Unlocking the 3rd ‘D’
A Spatial Design Manifesto:
Where are my XR UX people at? I wrote a manifesto just for you. That’s right. All five of you.
Is it just me? Or does almost every XR UI cling to tropes that came from web, mobile, and desktop?
Here are a few layouts from reputable XR companies:
These are fine, right? Easy. Intuitive. Obvious.
How about this one?
Look development aside, don’t the core concepts in the UIs above strike you as a little… old? Compared to this simple cocktail menu, they seem ancient to me. But all were created within a few short years of each other. What gives?
Why, in this age of incredible devices that can turn science fiction into reality, do we find ourselves clinging to legacy tropes? Why aren’t we sprinting toward a new horizon?
Honestly, I think we’re scared. We’re considered proficient if our designs are intuitive, but many new ideas have practically never been touched by human hands. How can a brand new XR interaction model be as intuitive as a scroll bar on first touch?
If you invent the scroll bar of XR, there’s no thirty years of institutional knowledge backing you up. Institutional knowledge works against you.
Hey you. Yeah you. So you finally got your chance to do some UX design work in XR. Not so fast. I know your secret. You may have thought you got away with it, but I know what you did.
I know you gave up on three ‘not-enough-time-for-that’ ideas before slapping that scroll bar into your XR UI. We both know that scroll bar is tired. Old. Safe.
But I also know you did it because it earned you high fives from your colleagues for delivering on time. I forgive you, because I can relate.
If we were to push the less conventional idea, we might get reprimanded or even lose our jobs for not delivering deploy-able work. OR… we might just stumble upon that new interaction model that’s about to define the next era of HCI.
Bold new steps require risk. Early success on this front has mostly come from designers who say…
“Screw the skeuomorphism haters!”
…as an auditorium full of who’s-who designers clutch their pearls.
We all felt a zing of “whoa, that’s cool!” when we pulled the drawers open in Robot Repair, or put a VR headset on our already-HMD-donned faces in Accounting in order to begin the experience. These are low-hanging skeuomorphic fruit. There are other examples that get deeper into the UX stack. Follow me. We’re on the hunt for new interaction models.
Why did underlined text become the insignia of hyperlinks?
It just did, right? The reason it was underline and not background color was arbitrary. Whatever the mechanic, it works really well as that thing, that totally arbitrary thing that can be anything.
What does a hyperlink look like in XR? Skeumorphism has been invaluable in helping us figure that out. If you ask some of our fellow devs on the social VR scene, they’ll say 3D hyperlinks are either portals a la Stargate or doorways a la Narnia.
Good start, but limited. You need a wall to put a portal on, or at least a large empty space so you can plop it in front of you. Plus, the implication of you moving through the thing is probably fine in VR since teleportation is already a commonplace model, but not so much in AR and MR.
Sure, portals are a neat trick in MR too, but a ubiquitous model? Probably not. Eddie Valiant needed a tunnel to get to Toon Town. He didn’t need one for Roger Rabbit to show up in his bed. In MR, the content mostly comes to you.
Hold on, though. That skeumorphic bath water might still have a baby in it. How about something more like a portkey from Harry Potter? A random thing could fill your room with new content when you touch it. It could be any object. Like the hyperlink applies to any word, portkey-ness could apply to any asset. Perhaps you recognize its portkey-ness by some sort of rim light or particle effect.
“Why did this candlestick fill the room with childhood memorabilia when I picked it up?”
“It’s a portkey. They can store anything inside anything. Pretty sweet, right?”
If you’re not ready to jump off the skeumorphism hater train, abstractions like portkeys can help you stay away from the digital thingy becoming too much like its physical counterpart. I don’t agree we should blanket-shun the style of real-world objects in a medium whose nature is to mimic the real world, but XR has plenty of room to go hog-wild with modernism and minimalism.
You know what could be super-duper modernist and not stale? Interaction models with abstract 3D iconography that provide the same value prop to 3D content that legacy models did for 2D.
Take the scroll bar. You 86'ed the drop shadow. You squashed that rect to a line. You squashed that line to a dot. You squashed all the things. It’s still not right. Yet we keep seeing scroll bars everywhere. They infest our XR UIs. Why?
What is this huge value we’re constantly chasing with scroll bars?
- Navigation
- Flow
- Infinite Content
Touché. But is there an existing model that could provide these same values natively in XR?
I’ll have a cheeseburger, fries, and a super-sized gallon of NOPE.
No such model exists, but slapping a scroll bar into an XR UI is not our only alternative. Imagine you’ve never seen a scroll bar. Clear your mind of any and all bar scroll-age.
Now imagine a mechanism emerging from the void. This magical doohickey gives you the ability to let 3D objects flow through your room endlessly…
Did your mind go immediately to conveyor belts? Toy trains? Pneumatic tubes? Now we’re scratching the surface of a trope we haven’t touched in XR yet.
I’m talking about spline rails, y’all. If spline rails don’t get your UX nerd gears turning, I don’t know what will.
Spline rails could help us unlock the 3rd ‘D’. They’re sort of like scroll bars, but they provide something scroll bars can’t: spatial context.
A spline rail exists in a certain orientation in world space. It does not re-position itself to be perpendicular to your perspective. It can be re-positioned manually, but once in place it should be equally accessible from any position in the room without it changing orientation. It can be a straight line, a complex curve, even a circle if you like. It can loop, but it doesn’t have to. Its purpose is to reveal and dismiss 3D content as you point and drag along it.
The key difference between a spline rail and a scroll bar is spatial reference. A spline rail is world-locked. A scroll bar is gaze-locked. A spline rail doesn’t need to billboard to face you. It is truly spatial.
By its very nature, doesn’t a scroll bar deny access to the 3rd ‘D’ by billboarding to face you at all times? It’s at best a 2.5D surface navigator. Does it count as a spatial interface?
I’m not saying non-spatial trends in XR UX are inherently wrong, but they limit us. We will only level up to XR 2.0 by breaking through limits like this.
Spline rails surely aren’t the only model that will help us unlock the 3rd ‘D’. Any model that enables objects to move toward us, away from us, or in any combination of all three axes will break us out of the legacy box we’ve built around ourselves.
“But dude. Bro. Dude-bro… the command line is wicked efficient. Pshaw, you use a mouse? That’s a fad. It’ll be gone in a couple years.”
- Random Software Dude-Bro, 1986
I don’t know about you, but I won’t be that software dude-bro who finds himself on the wrong side of UX history. This dude-bro intends to use all three dimensions to their full potential.
Think of the first time you played Tilt Brush:
1) You learned a few basic controls.
2) You doodled in mid air and thought, “Oh, it’s a paint app, but in VR.”
3) Someone said, “Now reach out, pull the trigger, and drag the controller toward you.”
Cue your mind being blown. In that moment, the 3rd dimension suddenly joined the first two in your brain canvas, and the world could never be the same again.
New possibilities opened like a floodgate the first time you painted the air. We haven’t even begun to open that gate in XR UX. Let’s open it.
Sincerely,
AJ Campbell, friendly neighborhood XR Dev
Way too late on a work night, May 1st, 2019
← If you found this article interesting, please pass it on by clicking “Share on Twitter.”
Some other spatial computing articles I’ve written in the past:
