Welcome to Last Minute Lecture.
This free chapter overview is designed to help students review and understand key concepts.
These summaries supplement, not replace the original textbook and may not be redistributed or resold.
For complete coverage, always consult the official text.
So if you're setting out to design a solution today,
the default, the knee -jerk answer is,
it's almost always an app.
An app, right.
And why not?
We live in a world where the number of apps is, I mean, it's counted in the millions.
But if you just stop thinking about the smartphone for a second, the whole universe of interaction design is just, it's exploding.
It's way beyond that little glass rectangle.
That's so true.
And our mission in this deep dive is really to systematically catalog that enormous diversity.
We're looking at a full spectrum.
I think it's 20 distinct types of interfaces.
It ranges from the decades old command structures right up to brain computer interfaces.
And we need to understand not just what they are, but what the unique design considerations are, the research questions for each one, and maybe figure out if this whole idea of a natural user interface is even achievable.
So we're moving way beyond the classic desktop metaphor.
We're going from like typing abbreviations on a black screen all the way to systems that have embedded AI and can read our behavior.
It's an entire catalog of possibilities.
And we are going to unpack it all, starting right at the beginning with, well, the foundation,
the desktop legacy.
Right, so we have to begin with the most basic form of digital interaction,
the command line interface.
The CLI.
Exactly.
This requires the user to type abbreviations, like using alls to list files at a terminal prompt.
Now the efficiency advantage here is it's profound, but really only for expert users.
Yeah, I always picture CLIs as something for system administrators or hardcore coders, but our sources point to this fantastic kind of unexpected application,
accessibility.
Tell me more.
There was a system called Text SL, and it allowed visually impaired users to interact in virtual worlds like Second Life.
Oh, wow.
Yeah, just by typing simple commands, they could make their avatar move or wave or even smile.
It really highlights that efficiency isn't just about speed.
Sometimes it's about the most direct path to just participating.
And if you're designing one of these, the core principle is consistency.
This goes all the way back to 1980s research.
If you name commands with the same structure,
experts can move at lightning speed.
They don't have to remember all these arbitrary names.
Okay, so that sets the stage for the big shift, the graphical user interface, the GUI.
This is the legacy born from concepts like the Xerox star, which gave us the original wimpy elements.
Windows, icons, menus, and the pointer.
Exactly.
And Windows were, well, they were purely a response to the physical limits of the screen.
They let users multitask, layer information.
But they also created a brand new problem.
Window management.
The research shows that users are constantly switching.
I mean, on average, a window is only active for about 20 seconds before the user swaps to another one.
20 seconds.
That constant swapping has a huge implication for focus.
A profound one.
Yeah, it means designers can't assume you're paying sustained attention.
And we see techniques trying to help, like on Mac OS where windows shrink down, or browser tabs that highlight a little thumbnail view.
Anything to make finding the right thing instantaneous.
And while we're in this GUI world, let's talk about forms.
Because I think everyone has felt the pain of a badly designed form.
Oh, absolutely.
I'm thinking of those ridiculously long scrolling menus when you have to pick your country.
The country picker.
Right.
If you live in, say, Zambia or Zimbabwe, you are scrolling forever.
Exactly.
Designers really should opt for predictive text, or at least group the options in a structured table.
Maybe use letter headings and some shading.
It makes scanning so much quicker.
It's such a huge lesson.
Friction, and even the simplest thing, drives users crazy.
So next up is menu design.
It's moved way beyond just simple dropdowns.
You have things like the mega menu, which is really popular in online shops.
Where everything just opens up in one big panel.
Yes, exactly.
It presents many options in a single large two -dimensional panel.
And usability testing heavily favors these because the human eye is just much, much better at scanning a wide array of options at once than trying to drill down through nested layers.
But the core difficulty is still categorization.
I mean, think about trying to find a simple command like sort in a huge program like Microsoft Word.
Where would that even be?
Who knows?
Is sort a table function, a tool function?
That ambiguity just shows how tricky it is to group things in a way that feels intuitive for every single person.
And finally, icons.
They originally relied on that desktop metaphor, a trash can for delete, a folder for grouping things.
And they've evolved so much.
From that detailed photorealistic look.
Right, to the flat 2D icons we see on smartphones today.
Or even just the simple grayscale indicators on a smartwatch.
So how do they work?
They succeed when they follow one of three mappings.
Isomorphic, which is a direct image, like file icon.
Analogical, like using scissors for cut.
Or arbitrary, like the X to close a window.
And when you look at how they work together, like the editing features on an iPhone footer, you see this blend.
Totally, you have a square with extended lines for cropping, which is kind of isomorphic and analogical.
Then you have the three overlapping circles for lenses or filters.
And my favorite, the little magic wand for auto enhance.
Right, just make it better.
It's a simple iconography, but it's communicating some highly complex software functions.
Okay, let's shift gears completely.
Let's move into type three, multimedia.
This is all about combining graphics, text, video, and sound with interactivity.
It follows what's called the multiplicity principle.
And the goal is always duper engagement, better learning.
Right, I'm thinking back to those early multimedia encyclopedias.
Exactly, where you can read some text, then click to see a video clip, or hear an audio file, or jump to a hypertext diagram.
We saw it in systems like the Cardiac Tutor.
And it's still around.
The Kids Discover app uses a sliding comparison tool, so you can slide back and forth to compare, say, the Coliseum today to a historical rendering of it.
But there's a big aha moment in the research here.
When you give users unlimited access to dynamic content video animation, they tend to channel hop.
So they just watch the flashy stuff.
They do.
They often ignore the static text or diagrams, even when that's where the critical information is.
So that's a huge challenge for learning design.
A major one.
To encourage more systematic engagement, designers use something called dino linking.
This means explicitly connecting content across different windows, so an action in one place directly updates information in another.
Or you just have to force users to complete structured tasks.
That too.
Okay, moving to type four, virtual reality, or VR.
This is the definition of immersion, right?
Creating the illusion of participation in a synthetic environment.
Yes, and the key psychological state that designers are chasing is presence.
That feeling of truly being in the virtual environment.
High fidelity systems, like flight simulators, are all about maximizing this feeling.
And we experience VR through two main viewpoints.
That's right.
First person, like in many games, where the camera is your eyes, and third person where you see your avatar on the screen.
And we've seen VR used for some really powerful things, like generating empathy.
For sure.
The We Wait experience, which used these deliberately sparse polygon graphics to put users in the shoes of refugees crossing the Mediterranean.
That brings up a really fascinating research question.
The tomato presence concept.
Does the world have to look completely realistic to maintain that sense of presence?
Or can more abstract, symbolic things still convince you that you're there?
The research suggests that fidelity isn't always the top priority for immersion.
The mind is really good at filling in the gaps, so you can maintain that sense of presence even with non -realistic objects.
Hence, the tomato presence challenge.
Okay, so let's move into the mobile world.
Type five, website design.
This area has had a fundamental shift.
We move from these really strict usability rules toward a balance of aesthetics and function.
Yeah, Steve Krug absolutely nailed the user reality.
He said, designers think they're writing great literature, but users are seeing the site like a billboard going by at 60 miles an hour.
They scan, they don't read.
Exactly.
And as sites get bigger, wayfinding becomes critical.
That's why we see things like breadcrumb navigation.
Right, the little trail.
It uses that Hansel and Gretel metaphor to always show you where you are in the site's hierarchy.
And of course, today, the standard is the responsive website that just automatically adjusts to whatever screen size you're on.
But it's not always about pure efficiency.
Take a brand like Nike.
Sometimes, they prioritize this big cinematic experience over traditional usability principles, like Instone's classic questions.
Where am I, what's here, where can I go?
And for some brands,
that cinematic, slightly mysterious experience,
that is part of the perceived value.
That idea of context brings us right to Type 6, mobile devices, they are everywhere.
And that ubiquity is defined by their embedded sensors.
Accelerometers are a great example.
Use brilliantly in that old iBeer app to simulate drinking.
Right, and input relies so heavily on contextual data, often through scanning.
We see that with barcodes and QR codes.
Yeah, probably.
But while QR codes have this massive appeal for companies, the practical usability is often just frustrating for users.
They can be slow, cumbersome.
You're always fumbling with the camera angle or the lighting.
And the ethical angle is immediate with mobile.
Just think about smartwatches.
Notifications are annoying on a phone, but a device on your wrist is ever present.
That constant susceptibility to pings and promotions,
just because you only have to glance down,
it raises real concerns about constant interruption.
And from a pure design standpoint, the physical constraints are severe.
The target size for your finger is crucial.
You need at least seven to 10 millimeters, and you really have to prioritize one primary action per screen to avoid clutter.
And finally, in this section, type seven, appliances.
These are what we call transient interfaces.
You use them for short, specific tasks.
Think a washing machine or a ticket dispenser.
They demand simplicity and high visibility.
And the trade -off here can be brutal.
Let's say you replace a standard toaster's physical lever and knob with an LCD screen.
You gain precision, you get extra features like keep warm or toast one side, but what do you lose?
You lose simplicity.
Absolutely.
The increased cognitive load for a simple task like making toast can actually make the whole experience way more frustrating.
Okay, let's pivot now to the input revolution.
Starting with type eight, voice user interfaces or VUIs.
The smart speakers.
Right, interaction can be command -based, play music or conversational, like in a call writing system.
And machine learning has made them so much more accurate.
Dictation systems can be three times faster than typing.
A really critical interaction mechanism here is barge in.
Tell us about that.
It's what lets you interrupt the system's prompt if you already know what you wanna say.
You just talk over it.
And the dialogue styles matter too.
Directed dialogue, where the system controls the conversation, is usually less error prone than flexible dialogue, where a user might just overwhelm the system with too many details at once.
And the social dimension is key.
A smart speaker sits in a shared space, like a living room.
It encourages multi -user interaction, which is a big shift from the very solitary model of a handheld device.
Then you have type nine, pen -based devices.
These capitalize on our ability to write.
Systems like digital ink can record your movements on special paper.
It's useful for form filling.
But the big design challenge is that your hand often covers the small screen you're trying to write on.
Next is type 10, touchscreens.
They've evolved so fast from single -touch kiosks to multi -touch surfaces that let you swipe and pinch with multiple fingers.
And the research concern here is developing new conceptual models, right?
Like cards and carousels to replace the old mouse and GUI actions, especially because, well, you can't feel anything.
Exactly.
To compensate for that lack of tactile feedback, some interfaces use dwelling touches, where you press and hold to confirm an action or bring up more options.
Okay, type 11, gesture -based systems.
We're talking about using arm and hand movements.
Yes, often captured by cameras.
Designers found that gestures need to be sequential, kind of like a noun followed by a verb.
And there's a powerful real -world example of this in hospital operating rooms.
Using tech like the Microsoft Kinect, surgeons can manipulate X -rays on a screen without ever breaking sterility.
They just gesture in the air.
That's incredible.
And finally, type 12, haptic interfaces.
This is all about feedback through vibration or force using actuators.
You see it in gaming steering wheels or even concepts for remote communication, like a simulated hug.
Haptics is also getting into specialized training.
There's something called the music jacket, which uses these subtle vibratactile nudges on a violent student's arm to correct their posture in real time.
And the emerging tech here is wild.
UltraHaptics creates the illusion of midair touch for virtual buttons.
And we're seeing exoskeletons with artificial muscles made of air bubbles to help with mobility.
So the design consideration is key.
The feedback has to be timely and targeted.
Intermittent buzzing works.
Continuous vibration is just an annoying distraction.
Okay, section five.
We're moving into shared spaces and blended realities.
Type 13 is multimodal interfaces.
So combining different senses.
Yes, combining inputs like speech, vision, touch, and haptics to make the experience richer.
The most common pairing is speech and vision.
A great early example is the Microsoft Kinect, which used a camera, depth sensor, and microphones to track your joints in 3D and map them perfectly to an avatar.
Then there's type 14, shareable interfaces.
These are designed for multiple users working together in the same physical space.
Like interactive tabletops or smart boards.
Exactly, and one really unique system called Diamond Touch can actually identify which user is touching the surface using distinct electrical signals.
That's crucial for equitable group work, and the research consistently shows that horizontal surfaces, like tables, support more collaboration than vertical ones.
Which brings us to type 15, tangible interfaces.
This is where physical objects like bricks or models are coupled with digital effects.
I love this idea.
So in urban planning, there are systems like ERP where planners can move physical model buildings around.
And the digital shadows on the projection change instantly in response.
This fosters such deep collaboration because manipulating physical things just encourages problem solving.
There's also a great, pluriple example, the VoxBox.
It's a tangible questionnaire machine.
A tangible questionnaire.
Yeah, it uses familiar physical things, sliders and knobs to let people give their opinions.
And then it shows your progress with a physical indicator, like a ball dropping down a tube.
It makes a mundane survey way more engaging.
And that leads us to type 16, augmented reality or AR.
This is where you superimpose digital elements onto the real world.
Pokemon Go is the huge success story, but it's used in high -end cars with head -up displays that show navigation arrows right on the road ahead.
And we also see it in virtual trions, from Snapchat filters to retail mirrors.
For example, the magic face AR mirror let people virtually try on complex opera makeup in a dressing room setting.
But the research concern here is critical.
The digital augmentation has to be simple and perfectly aligned with the physical world.
It needs to stand out enough to be informative, but not so much that it distracts you from your real environment.
Okay, we have finally reached the future interfaces.
We're starting with type 17, wearables.
Devices worn on the body watches, trackers, smart glasses.
But these devices raise immediate ethical questions.
They certainly do.
Take Google Glass.
It sparked so much social discomfort because the wearer looked like they were staring.
And the idea of constant recording raised huge privacy concerns.
So the research here focuses on really fundamental issues like comfort, hygiene, and where on earth you're supposed to put the batteries and electronics.
Type 18 covers robots and drones.
Beyond industrial uses, we're seeing therapeutic companion robots like Pero the Seal, which is used really effectively with dementia patients to encourage social interaction.
And drones are being used for everything from entertainment to agriculture and wildlife tracking.
For robots, the ethical concern is anthropomorphism.
Should they look human -like, pet -like, or just like a machine?
That design decision really impacts social acceptance.
And for drones, the challenge is noise and privacy, especially when they're hovering over residential areas.
Type 19, brain -computer interfaces, or BCI.
This is a direct communication pathway between brainwaves detected via electrodes and an external device.
Medically, the BrainGate project has shown incredible success, letting paralyzed patients control robots and cursors just with their thoughts.
And other companies are now focusing on non -invasive mass -market applications, like games.
And finally, Type 20, smart interfaces.
These are context -aware devices with embedded AI.
We see them everywhere from smart buildings to smart homes with thermostats like Nest.
But the core debate in this area is massive.
Should the goal be to take the human out of the loop entirely, which can be really frustrating when the controls are centralized?
Or should the focus be on human -building interaction, which prioritizes the needs and values of the people living there?
So after charting all 20 types, we have to synthesize this into the big idea of the natural user interface, the NUI.
Right, the NUI relies on interacting with voice, hands, and your body using our everyday human skills.
The theory is that they should be inherently easier to learn than GUIs.
The renowned designer Don Norman argues that NUI is not inherently natural.
There's a catch.
Yeah, think about a sensor -based faucet.
On and off is simple.
But if you have to design complex gestures just to control temperature and flow, it gets clumsy and arbitrary really fast.
So the naturalness of the NUI is highly dependent on the complexity of the task and the context.
So the interface that feels most natural is always context -dependent.
But that said, NUIs have undeniably improved accessibility, dramatically improving things like the iPhone's voiceover for visually impaired users.
And the future of design is pushing far beyond just smart.
We're looking toward interfaces based on subtle, continuous inputs data streams from brain sensors, body sensors, behavioral sensors, all combined with advanced AI.
Which leads toward augmented cognition, basically giving users technological superpowers.
And that brings us back to choice.
Which interface should you choose?
It always depends on the requirements mobile for dietary monitoring, a shareable one for group work.
But the final selection has to account for these essential factors.
Reliability, social acceptability, privacy, and ethics.
And we see that blend of disciplines being championed by designers like Leah Butchley, whose work focuses on blending technology with art and craft.
Like the Lily Patarduinos, which are sewable electronics or computational ceramics.
These tangible wearable case studies really broaden the conversation about who technology is for and how it should be used.
It's so clear that the Lang era is long gone.
It's been replaced by this dense ecosystem where the best interface is always the one that perfectly suits the context.
So for you, the listener, think about an activity you do every single day.
Maybe it's studying, maybe it's cooking a complex meal.
Now imagine redesigning its entire interface using a brain computer interface or a truly engaging tangible system.
How would that fundamental shift in interaction change the entire experience of that routine task?
It raises critical questions about how we define what's necessary and what is truly possible.
Thanks for joining us for this deep dive into the 20 faces of interaction design.
We'll see you next time.