Conversational AI is the skeuomorphism of VUI
Will we ever move towards a minimal approach to VUI?
You surely remember when about 10 years ago, around the time the first iPhone came out, UI design was all about skeuomorphism (a.k.a. designing interfaces like they were made of real materials, rendering realistic textures and using volumes and shadows, etc).
Why was it like that?
I don’t think we can really pin down a single specific reason. I think it’s a combination of:
- Trends. UI design, as all the disciplines within the big realm of design, is somehow cyclical. Like bell-bottom trousers, they disappear for a decade and then here they are again.
- Just because we could. At the beginning of GUIs, screen resolutions and color range were very limited, the same goes for the first cell-phones. Once we got good-enough screens to show off our design skills and Photoshop rendering ninja tricks we went overboard.
- The excuse we all use: people were still new to that kind of device and apps and having UIs resembling the real things (i.e. the Notes app, looking like a real notepad) was a way to ease the transfer from a real-life experience to a digital one. And this actually makes a lot of sense.
I think the same is happening with VUI. We’re trying to make them look like real people to get accustomed to them.
Will we ever go towards a minimalism of VUI? Like we went from realistic looking apps to the opposite of the spectrum with super minimal, just flat shapes, thin lines.
Until some time ago, the technology was not mature enough to let us go “skeuomorphic” with the design of Voice User Interfaces, but we’re getting there. So is this conversational approach a way to get us accustomed to talking to objects? It might be.
It would actually be the right thing to do. Just like the depiction of realistic knobs and handles were the affordances in our GUI that let us understand what kind of actions we could perform, maybe the “empathetic” and human-like chatting of a smart speaker is the affordance to teach us how to ask things and give commands.
What will come next?
In Graphical User Interfaces, we moved from a realistic approach to a minimal and flat design. We stripped down all the things and effects that weren’t really necessary and evolved (?) to a bare minimum, tasteful (ideally) layout.
Is this gonna happen with VUI as well? After we will get used to talking with our devices, after showing off how life-like AIs can be, will we just get bored with toying with VUIs and just stick to a more basic exchange of commands and answers?
It would be faster, probably less prone to misunderstandings and less cringy.
Maybe it’s just me though. I admit I’m not a big fan of voice commands. I use them mainly to set timers or alarms and I’m not really investing many feelings when doing that. “Set alarm at 7:30” “Timer: 10 minutes”, that’s it. Faster, clear, no chances of being misinterpreted, gets the result.
There are also concerns about AIs pretending to be humans, and Google had to change the way its Duplex service approaches people. So maybe a future with a more minimalistic tendency to VUIs is not so backward.
Less imaginative, less sci-fi maybe, but probably more efficient. After all, they’re still machines, they won’t take offense if we don’t say “please” and “thanks”. For now.
⬜️⬜️⬜️⬜️⬜️⬜️⬜ 👧🏻👧🏼🧒🏻👦🏼👧🏾👦🏿👧🏽👧🏻👧🏼🧒🏻👦🏼👧🏾👦🏿👧🏽👧🏻👧🏼 ⬜️⬜️⬜️⬜️⬜⬜️⬜
My new book “Designing Digital Products for Kids”! Out December the 4th!
Learn the secrets to design successful digital products for children.
You’ll find answers to all your questions regarding the industry, and its peculiarities in 📐 UX design, 🎨 UI design, 🔍 user testing, 📈 business strategies, and much more.
⬜️⬜️⬜️⬜️⬜️⬜️⬜ 👧🏻👧🏼🧒🏻👦🏼👧🏾👦🏿👧🏽👧🏻👧🏼🧒🏻👦🏼👧🏾👦🏿👧🏽👧🏻👧🏼 ⬜️⬜️⬜️⬜️⬜️⬜️⬜
