting a lot of weight on language to define the shape of what we want to produce. These stories are supposed to be quick, do them quick. But you aren’t writing crappy cards, right? Because you’re not really talking about wait, wait, what is that? Or at least get around having that conversation, because what happens is that’s a mold that creates a thing in its image.
Once that thing’s made, it’s really hard to unmake it. 10 years ago at this conference, Jesse James Garrett talked about how an organization’s brand depends on the foundational meanings of the language it uses, and how the structures we make need to take that language seriously.
We’re still figuring this out. In the classic new management advice book “The first 90 Days”, the author stresses that to work effectively has a manager or executive, you have to understand the cultural moors and fundamental values of the organization. Management has much to do with the direction of the company as the budgets and reporting employees. Culture is a critical business concern, and our interface of that system is language. Organizations also have collective memories and share narratives, and the arbitrary pathways of the past are often remembered as a linear story that can be very convincing, because it leaves out all the paths that were taken.
That story can influence future decisions. And likewise, even on a project level, that underlying story ends up being articulated in terms of interfaces, right? We need to add another tab to our global nav with my department on it. I need my thing at the top of that thing, right? Whether it makes sense or not. We need an app, we need to be more like Pinterest, I’m still hearing that one. Often, it’s the best way stakeholders know how to talk about their issues, is through that. It’s our job to help them not step on that landmine, right? It’s our job to say OK, let’s talk about the story underneath all that.
Language plays a critical role, both vertically and horizontally in organizational life. Vertically, it starts as deep as the social structure going all the way through to the software services work. The business rules that the software services are putting out in the world. The corporate languages, which is anything from the colloquial way that the organization talks to their world, to the way that one employee that I’m bringing in now, when they get trained, they get trained on all of the crappy systems and the company first, so there binder is organized, the table of contents is organized by crappy system one, crappy system two, crappy system three…I would say the names, but that would be out of NDA. That’s just a variable.
In other words, it’s creating a map of what that organization is that sticks with that person all the way through their career. How’s that getting on the right foot?
I’ve been talking about language like it’s stuff, it’s out in the world, because it is. It has a tangible effect on the work we do. It is environment.
Language is physical. We make it. We consume it with our bodies. When we speak, we’re using a great deal of our body to breathe, create the sound vibrations needed for articulation, not to mention gesturing like I’m doing right now, not on purpose but I’m just doing it.
When we write, we’re adding physical information to the environment that we depend on being picked up and interpreted correctly by a reader. We learn to read by reading aloud.
Even when you read to yourself silently, when you get good at reading, science shows us that our body is still subvocalizing. Our body is still firing neurons in doing things that it would do if we were saying it out loud.
There’s really no such thing as reading truly silently. Language is physical. There’s mounting evidence that language’s been with our species long enough to be a factor in natural selection, part of the environment that actually shaped our evolution.
For example, anthropologists have shown there’s a connection between language use and the ability to create sophisticated tools and weapons that can’t just be figured out just by watching somebody. There has to be some kind of collaborative effort.
This stone arrow tip that was found as one of these studies, dates to many thousands of years before the actual emergence of Homo sapiens. Language has been with us since before we were us. Language is environment, it’s stuff that we put into the world that allows us to make more stuff, including whole civilizations.
Philosopher and writer Andy Clark, in his book on embodying extended cognition, talks about how language is a kind of scaffolding. We create it so that we can go on to create more environment together.
Likewise, names and categories allow us to rearrange the world magically. We could take all this stuff in our world and just categorize it and move it around in the abstract, but it’s not really that abstract because it’s a part of our world.
They add new structures to the environment, new objects, new places. Labels are powerful things. They’re like a tardis. They’re innocuously small and mundane on the outside, but immensely powerful, mysterious, and huge on the inside.
Labels carry with them much of what the world is to us. They allow us to rearrange it, move whole universes of meaning with a breath or a scribble.
When we put labels in the structures, such as lists, we create new parts of our environment that enable new kinds of human activity. As Eco has said, it is the origin of culture.
In archaeological digs, most of the lists of the common source of writing found are about commerce and mundane record keeping. Why? Because writing was a necessary infrastructure for markets to expand over space in time, past the natural limits of human memory.
Poets didn’t need to write stuff down. They memorized it and they were out talking and saying it. It was these flunkies that were just trying to get work done. They were making language writing what it is.
As Patrick Lambe says, in this really amazingly wonderful book that is expensive but you should all buy and read, “Organizing Knowledge,” taxonomies provide the lenses by which we perceive and talk about the world we live in.
Keep in mind that taxonomies aren’t just hierarchies. Taxonomies can be lists, spectrums, facets, polyhierarchies, all sorts of arrangements of language that create environmental structures for action and understanding.
The reason why language can behave as environment is because it provides us a kind of a affordance. James J. Gibson, who invented the concept of affordance, talked about how the structure of the surfaces and objects of our environment has a direct, mutually coupled relationship with our bodies.
That’s distinct from consciously uncoupled, which is a whole other thing, involving coldplay or something.
Andrew: This is mutually coupled relationship with our bodies, which pick up information about what our bodies can do with those structures. These stairs are picked up by our perceiving bodies as a structure that will take us upward in space if we climb them.
We don’t know where it’s going to take us. We don’t know the contextual significance of the stairs, unless we go up them and find out for ourselves or unless somebody tells us.
The intrinsic affordance of the stairs is supplemented by a label, a semantic affordance. There’s literature out there about semantic and linguistic affordance. There’s different takes on it, but this is the central issue.
Donald Norman calls these signifiers. He’s worked for years, trying to untangle the tangle that happened when he tried to use Gibson’s affordance concept, or did wonderfully use Gibson’s affordances concept for the design world.
I agree. These aren’t signifiers. That’s a signifier, but its signification is a species of affordance. It’s still structured, it’s in the world. We’ve just had to learn what it means at one degree removed.
Gibson, when developing the theory of affordances, explained that affordance isn’t just about simple perception between a body and a surface. He argued that affordances between people, because of language in particular, create extremely high levels of behavioral complexity in meaning.
These affordances coalesce to become compounded in variant structures in our environment that are cultural constructs rather than physical ones.
He talks about how a mailbox affords mailing a letter to another place. The intrinsic structure of the mailbox is just to hold that object that affords putting things into, not even just letters. If you’ve got little kids, you know…peanut butter!
Andrew: For people encultured in a complex system of postal mail, we see right through just the mere physicality of the subject. We see it as something that does this really complex thing.
Of course, that construct allows us to completely abstract those affordances into something made entirely with bits and pixels. Here we have an even more complex system, made entirely out of representations and metaphors of that box. It’s made entirely of language, it’s not just depending on it. Software, that’s what we made.
Placemaking and sensemaking are two words that keep coming up in our community. Good, because we need to keep working with these things and what they are.
When you see a city scene like this, you see placemaking happening everywhere, through the signs, labels and buildings, but also through the conversations, publications, arguments, the narratives.
In one sense, I get Clark’s metaphor about language as scaffolding. When I see this, what I really see is almost the opposite. I see culture and language that needed all of this stuff built to support it. It came first.
This is the scaffolding as far as I’m concerned, the built stuff. All that it takes for humans to make something into a place is to say something about it, to name it or mark it in some way.
There were names for craters and mountains on the Moon, long before anybody walked on it. When the first human set foot, they marked it with this flag. There’s no text on it, but it’s still a rhetorical act. It’s a semantic utterance.
The official line was, “We come in peace for all mankind,” but the subtext was, “Hey, Russia! Look what we can do!”
Andrew: It’s a wonderful irony that all of these flags have bleached white now.
Digital networks make semantic environments a super-charged geometrically more immersive and influential infrastructure that ends up changing the physical realities of behavior and place.
Yelp adds a meaningful layer to the environment that changes the sort of places that these are, as a real physical effect that increases businesses to independent restaurants. People don’t feel like they’re risking anything to go.
They know something about it, the change their known quantities. Why go there? Why go to Olive Garden if I can go to this wholesome mom-and-pop Italian joint?
In bit-based digital environments, we have very little physical structure to help us clarify what language means. We depend on the language for all of our structural affordances. The stairs to the poetry room and the poetry room itself end up being conflated into these two-dimensional bits of language.
This was the generative problem that spawned web-oriented information architecture. How to organize space where everything is made of language without the intrinsic structures of physical stuff to guide us.
When labels don’t make sense in relation to one another, when the labels, connections, and rules are muddled we’re disoriented. Sense-making is stymied. Place-making is dissonant. Here is the Facebook mobile app. Just look at that for a minute and see if you can figure out what those things mean.
We have a, “Hey, yo dog. If you like newsfeed, I’m going to put a newsfeed in your newsfeed.”
Andrew: I apologize for using a five-year-old meme but I think it’s apropos here. Why is there a newsfeed in the newsfeed? If I select most recent I’m still in a newsfeed. What happened to new newsfeed feed?
Andrew: Is it news new, but most recent is different? I’m going to see all my friends and some sort of like, weird not-chronological order? These make no sense.
Andrew: They just don’t. Actually, like a week ago I noticed it disappeared off the app. Now you’re locked into newsfeed which is some mysterious Facebook algorithm. I have no idea what the hell they’re showing me. At least I had some control here once I figured out what these things meant.
These labels are not structurally clear, but what this is, is a symptom of an organization. I know wonderful people work at Facebook. I understand. But we all work at organizations that get this stuff wrong. They’re like, “Here’s a little silo of functionality. Here’s a little silo. Here’s a feature. Here’s a manager or a director that needs to get something done this quarter so that they can get their bonus. I want the most recent thing in there. I want the thing in the thing.”
You just jam all this Frankenstein crap together and it just doesn’t make sense. Nobody talked about what is this. In Google+, which has gone out of its way to be very clear about context and the rules of privacy, there’s still some wrinkles.
If I’m in my family circle, which is blatantly constructed like it’s a room, like a place, if I go into the family circle on the desktop and I’m reading my family circle and I post, it automatically goes to the family circle. It makes sense, right? I’m in this room. I’m talking to you. You’re talking to me, or watching, whatever. We’re in the same room. I don’t then say something and then it’s suddenly somewhere else.
On the mobile version it takes like six interactions even if I’m in one circle to post something in that circle. If I’ve gone from one to the other it’s just all kinds of confusing. Again, what this sort of comes down to is there are lots of requirements obviously that got written. There are genius engineers in this organization and some amazing designers. But the architecture is fractious. The architecture is dissonant.
The ways we enable and influence sense-making and place-making aren’t just about a single website or app. It has to do with all the context that we encounter in something like this grocery store service experience where you’ve got an ID card that’s both digital and physical. It’s connected to other information about you that ties all of this stuff together.
It has to be surfaced to you in a particular way so you understand what it is. You can go and find a specific kind of peanut butter on a specific aisle in a store. You can create recipes. All of these things working together, all this machinery, it’s all language.
Consider the whole ecosystem of information environments that make up the whole retail customer experience. All these conversations, all the labels in the store and the organization there, and the expert reviews, and price comparisons, all of that, language. All of this connects to systems. All of this is part of complex things in the world, whether we made them or not. Most of what we deal with everyday is stuff that we made. It’s on us.
Let’s talk about systems for a minute. Think about how the staring game works. If I just walk up to one of you, I was going to just walk up to somebody and start staring, but I’m afraid I would actually trip off of this thing or something. Let’s pretend I just walk up to somebody an start starting. It would be off-putting I’m guessing. I’ve discovered this, by the way. It was hard to unlearn doing that.
If you just walk up to somebody, “Hey, let’s play the staring game.” “What is that?” “Well, we have to stare and the first one to blink loses.” “OK.” Suddenly there is an architecture, a simple little architecture in between you that’s a structure that you’re both living in together.
All games work this way. In fact, all human systems that we create together work this way. Think about baseball. That fact that all of this activity in baseball is guided by a collective agreement to follow a set of rules. There are objects like bases, but almost everything else is just drawn on the field. It’s just language.
There are very few physical structures other than that. There’s no rails that you have to stay on that connect to bases. There’s no big robotic arm that takes the failed batter and plops him back in the bullpen. They just go. It’s like there are these structures that you don’t see, but everybody just assumes they’re kind of there. This is how civilization works.
When rules change without everybody being on board by the way, bad things can happen. This is Sweden in 1967, the day after a law passed that switched the side of the road. Not everybody knew apparently.
Andrew: This is basically what happened at Facebook when they change their privacy stuff. Jon Colman, I owe you a drink if you’re in here. I’m sorry I’m beating up on your beloved employer. I really apologize.
Even our economies depend on the way that we define linguistic structures and objects that are a part of that system. Money is made of language. It’s just agreements between people. It’s just people saying, “OK. That’s worth that. Sure.” Then we abstracted it into currency.
Just the other day the US Internal Revenue Service ruled that they will define Bitcoin as property, not currency, which literally changes what it is. It literally changes what we can do or not do legally, structurally, systemically with Bitcoin.
Now that hideous chunk of content on the right is the official IRS notice. It’s hideous because it’s not really written for consumers. It’s new machinery added to the vast cultural machine of documentation that has to be there to sort of shore up all of this activity. It has to be articulated. The law is essentially language machinery.
What we’re creating is new legislation. We’re creating new structures, new places. We poo-poo stuff like this, but at least somebody is sitting down thinking about the what the hell this stuff means. Maybe just to cover their ass, but they’re thinking about it. Actually being able to articulate what something is, extremely important.
Business rules, systems, maps, environments, they all depend on or are inseparable from language. This means that we have to take language seriously and understand the nature of this material and the infrastructures and influences it creates.
I foresee a near future where design schools, programs, or mentorship or whatever, involves talking about linguistics and semiotics. Not just to sit around sounding smart, but I didn’t study this stuff really in school. I’ve had to pick up on it since because I started realizing I need to understand how this stuff works. The clay under my hands, what is it? It’s doing weird things. I’m not sure why.
I’m assuming things about it that it ends up not being. The same way we need to understand how CSS and HTML relate to each other, how a data stack works, or a presentation layer is connected to it. This is what this is. It’s the stack of language. Ultimately all of these things tie back to it.
I’m going to finish with three thoughts to sort of riff on this whole thing. First thought, modeling is making. In a previous article I wrote on this topic I argued that planning is making. I still think that good planning is making, but that’s a harder argument to make in a two minute slide.
Good planning involves modeling. Taking tacit abstractions, complex entities, and dynamics and making them into objects that we can work with in time and space with our bodies. Good modeling is actually prototyping. It is. Everybody thinks that prototyping isn’t prototyping unless you’re making something that looks like a product. But to people that make products, this looks a hell of a lot like making a product.
After awhile I’m assuming you get to the point where you’re like, “Oh, we keep making crap. Maybe we should think about it.” This is important. You’re prototyping your understanding of the problem, and eventually the solution. It’s not valuable in and of itself. It’s valuable as a way to achieve a collective understanding and a shared vocabulary.
Scaffolding for collaboratively knowing. This is hard work. It is design work. Whether it’s a month or workshops or a five minute conversation over a bar napkin, it is important work.
Second thought, business rules are everybody’s responsibility. All of these symptoms are built up from what companies call business rules. I want to say clearly that these are your responsibility no matter where you are in the line of the work that’s being done.
If you’re the wireframe flunky who is just like, “Don’t think. Just take these requirements and make a thing.” Figure out a way to ask. “Hey, why this business rule like this? Because here’s what going to cause to happen. Maybe we could rethink that.” Even if UI is your focus, this is a conversation that’s important. Not just for that project, but for your career.
As our practice matures we need to be able to participate in these discussions that determine the nature and overall direction of our work. Business rules are a story the business is telling itself. It’s often under us to dig underneath those stories and figure out what’s driving them.
Last, there’s really nothing that matter to humans that isn’t somehow wrapped up in language. The ancients who first used the word poet were using a word that meant maker because they understood that the poet makes more than poems. The poet makes worlds. We are a linguistic species. Grit and gristle, halos and horns, we’re immersed in symbol and suffused with story. It’s language all the way down. Thanks!
David: It’s just a quick plug in a sense. I’m glad that your diagram had pragmatics on the outer circle.
Andrew: I stole that diagram. Whoever I stole that from gets credit.
David: It’s awesome. Too often in linguistics it stops at semantics. Linguistic anthropology in particular, that’s the plug for an entire discipline.
Andrew: David is plugging linguistic anthropology, just to make sure you’re clear.
David: It goes really deep on pragmatics. I think for us, as user experience information architects, whatever the hell we’re calling ourselves today, that pragmatics layer is where it’s at.
Andrew: That’s where it’s at. Tell us what pragmatics is.
David: It’s language use in context. By observing language use in context, contextual inquiry, ethnographic interviewing, we can get a better sense of meaning. When we try to communication we can actually align our meaning to the meaning of our audience through studying pragmatics.
Andrew: I just want to riff on it a second. It wasn’t actually a question, but I’ll pretend like it was.
Andrew: I’m kidding. David and I go way back. We’re good. That’s a great point. Here’s the thing. Most classic IA method doesn’t acknowledge that the context of the language changes what it means. Card sorts are done out of the context of the actual work that the semantics stuff is intended for. It doesn’t mean you can’t learn something from them, but you can go really wrong with it. You should depend on it entirely.
There’s lots of other situations. The mental modeling method that a lot of us have used, takes a lot of the semantics the we hear in interviews and brings it into a vacuum where we figure out a platonic clustering. The truth is that actually has to work for people coming at it from all kinds of different pragmatic angles and contextual behavior.
That’s a personal thing that I’m obsessed with, how do we evolve these tools that we’ve been using for the last 15 years to acknowledge. Because we didn’t have to think about this so much, because everybody was sitting in front of a desktop computer. We should have been thinking about it more, but you do what you do. Now we have to, we really have to. Anybody else?
Abby: It’s almost like Andrew should write a book about something like this.
Andrew: About this context thing? I think it’s just a buzzword.
Abby: That could sell. We have another question here.
Audience Member: Thanks for the great talk. Suppose I am the CEO of a major information providing company, and I come to this lecture and I drink the Kool Aid 100 percent and I said “Andrew, I’m going to put you in charge of every single thing having to do with data in my organization. What are you going to do?”
Andrew: Honestly, I’m going to say that’s a bad idea. Because it depends how you do it, but information does not seem to me to be something. It’s like having a chief of innovation, it’s just stupid. It’s ghetto-izing the thing.
Audience Member: Don’t dodge the question. Tell us what this really means.
Andrew: It’s not a dodge. I’m not being a dick. That kind of decision, use all the work chart. That’s an infrastructure decision. My first call as that hire would be, let’s rethink what you think this is because what we need to do is actually educate the organization in thinking architecturally and thinking linguistically, and having some rigor and discipline about what they’re doing. It’s an uphill battle, though, because everybody’s just trying to get stuff cranked out, right?
We have a hard time just getting people to sit down long enough to do a usability test, much less having a meeting where we are like “What do you mean by cart, here?” I don’t know if I’ve got a great answer to that, other than that would be a conversation to do some hard work, scribbles and modeling on the white board. You have to sit down with that executive and go “Hey, let’s figure out how to solve the problem that you brought to me in the form of the solution.”
Audience Member: Thanks.
Abby: If anybody does want Andrew to work on their information, I believe you talk to Bob Rice in the understanding group about that.
Andrew: [laughs] We didn’t pay her for that plug.
Audience Member: You touched a lot on the similarity of language, how something can go across ages like the spearhead. But there’s also many changes. I saw an article this morning where an airline employee who had to give a note that someone was deaf, deaf and dumb, which offended the person and now the airline is going to take action on that individual because it is a disparaging term now. But the word “dumb”, the original word for dumb was “mute.” Now it has changed to something where it’s simple or stupid, but the person was actually using the word in the original context.
I guess in the bit world, you have the disc as a save, and we have 23 year olds in our shop that have never seen a disc. How do you account for the change of not only the visual language, the semiotics where you’re using things that have no context for people now, and the actual changes through all of the layers of language?
Andrew: I don’t have an answer for that that would fit the time that I have to give an answer to that, other than I feel like part of the curriculum of design and part of the design thinking and all that other stuff, it needs to be a little smarter about meaning. Again, it used to be that big companies had a lot more control over the context in which their communication is happening.
Most of it was one way. Now, it’s gotten more complicated. I don’t know what else to do other than to say you keep developing. Just like we have other models we can point to that say a color wheel or whatever, we’ve got stuff. We need stuff about this, and we need some surface understanding from people to think about it.
That original issue about the deaf and dumb thing, I don’t know what to do about that other than cut somebody a little bit of slack and just apologize. The save icon issue is really, this whole flat vs. square whatever thing is really about semiotics. It’s really about language affordance and how language affords things. It’sreally not about aesthetics or tastes, but that has been the red herring, so I think that’s what that is getting at.
Abby: All right. Last question, Peter.
Peter: I don’t know if it’s much of a question, more than a comment. First, thank you, this was excellent. The language all the way down, and I love Yertle the Turtle, it’s probably my favorite Dr. Seuss story. I wrote a post about how UX shunted information architecture, and it was probably mistitled. It should have been how IA stunted itself by falling under the umbrella of UX.
I think language all the way down into your chart addressed two parts of that in my company. I work a group on where people sit is determined by our taxonomy. That taxonomy was not meaningfully constructed at the outset, it was just thrown together really quickly because we needed to scale crazy fast.
Now that we’re trying to overcome some of the limitations of our taxonomy, it is so deeply embedded in the core operations of our system that it’s actually really difficult. Just wanting to basically support what you are saying and encourage folks, to encourage your organizations, particularly early on to be exceedingly mindful of what Andrew has been talking about, because it has real world, honest, deep impact, particularly if you find yourself succeeding.
In this community, the IA community has an opportunity to influence that impact that I don’t think it’s really taken advantage of because it’s gotten too associated with interfaces, and not enough with those organizational realities that you referred to at the outset.
Andrew: Thanks for that. No, that’s totally true. This wasn’t all clear to me from the beginning. In trying to write this book that Abby was referring to, which I’m hopefully finishing sometime this year about context. That it really is all for humans, it’s all bound up in language. When I started, they gave me these X-Ray specs, and then when I started seeing things like you were talking about, I was like “Oh my God, it’s everywhere.” What’s that movie with Roddy Piper?
Audience Member: They Live.
Andrew: Yeah, it’s like They Live. It’s like that. You put them on and you’re like “Oh shit, there’s these crazy things everywhere that are making people do weird stuff, and they don’t see it. They don’t see it.” And sometimes in order to do good work, especially information architecture work, you have to make them see that stuff so they understand how much crap they really need to bust down and redo. They can’t just layer new semantics on top of it, right? That’s not going to fix it. You sure as hell aren’t going to fix it with UI. Thanks for that point Peter, that’s great. Thanks everybody.
Abby: That was amazing, right? Did I promise amazing? Can we get louder? Woo! Yes, we can. Awesome, thank you guys.