----------------------------Original message----------------------------
Apologies to those of you who receive both MAPS-L and GIS-L. I am deliberately
cross-posting, because I am equally interested in responses from the GIS and the
cartographic communities (to whatever extent they may be different!).
 
A recent article in New Scientist (July 3rd I think) carried an article about
a new atlas produced by Hammond. The consultant on this was, apparently,
Mitchell Feigenbaum, a mathematician better known to many for his work on
Chaos theory. I was sufficiently intrigued by this that I thought I would
share some of the details with you, not only for your information, but also
in the hope that someone out there may know more about the concepts and methods
used in creation of this new mapping.
 
The cartography for the atlas was, it seems, almost all done using automated/
digital methods developed by Feigenbaum and, apparently, draws quite a bit on
fractal methods, chaos theory, etc. A new (?) conformal projecton was developed
which, they claim, has fewer distortions than any other (I guess they would
claim that, wouldn't they?  :-)  and which appears based on equations derived
from working backwards from the desired scale factor (is this correct? I
didn't quite understand it!).
 
Techniques of label placement seems to have borrowed from sub-atomic physics
(?) in that rays are extended out from objects, and interaction of these with
rays from other objects determines the location, orientation, etc., of the text
message. The message was that all label placement was automated...
 
What I have given is a very sketchy skeleton, which has probably picked up
wrongly on certain points. I would be most interested in any expansion or
clarification that list readers can offer. Also, what are the maps produced
by all of this "new technology" like? Do they work? What are their properties?
Does anyone have a reference / citation for the atlas(es) in which they appear?
 
Darius Bartlett
[log in to unmask]