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