[Proj] Oblique Cyl Eq Area

Gerald I. Evenden gerald.evenden at verizon.net
Thu Jul 7 15:57:35 EDT 2005

On Thu, 2005-07-07 at 13:02 -0500, William K wrote:

> On Jul 7, 2005, at 12:31 PM, Gerald I. Evenden wrote:
> > First. "trillion" seems to have a ambiguous definition in Kdict so  
> > I am not sure about
> > your what you mean here.  Is it greater or lesser than a  
> > gazillion?  :-) sorry about my humor.
> >
> I think trillion is a standard thing - the next step beyond a  
> billion, 1,000,000,000,000.

Perhaps it is.  I've only seen reference to the number describing our US
national debt---
far too large a value to comprehend.  In practice here I prefer either
the number or range
in powers of 10.

Alas, the solution to the problem if you wish to fix your own copy of
the source is to
remove the reference to the earth's radius in PJ_ocea.c.  The first two
statements just after the entry ENTRY0(ocea) read

P->rok = P->a / P->k0;
P->rtk = P->a * P->k0;

Change the "P->a" to "1." on both lines.  Simplify the second line if
you like.  I suspect
that this error has persisted over a long line of releases and versions
and points to the
folly of using a unity earth radius as a test value (see p. 279, PP
1395).  Also shows how
often ocea is used.  This is such a dumb error.  Only one projection
references the
earth's radius (or major axis).

> 'proj -le' lists all the ellipoids supported, and sphere is there  
> ('Normal sphere', r=6370997).  I have no seg faults with it.  I get a  
> more graceful bail with unknown ellipsoids.

That is old PROJ4 stuff.  Libproj4 does not list "sphere" as the value
given is really only of interest
to NA users and a holdover from Snyder.  The equivalent in more recent
versions is
"+ellps=clrk66 +R_A" which generates a radius of a sphere with
equivalent surface
area to the Clarke ellipsoid.  See p. 7 of the Release 4 Interim Report.
There are several
other variants of radius being derived from the ellipsoid that are
described on page 7.
You can test the results with the -V option.


I have a handle on the error I was having which was related to error
reporting by lproj.
I believe it is a problem with my inhouse version and will be fixed
before the next release.

> _______________________________________________
> Proj mailing list
> Proj at xserve.flids.com
> http://xserve.flids.com/mailman/listinfo/proj

Jerry and the Low Riders: Daisy Mae and Joshua
"The most certain test by which we judge whether a country is
really free is the amount of security enjoyed by minorities"
---Lord Acton, 1907 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.maptools.org/pipermail/proj/attachments/20050707/129b2adf/attachment.html

More information about the Proj mailing list