Bitching about Transparent Screens

I’ve been tagged a number of times on Twitter from people are asking me to weigh in on the following comic by beloved Parisian comic artist Boulet.

Since folks are asking (and it warms my robotic heart that you do), here’s my take on this issue. Boulet, this is for you.

Sci-fi serves different masters

Interaction and interface design answers to one set of masters: User feedback sessions, long-term user loyalty, competition, procurement channels, app reviews, security, regulation, product management tradeoffs of custom-built vs. off-the-shelf, and, ideally, how well it helps the user achieve their goals.

But technology in movies and television shows don’t have to answer to any of these things. The cause-and-effect is scripted. It could be the most unusable piece of junk tech in that universe and it will still do exactly what it is supposed to do. Hell, it’s entirely likely that the actor was “interacting” with a blank screen on set and the interface painted on afterward (in “post”). Sci-fi interfaces answer to the masters of story, worldbuilding, and often, spectacle.

I have even interviewed one of the darlings of the FUI world about their artistic motivations, and was told explicitly that they got into the business because they hated having to deal with the pesky constraints of usability. (Don’t bother looking for it, I have not published that interview because I could not see how to do so without lambasting it.) Most of these things are pointedly baroque where usability is a luxury priority.

So for goodness’ sake, get rid of the notion that the interfaces in sci-fi are a model for usability. They are not.

They are technology in narrative

We can understand how they became a trope by looking at things from the makers’ perspective. (In this case “maker” means the people who make the sci-fi.)

thankthemaker.gif

Not this Maker.

Transparent screens provide two major benefits to screen sci-fi makers.

First, they quickly inform the audience that this is a high-tech world, simply because we don’t have transparent screens in our everyday lives. Sci-fi makers have to choose very carefully how many new things they want to introduce and explain to the audience over the course of a show. (A pattern that, in the past, I have called What You Know +1.) No one wants to sit through lengthy exposition about how the world works. We want to get to the action.

buckrogers

With some notable exceptions.

So what mostly gets budgeted-for-reimagining and budgeted-for-explanation in a script are technologies that are a) important to the diegesis or b) pivotal to the plot. The display hardware is rarely, if ever, either. Everything else usually falls to trope, because tropes don’t require pausing the action to explain.

Secondly (and moreover) transparent screens allow a cinematographer to show the on-screen action and the actor’s face simultaneously, giving us both the emotional frame of the shot as well as an advancement of plot. The technology is speculative anyway, why would the cinematographer focus on it? Why cut back and forth from opaque screen to an actor’s face? Better to give audiences a single combined shot that subordinates the interface to the actors’ faces.

minrep-155

We should not get any more bent out of shape for this narrative convention than any of these others.

  • My god, these beings, who, though they lived a long time ago and in a galaxy far, far away look identical to humans! What frozen evolution or panspermia resulted in this?
  • They’re speaking languages that are identical to some on modern Earth! How?
  • Hasn’t anyone noticed the insane coincidence that these characters from the future happen to look exactly like certain modern actors?
  • How are there cameras everywhere that capture these events as they unfold? Who is controlling them? Why aren’t the villains smashing them?
  • Where the hell is that orchestra music coming from?
  • This happens in the future, how are we learning about it here in their past?

The Matter of Believability

It could be, that what we are actually complaining about is not usability, but believability. It may be that the problems of eye strain, privacy, and orientation are so obvious that it takes us out of the story. Breaking immersion is a cardinal sin in narrative. But it’s pretty easy (and fun) to write some simple apologetics to explain away these particular concerns.

eye-strain

Why is eye strain not a problem? Maybe the screens actually do go opaque when seen from a human eye, we just never see them that way because we see them from the POV of the camera.

privacy

Why is privacy not a problem? Maybe the loss of privacy is a feature, not a bug, for the fascist society being depicted; a way to keep citizens in line. Or maybe there is an opaque mode, we just don’t see any scenes where characters send dick pics, or browse porn, and would thereby need it. Or maybe characters have other, opaque devices at home specifically designed for the private stuff.

orientation

Why isn’t orientation a problem? Tech would only require face recognition for such an object to automatically orient itself correctly no matter how it is being picked up or held. The Appel Maman would only present itself downwards to the table if it was broken.

So it’s not a given that transparent screens just won’t work. Admittedly, this is some pretty heavy backworlding. But they could work.

But let’s address the other side of believability. Sci-fi makers are in a continual second-guess dance with their audience’s evolving technological literacy. It may be that Boulet’s cartoon is a bellwether, a signal that non-technological audiences are becoming so familiar with the real-world challenges of this trope that is it time for either some replacement, or some palliative hints as to why the issues he illustrates aren’t actually issues. As audience members—instead of makers—we just have to wait and see.

Sci-fi is not a usability manual.

It never was. If you look to sci-fi for what is “good” design for the real-world, you will cause frustration, maybe suffering, maybe the end of all good in the ’verse. Please see the talk I gave at the Reaktor conference a few years ago for examples, presented in increasing degrees of catastrophe.

I would say—to pointedly use the French—that the “raison d’être” of this site is exactly this. Sci-fi is so pervasive, so spectacular, so “cool,” that designers must build up a skeptical immunity to prevent its undue influence on their work.

I hope you join me on that journey. There’s sci-fi and popcorn in it for everyone.

Brain interfaces as wearables

There are lots of brain devices, and the book has a whole chapter dedicated to them. Most of these brain devices are passive, merely needing to be near the brain to have whatever effect they are meant to have (the chapter discusses in turn: reading from the brain, writing to the brain, telexperience, telepresence, manifesting thought, virtual sex, piloting a spaceship, and playing an addictive game. It’s a good chapter that never got that much love. Check it out.)

This is a composite SketchUp rendering of the shapes of all wearable brain control devices in the survey.

This is a composite rendering of the shapes of most of the wearable brain control devices in the survey. Who can name the “tophat”?

Since the vast majority of these devices are activated by, well, you know, invisible brain waves, the most that can be pulled from them are sartorial– and social-ness of their industrial design. But there are two with genuine state-change interactions of note for interaction designers.

Star Trek: The Next Generation

The eponymous Game of S05E06 is delivered through a wearable headset. It is a thin band that arcs over the head from ear to ear, with two extensions out in front of the face that project visuals into the wearer’s eyes.

STTNG The Game-02

The only physical interaction with the device is activation, which is accomplished by depressing a momentary button located at the top of one of the temples. It’s a nice placement since the temple affords placing a thumb beneath it to provide a brace against which a forefinger can push the button. And even if you didn’t want to brace with the thumb, the friction of the arc across the head provides enough resistance on its own to keep the thing in place against the pressure. Simple, but notable. Contrast this with the buttons on the wearable control panels that are sometimes quite awkward to press into skin.

Minority Report (2002)

The second is the Halo coercion device from Minority Report. This is barely worth mentioning, since the interaction is by the PreCrime cop, and it is only to extend it from a compact shape to one suitable for placing on a PreCriminal’s head. Push the button and pop! it opens. While it’s actually being worn there is no interacting with it…or much of anything, really.

MinRep-313

MinRep-314

Head: Y U No house interactions?

There is a solid physiological reason why the head isn’t a common place for interactions, and that’s that raising the hands above the heart requires a small bit of cardiac effort, and wouldn’t be suitable for frequent interactions simply because over time it would add up to work. Google Glass faced similar challenges, and my guess is that’s why it uses a blended interface of voice, head gestures, and a few manual gestures. Relying on purely manual interactions would violate the wearable principle of apposite I/O.

At least as far as sci-fi is telling us, the head is not often a fitting place for manual interactions.

Precrime forearm-comm

MinRep-068

Though most everyone in the audience left Minority Report with the precrime scrubber interface burned into their minds (see Chapter 5 of the book for more on that interface), the film was loaded with lots of other interfaces to consider, not the least of which were the wearable devices.

Precrime forearm devices

These devices are worn when Anderton is in his field uniform while on duty, and are built into the material across the left forearm. On the anterior side just at the wrist is a microphone for communications with dispatch and other officers. By simply raising that side of his forearm near his mouth, Anderton opens the channel for communication. (See the image above.)

MinRep-080

There is also a basic circular display in the middle of the posterior left forearm that displays a countdown for the current mission: The time remaining before the crime that was predicted to occur should take place. The text is large white characters against a dark background. Although the translucency provides some visual challenge to the noisy background of the watch (what is that in there, a Joule heating coil?), the jump-cut transitions of the seconds ticking by commands the user’s visual attention.

On the anterior forearm there are two visual output devices: one rectangular perpetrator information (and general display?) and one amber-colored circular one we never see up close. In the beginning of the film Anderton has a man pinned to the ground and scans his eyes with a handheld Eyedentiscan device. Through retinal biometrics, the pre-offender’s identity is confirmed and sent to the rectangular display, where Anderton can confirm that the man is a citizen named Howard Marks.

Wearable analysis

Checking these devices against the criteria established in the combadge writeup, it fares well. This is partially because it builds on a century of product evolution for the wristwatch.

It is sartorial, bearing displays that lay flat against the skin connected to soft parts that hold them in place.

They are social, being in a location other people are used to seeing similar technology.

It is easy to access and use for being along the forearm. Placing different kinds of information at different spots of the body means the officer can count on body memory to access particular data, e.g. Perp info is anterior middle forearm. That saves him the cognitive load of managing modes on the device.

The display size for this rectangle is smallish considering the amount of data being displayed, but being on the forearm means that Anderton can adjust its apparent size by bringing it closer or farther from his face. (Though we see no evidence of this in the film, it would be cool if the amount of information changed based on distance-to-the-observer’s face. Writing that distanceFromFace() algorithm might be tricky though.)

There might be some question about accidental activation, since Anderton could be shooting the breeze with his buddies while scratching his nose and mistakenly send a dirty joke to a dispatcher, but this seems like an unlikely and uncommon enough occurrence to simply not worry about it.

Using voice as the input is cinemagenic, but especially in his line of work a subvocalization input would keep him more quiet—and therefore safer— in the field. Still, voice inputs are fast and intuitive, making for fairly apposite I/O. Ideally he might have some haptic augmentation of the countdown, and audio augmentation of the info so Anderton wouldn’t have to pull his arm and attention away from the perpetrator, but as long as the information is glanceable and Anderton is merely confirming data (rather than new information), recognition is a fast enough cognitive process that this isn’t too much of a problem.

All in all, not bad for a “throwaway” wearable technology.