Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!know!samsung!zaphod.mps.ohio-state.edu!sdd.hp.com!hplabs!hpfcso!hpldola!paul
From: paul@hpldola.HP.COM (Paul Bame)
Newsgroups: comp.graphics
Subject: Re: Bureau of Census Dimeco Map Files
Message-ID: <11390038@hpldola.HP.COM>
Date: 27 Jul 90 18:44:31 GMT
References: <1224@umriscc.isc.umr.edu>
Organization: HP Elec. Design Div. -ColoSpgs
Lines: 30

If this is the same Census data I have seen, you get two endpoints per
record containing first the unprojected lat/lon and then the projected
versions:

lon1    Geodetic "From" Longitude (Radians, negative sign assumed)
lat1    "From" Latitude (Radians)
lon2    "To" ...
lat2    
x1
y1
x2
y2

The projected coordinates are supposed to be an Albers projection scaled to
fit a 20x30" map.

A good idea might be to re-project the lat/lon pairs so you *know* what
projection and scale is being used.  A suitable projection is the Albers (as
originally used supposedly) as described in one of the "Gems" in Glassner's
new book (I wrote the gem) and have posted code before.  Another person here
in comp.graphics recently posted the updated version of the book from which
I got the algorithm.

You may also need a UTM <==> lat/lon converter which I'd like to collect
as well.  Shouldn't be too hard to write but I lost my UTM references years
ago (ROTC map class believe it or not).

			-Paul "Spice is the Variety of Life"
			paul@hpldola.hp.com	N0KCL
			719 590 5557