[Proj] The problem with proj.4

[Frank Warmerdam]

support.mn at elisanet.fi wrote:
  > So fuzzy calculations are not acceptable.

Janne,

I'm not sure what calculation you consider to be fuzzy.  There
are things that are not currently expressable in in a one
step transformation but I don't understand how anything is fuzzy.

> There have
> to be a way to define the calculations clearly in a standardised
> way. 

I don't disagree with this statement.  But one of my concerns is
that if this ability (distinct ellipsoid def for datum vs. projection
calcs) is added to PROJ.4 there will still be no analog in most
other coordinate system definitions I work with (GML, OGC WKT,
GeoTIFF, mapinfo, etc).  This significantly decreases my
enthusiasm.

I will note that anyone desiring more precise control over processing
steps can still make multiple calls into PROJ.4 - doing individual steps
(projection, then datum shift, then projection).   The downside of this
is that it puts an onus on the application to decompose the steps.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent