[Sigia-l] 3-D workflows

[Christopher Fahey [askrom]]

{ Sorry for double post - please ignore my previous misfired email! }


Clifton Evans wrote:
> It should be said that I didn't necessarily mean 3d 'Fly
> Through' or first person 'Shooter' experiences like in video
> games or on TV.

You've put your finger on it there: you suggest that there is a
difference between "3D" as a simple axis of information and "3D" as a
real or virtual space with X, Y, and Z axes. Let me explain:

As George Olsen has pointed out, 2D has worked well for us for
centuries. The advantage of viewing information in two dimensions is
that it allows the user to browse through data by simply moving their
eyes around in their sockets, by sliding a flat sheet of paper around on
a table, or by moving scrollbars on the sides of a browser window.
Throughout history, people have made flat maps, charts, tables,
diagrams, and illustrations in order to translate information and
knowledge into this easily digestable format.  

But what is the real difference between a 2-dimensional display of
information and a 3-dimensional display of information? The difference
is in your ability to change your perspective in three dimensions.

Let me restate that: a so-called 3D information space is NOT really a 3D
information space unless the viewer can move their perspective around
the information space in three dimensions. A 3D pie chart is just a 2D
pie chart unless you, the viewer, are able to move your perspective
around to view the far side of the pie chart, to rotate around it.

In other words, I don't think a digital information space can truly be
called 3D unless it is like Quake or like a flight simulator. A 3D
display of information *must* be exceptionally interactive, not just a
2D image with 3D styling.

A hyperlinked information space (like the Web) can be loosely called a
"3D" space insofar as a user can "move" around the information, not
limited to viewing the information on a single plane of a single page.
While applicable, this is a slippery definition: I think we should
reserve the term "3D" to refer to information that is stored in a format
(like CAD or VRML or whatever) that can be used to duplicate the
real-world experience of moving through space.

This is a 3D dataspace:
   (1,1,1)=x
   (1,1,2)=x
   (1,2,1)=x
   (1,2,2)=x
   (2,1,1)=x
   (2,1,2)=x
   (2,2,1)=x
   (2,2,2)=x
   (2,1,1)=x
Every point in this dataspace has three "dimensions". You could in
theory build an interface that could "browse" through this information
through a small three-field "viewport" that can reveal the contents (x)
of each point by simply supplying the X, Y, and Z coordinates of each
point in the dataset.


This is not a 3D dataspace:
   ___
 /___/|
|___| |
|___| |
|___|/
It is just some 2D information that looks like 3D. You cannot look
behind it or rotate it.

This is the difference I'm talking about.

-Cf

[christopher eli fahey]
art: http://www.graphpaper.com
sci: http://www.askrom.com
biz: http://www.behaviordesign.com