In the middle of interesting article about criminal misbehavior by a participant in an online game, Clay Shirky has an intriguing insight about online interaction.
MUDs and MOOs -- text-based virtual worlds -- were common early genres of online interaction. Prophets and business people extrapolated that future online interaction would be much like these virtual worlds, but with sound and color and 3D. It wasn't that long ago. Remember the early online malls with pictures of buildings and streets?
My label for this was the Whole Worlds hypothesis — interactions would be through the lens of characters playing roles in immersive environments. Two additional assumptions made by the Whole Worlds camp were that the MUD/MUSH/MOO style of interaction would give way to more visually immersive environments, and that the interactions would move outside the realm of games and hang-out spaces, and become normal modes of business interaction.
These hypotheses seem to be right and wrong respectively — the MMO is the logical inheritor of the MUD, but MMOs have stayed game- and hang-out oriented.
Business interaction, by contrast, has remained largely text and voice-based, and has moved in the opposite direction from the Whole Worlds model, towards fragmentation and multi-tasking. Not only are we not immersed in purpose-built online spaces at work, we aren’t even immersed in the real world anymore, as the rise of continuous partial attention (tip of the hat to Linda Stone) means that our presence in reality is lessened by interrupts from phones, IM, the Treo, and so on.
So the failure of the Whole Worlds model outside games is pretty obvious, leaving game worlds as the principal site of that theory.
My question is, why is that? Is it because we in business lack creativity and imagination, and prefer to live in a duller world? Then again, game designers don't use games to design games (to the best of my incomplete knowledge) -- they use text-editors, IDEs, storyboards.
Web usability folk had explanations at the time about the failings of the early cartoon interfaces, such as Jakob Nielsen's rant. Nielsen argues, alternatively:
* that our preference is biological. "If we had been frogs with eyes sitting on the side of the head, the story might have been different, but humans have their eyes."
* that the problem is the limits of web graphics and navigation tools: ""Users need to pay attention to the navigation of the 3D view in addition to the navigation of the underlying model: the extra controls for flying, zooming, etc. get in the way of the user's primary task".... Poor screen resolution makes it impossible to render remote objects in sufficient detail to be recognizable; any text that is in the background is unreadable
* that 3d is confusing because the space being modeled has more than three dimensions. "Most abstract information spaces work poorly in 3D because they are non-physical. If anything, they have at least a hundred dimensions, so visualizing an information space in 3D means throwing away 97 dimensions instead of 98: hardly a big enough improvement to justify the added interface complexity."
But those explanations are just-so stories, providing backfill for something that we already knew -- text-heavy interfaces like Amazon and Google were more popular than "Virtual Main Street."
Is this another manifestation of Tim Bray's motto, "knowledge its a text-based application?"
I don't know. What do y'all think?
Posted by alevin at August 18, 2004 09:20 AM | TrackBackWhat is this "mystery" you speak of? :) The "3D" we're used to - the real world - is slow and inefficient, and ironically, due to its increased dimensional capacity, more restrictive in many ways.
Fortunately, our brain doesn't (so much) work in such ways. Information is - for the most part - instantly accessible, and stored in pretty abstract, freeflowing ways. In coding terms, "processing" in the real world is like trying to find a particular item in a 3-dimensional array - you have to find/know where something is *and* then get to it, whereas processing in the brain seems to act more like a hashtable, or at least an extremely-linked graph - things are tied together within it semantically rather than geographically.
The last Nielsen quote above ("3d is confusing because the space being modeled has more than three dimensions") could also be considered in inverse. 3d is confusing because the space being modeled has *one* (or even, perhaps, no) dimension, and tring to extrapolate it into a 3d world adds unnecessary complexity.
I think there *could* be some... "interesting" aspects to business-related 3d worlds. For instance, I'd like to see an interface that represented my *focus of attention* in 3d terms, along with others. By using, say, proximity (of someone's avatar to mine) as a measure of how related our current activities are (e.g. 2 people coding may be close together), and some kind of directional aspect (e.g. how much are people facing me) to indicate how "available" people are (how much they're willing to break from their activity and talk), I think you could get some useful communication ideas emerging.
But maybe you could do that with a 2D list, as well...
The proximity idea is cool. Typical recommendation engines are focused on individuals. It would be interesting to see current and prospective links based on shared activity.
The UI ideas that suggest themselves for that are textual and 2.x d graphical (a flat space, using color and/or size to show proximity).
More ideas on this in a new post.
Posted by: Adina Levin on August 27, 2004 08:17 AMAlso, moving 3d images invoke the parts of our brains dealing with motor control and balance. In a game, this is exciting - it stimulates a feeling of physical action even when the gamer is sitting in a chair moving fingers.
In simulations of activities requiring motion and excitement -- flying a plane, emergency response -- this simulated physical engagement is useful.
In a social environment, it's distracting. For analytical tasks, it's distracting unless the 3d is very well-designed and well-abstracted.
So, a 3d model of a body part or chemical structure is well-abstracted, removing irrelevant detail. The ability to move the object, and engage the balance/motion parts of the brain, helps to understand.
But a 3d model of a grocery store is poorly abstracted, adding spacial detail about shelf height that's irrelevant to choosing one's groceries.
Posted by: Adina Levin on August 27, 2004 08:35 AMThere seems to be a rift between what a 3D interface is for.
On the one hand, they could be extremely useful for merely presenting information. On the other, traditional approaches have meant that they are poor to navigate by. (That's not to say that navigation within a 3D environment must inherently be difficult.)
2 questions spring to mind...
1. Are there certain "groups" of information that would benefit from having a 3D representation rather than a 2D/1D one? I think this also entails thinking differently about how we transition between "sets" of information - i.e. should we use a web-like "linked page" system if we have the extra dimension?
2. Are 2D working environments more efficient to use because our traditional inputs and interfaces (e.g. the keyboard) are designed for 2D environments? How popular, say, would 3D methods become if we had a). semantically-driven, usably-accurate speech recognition, and b). more touch screen technology, so that we could point to where we want to navigate to rather than use clunky mouse/keyboard interfaces?
Posted by: Scribe on August 27, 2004 12:02 PMHum, just noticed that question 2 pretty much mirrors Nielsen's first two points regarding difficulty. Still, more touch screens would be cool...
Remember, remember: Read article first, drink coffee and comment later.
I think text language is much more mature and agreed-upon than visual language. Short learning curves and immediate "productivity enhancement" (not sure I like that term) are paramount in the business environment, which pretty much forces us to bootstrap from the almost-universal solvent that is text.
But I'm not betting that the interfaces of 20 years hence won't make more efficient use of our 3d perception capabilities. (That interface in Minority Report was just wicked.)
Posted by: Seb on November 1, 2004 01:20 PM