Why is "3d zones with buildings" preferrable to "2d web pages with
lists of hyperlinks?" You are not adding any useful information by
mapping it into 3d. You are merely creating a jazzy 3d presentation.
You are also slowing down access to the data tremendously, because you
would have people waste time "flying through" a 3d city to get what
they want, instead of just entering a search keyword into a form.
4. The size of those buildings should be relative to the amount of
information they contain. If size is left free to de decision of the building
"owner", every building would be twice as bigger than the previous one
inserted in the city, and the city would grow with no sense.
It is impossible to determine how much "information" a file holds
based on its size. 2 MB of mindless chatter from some alt.* newsgroup
contains a lot less useful information than a 5K technical document.
In general, there is no linear relationship between the words in a
document and the "content value" of the document. "Content" is a
collection of symbolic relationships, and indexing the symbolic
relationships in hypertext fashion is the proper way to deal with
content. You can't pour "50 milligrams of content" into a test tube,
nor can you conjure up a building that contains "3000 cubic meters of
content." So it makes no sense for a user to measure the available
content by the sizes of the buildings.
5. The "adresses" or phisical coordinates on the 3D model of the virtual city
should be fixed, so anyone could find a resource by remembering where it was
phisically located into cyberespace. If coordinates of a resource change
often, navigation will be confusing and quite dificult. On the other hand,
fixed coordinates do not allow continuous reorganization of the city to allow
an efficient content oriented search, so the city can finally grow -like a
real city- in a quite anarchic way. That would make those searchs inefficient
if you want to try a Yahoo-kind search with the suposed advantages of 3D
visualization. Don't really know the best solution.
Alexandre Soler M.
alexs@servicom.es
Barcelona, Spain
Do they have many malls or "superstores" in Spain - or have you spent
much of your life shopping in gargantuan buildings? Big 3d spaces
don't make it any easier to shop. The first several times you visit a
mall, you have no idea where anything is. To get anything done you
have to ask salespeople where things are. If the store doesn't have
what you're looking for, it takes an awful lot of footwork to
determine that. Only if you go to the same stores over and over
again, and you buy the same things all the time, and the store never
really changes, do you get any speed benefit at all from a big 3d
layout. Given that the content of the Web changes rapidly and
continuously, a big 3d layout will never be useful for accessing
information in general.
IMHO, this idea of virtual 3d "mirror-cities" for displaying Internet
resources is seriously misguided. To improve a user's access to
useful information, what we really need are underlying distributed
database technologies that can automatically correlate symbolically
related information, and documents that can broadcast that they have
certain kinds of information on widely recognized subject keywords.
In other words, this is a library technologies problem, not a 3d
visualization problem.
Cheers,
Brandon J. Van Every | Computer Graphics | The sun attempts
| | to be white,
vanevery@rbdc.rbdc.com | C++ UNIX X11 Motif | as white as
http://rbdc.rbdc.com/~vanevery | HTML CGI Perl TCL/Tk | daytime.