[MetaCRS] EPSG:3785

[Martin Desruisseaux]

Hello all

Frank Warmerdam a écrit :

> I have to admit I am leery of treating this coordinate system as a special
> projection method.  It is just normal mercator except that we want to
> use different ellipsoidal parameters in the projection calculation than
> we want to use in geographic interpretation of the lat/long (which we
> wish to treat as WGS84).

I have the same feeling... Geotoolkit and GeoTools have an approach applicable 
to arbitrary projections while sticking to the standard WKT format, except for 
the occurence of two PARAMETER elements in addition of the usual ones:

     http://www.geotoolkit.org/modules/referencing/faq.html#Google

This approach avoid the introduction of new WKT keywords or projection name. But 
the EPSG approach is also supported. Given that we don't have much choice to 
support the EPSG approach anyway, maybe the incensitive to agree on an 
alternative approach has been slightly reduced compared to the time before EPSG 
defined 3857...

But if this group still feel the need to agree on an alternative approach to 
EPSG:3857, maybe the starting point would be to list on a wiki page the various 
known approaches with their pros and cons?

	Martin