I'm a huge fan of that algorithm (Gastner-Newman, I worked on a
stand-alone Java version),  but there's something that users should
know about it from a data-representation statement: It produces
results that may not be accurate for the input data.

The algorithm is an optimization one, and one of the things that it
optimizes is getting the areas accurate. So, particular areas or the
whole map may diverge from the target values by quite a bit.

As I say, I think the algorithm is great, but you need to calculate
the areas of the resulting cartograms and compare with the original
input to make sure they are within your tolerance for error.

regards,
-Frank

On Thu, Nov 6, 2008 at 1:59 PM, Tom Longson (nym) <[EMAIL PROTECTED]> wrote:
> This is the first time I've seen something quite like this, so I
> thought I'd share it with the group:
>
> http://www-personal.umich.edu/~mejn/election/2008/
>
> I'd be curious to see this with income / donations data as well.
>
> nym
>
> _______________________________________________
> Geowanking mailing list
> [email protected]
> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
>



-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Frank Hardisty
[EMAIL PROTECTED]

GeoVISTA Center
212 Walker Building

Dutton e-Education Institute
434 Earth and Engineering Sciences Building

814-867-1471
http://www.geovista.psu.edu/grants/cdcesda/software/

_______________________________________________
Geowanking mailing list
[email protected]
http://geowanking.org/mailman/listinfo/geowanking_geowanking.org

Reply via email to