Re: Star Maps

On Tue, 12 Oct 1999, Nyrath the nearly wise wrote:

> > Phillip Pournelle wrote:
>
> >	Unfortunately, the three D view is not effective for a campaign
map.  Is
> > it possible to generate a 2D map, perhaps on a hex sheet, where
there are
> > distortions of the star's actual position,	but positioned to create
a sense
> > of how far apart they would be for	a strategic view?
> 
>	Mr Joel H. Crook <joel@invbank.com> suggested using something
>	called Kohonen mapping in an attempt to preserve the
>	distance information when flattening a map.
>	http://odur.let.rug.nl/~kleiweg/kohonen/kohonen.html

i've had a quick look at this; it looks rather keen. sadly, the only
hard
info on the algorithm is a c program (c is a write-only language :(, and
it doesn't help that the programmer's first language is Dutch ) so i
have
no real idea how it works.

basically, Kohonen mapping can be used to 'flatten' a set of points in 3
dimensions to a set of points in 2 dimensions such that the distance
relations between them are more or less preserved. it can't be used to
make a real 2d map (where measured distance is proportional to real
distance), but it can be used to lay out points on a chart in such a way
that jump paths (which are to nearby stars) will tend to be to nearby
points.

is there any chance Mr Crook will do any work on a 3d -> 2d flattener
himself? if not, i'll put this on my almost infinite pile of things to
do.
actually, it turns out Kohonen maps have applications in visualising
protein structure, so i can sort of justify learning about them!

tom