Where Computer Interfaces Are Going : 3D Beyond Games

This article serves two purposes. It is a discussioncapitalize on what 3D has to offer. The precise
about the future of computer interfaces; and it is anature of how and where 3D can best be
means by which I can purge myself of thoughts thatincorporated is an open question, and a framework
have been accumulating on this topic for quite a fewto evaluate these questions seems appropriate. As a
years. Even if it fails as intelligent discourse on therough starting point it seems reasonable to divide the
first, it will have succeeded in the second. Previouslyattempts into two broad categories: those that are
the title was Where Are Computer Interfaces Going?trying to simulate the physical world and those that
but after writing it I noticed a significant number ofprefer more abstract representations. If you'll indulge
predictive passages and decided to be bold andme, I'd like to call these two approaches,
move the "are". Of course now I feel obliged to addrespectively, the "Physical Simulation Approach" (PSA)
a disclaimer. I admit right here, or at least in the nextand the "Abstract Representation Approach"
sentence, that I don't know where computer(ARA).Developers in the PSA camp are taking
interfaces are going. I don't know.With that out ofphysical simulations and hanging applications, web
the way, I'd like to start, as many interfaces do, withsites, movies, and pictures on simulated walls.
the metaphor. In the 80s and 90s successfulSimulated desks have functional simulated calculators
interface design and an appropriate metaphor wereon them. And, perhaps, there is a simulated sun
taken to be nearly synonymous. Although a goodoutside. It's all very familiar and comes with a nice
metaphor is important, it imposes unnecessary andminimal learning curve.The ARA camp are working on
artificial restrictions. So why is it so important? Thestrange visualization techniques to view complexity
best, perhaps only, reason is familiarity. Unfortunately,and patterns in large amounts of data. They have
familiarity comes at a cost: the shorter learning curvegeneral graphs floating around in space with links
can require speed and ability to be sacrificed.Considerjoining concepts and words together in arbitrary
the ubiquitous desktop metaphor. What is moreways. They have nifty algorithms that filter the
powerful, the abstract construct of a tree, or a singlesalient characteristics of large data sets so you don't
flat surface to place your papers on? Well, a tree is.get overwhelmed. Their attempts are, by far, much
In fact it is so much more powerful that it is theharder to describe with these mere words.In practice
cornerstone of all modern file systems. Trees aremany attempts will combine aspects of both
great, they impose an organizational order that isphilosophies. I suspect that successful attempts at a
common in natural systems. General graphs are,3D interface will have to balance these two
perhaps, too general. DAGs (Directed Acyclic Graphs)extremes in appropriate ways. Objects in a functional
are a good contender; largely because of their3D interface should probably be represented with
acyclicness, but also because they extend trees in amodels that are familiar, just like the icons on your
well defined way. I suspect that trees are so usefuldesktop are often imitations of familiar real-world
because we can't move backwards in time. Speciesobjects. This is a PSA property. On the other hand,
speciate, languages extend, and software bloats. Totree-based organizational systems would be well
fight these is to fight the increasing entropy of theadvised. Very much an ARA concept.Text should
universe.Would it be a good idea not to allow foldersalways be view-plane aligned, as should images. This
within folders within folders just because it would beis one of those 2D features mentioned earlier.
physically cumbersome, and at some point impossible?Images and text may be scaled, but they should not
Probably not. Do icons have a real-world counterpart?present themselves at an angle. Vertical and
Not really. Metaphors should be, and have been,horizontal edges need to remain vertical and
taken only so far.So what does the future hold? Willhorizontal. Of course, these features are trivially
interfaces be 3D? Will we be stuck with rectanglespresent with your desktop interface as well.And
forever? I think it's reasonable to say both have theirthere's an important lesson: build on the backs of
place. People on the 3D side think that we humansgiants. The desktop UI is successful for a reason, not
see, work, live, and play in 3D. We don't. They saysimply because it has a familiar analogue in the
they can't wait until there are fully 3D monitors thatphysical world, but rather because it behaves in that
you can walk around. Why? Our retinas, as well assame useful way that real desks behave. It takes
birds whose eyes are plastered on the side of theiradvantage of a well-established ability; spatial
heads, are two-dimensional surfaces. Birds havememory. You put something down and it stays
flatter vision than we do, if not as Euclidean, becausethere.Useable interfaces need a certain amount of
they don't have the benefit of the tiny bit of 3Dpersistence in their structure. Having objects stay
depth perception a predator gets by overlappingwhere you leave them is one good way to achieve
images. I've heard graphics programmers explain thatpersistence. Placing objects manually, whether on
their 3D scene was being projected onto a flat 2Dyour desktop or in a 3D environment, takes
screen and so it was no longer really 3D. But consideradvantage of spatial memory. We can remember, in
this: everything you see in this world is like that. It allcontext, where we've left hundreds of objects
gets projected onto our flat retinas. We just have(notwithstanding car keys; they get moved around
really big brains. A 3D scene is constructed in ourtoo much). You probably know where your camera is
mind regardless of whether what we're viewing is onand where the light switches in your home are. By
a flat computer monitor or in that nether-worldpositioning objects manually you can give them some
known as real life. In fact, most brains do a decentcontext; perhaps by placing pictures of your family to
job of scene construction even with one eye closed.the left, and panoramic vistas to the right. Contextual
From 2D to 3D. Impressive!People on the 2D sideclues help you remember.I've heard the assertion
think that we humans see, work, live, and play in 2D.that adding a single extra dimension doesn't buy you
We do, after all, have flat retinas, like playing tennismuch organizational power and that the added
on flat tennis courts, and eat dinners from flat platesnavigational complexity isn't worth it. Others think
on flat tables. But we don't live in 2D. Our brains arethat we need an n-dimensional space to do a good
really big. 1.3 litres big. More than enough dendrites,job. Aside from the obvious observation that we
axons and other brain-things to contain a nice 3Dseem to exist in a macroscopically three-dimensional
representation of the world we live in. Clues to buildworld (macroscopically was added just to keep any
the scene abound: motion, foreshortening, and thephysicists-who-may-know-better reading) and are
aforementioned depth perception.The truth is sometherefore good at 3D manipulations, there is evidence
things are better in 2D and some 3D. Writing a letter?that the jump from 2 to 3 dimensions is of a more
Use a desk. Put a flat piece of paper on it. Want tofundamental significance. If you draw a bunch of dots
file that letter away? Wouldn't it be cool if you couldon a piece of paper you will not be able to draw lines
just let it hover in some large 3D organizationaljoining the dots in all possible configurations unless the
space? Here's what I think.lines cross (given some sufficiently large number of
Text: 2Ddots. I think 5 might do it). However, once you hit
Reading and Writing: 2Dthree dimensions, all configurations are possible
Organizing and Grouping: 3Dwithout crossings. Adding a fourth or fifth doesn't
Visualization of Data: Dependshave any further beneficial effect. Admittedly there is
It has occurred to me that 2D representationssome hand-waving going on here; but the result has
should be considered a feature of an interface. It'simplications for some possible interface designs; and it
beneficial that text documents are lined up nicely forpoints to using three dimensions.So why haven't
you in a window. If head-or-eye-tracking hardwareinterfaces changed much in the last 20 years? One
were more widespread, we'd have software thatpossibility is that the desktop is in some way an
could compensate for (single) users who are notoptimal representation. More likely, however, is that it
directly in front of their screens. Imagine looking atis simply a functional representation; no need to
your monitor from an angle but still having the textchange when change takes effort, right? We expect
of this article appear flat. That would be a prettyto be able to sit down in front of a new interface
neat feature (on the other hand, it might just lookand immediately be as productive as we were
strange and make you sick; hard to tell without tryingbefore. We have all learned to use the desktop and
it).Because the input is essentially 2D, I predict puremenu-driven interfaces because we haven't had a
3D imaging devices will prove to be a novelty even ifchoice. It has taken time; just as learning to read and
the enormous bandwidth problems can be solved. Awrite took years when we were younger. Even the
graphics card that draws a 480x480 pixel scene atkeyboard and the mouse, although perhaps easier
60fps would take 8 seconds to update athan writing, have taken time and effort to master.
480x480x480 cube. Yes, I understand this is a vastNew interfaces will face the same hurdles. Their
simplification. Somehow restricting rendering to thedesigns will need tweaking to reduce the learning
surfaces of an object might help, but it sounds tricky.curve as much as possible. The users of these new
Regardless, the same or better effect will beinterfaces will need the patience to develop efficient
achievable by feeding a couple of 2D images to eachusage patterns; and the interfaces themselves will
eye. Technology that takes this approach will beneed to be entertaining enough to mitigate the
more successful. Devices that project images directlypatience required. All these efforts will yield interfaces
onto the retina seem like a reasonable approach;that are not only more enjoyable, but faster and
along with any tracking technology that may go withmore useful.Tristan Grimmer is technical director at
them.The next 10 years will be a transitional phaseUpper Bounds Interactive Inc. Previously a
for interface design. 3D rendering technologiesvideo-game programmer, and much earlier than that
already have a stable home in the entertainment,a young boy trying to get his Vic-20 to accurately
video game, simulation, and design sectors. Althoughcompute Pi, Tristan now spends his time working on
2D interfaces have dominated everything else, ITactile 3D in an attempt to rid the world of
expect we will start seeing more 3D incursions.rectangles.
Operating systems and applications are beginning to