[Proj] "Double Ellipsoid" error, reproduction

[support.mn at elisanet.fi]

I agree,

and especially there that the problem might rise mostly
when trying to reproject old maps to more accurate datums. If the job
of the pgm is only to create new maps and just draw them with a
plotter, there never exists this problem, since one can always use the
same ellipsoid. But in the world of old and different wild maps this
must be a rather common case.There it would be very nice to have
the possibility to adjust all parameters that affect the end product.
People who don't need that freedom, just would not use it...
(nothing more to learn!)

So:

dellps=something
pellps=something

or

ellps=something , when both are  the same.

To handle the Google Earth projection (when not rotated, north is upwards), one
might now write:

proj=nsper
pellps=sphere
lat_0=....
lon_0=....
h=....
dellps=WGS84
datum=WGS84
no_defs

And the dutch case:

proj=.....
pellps=the old special one
dellps=the current dutch ellipsoid
towgs84=something currently mostly used there, 7 par.
no_defs

Any suggestions?

Regards: Janne.

"If there is a weak point in the programming, at least 10% of
the population will always be there.."

--------------------------------------------------------------------------

Noel Zinn [ndzinn at comcast.net] kirjoitti: 
> Yes, and there is another inappropriate coupling, that of a datum with a
> specific transformation to WGS84.  A local datum is not defined by its
> transformation to WGS84 (but by some adjustment that probably happened
> before WGS84 was defined).  That (derived, empirical) transformation to
> WGS84 will depend upon the specific survey monuments occupied (spatial
> variance and distortion in the local datum), the instruments used (GPS,
> Doppler, how much data and how processed), and the date of derivation (since
> the world datums vary over time).  Happily, many software vendors have come
> to understand this and separate the transformation from the datum.  -Noel
> 
> 
> -----Original Message-----
> From: proj-bounces at lists.maptools.org
> [mailto:proj-bounces at lists.maptools.org] On Behalf Of Gerald I. Evenden
> Sent: Sunday, December 07, 2008 11:10 AM
> To: PROJ.4 and general Projections Discussions
> Subject: Re: [Proj] "Double Ellipsoid" error, reproduction
> 
> On Sunday 07 December 2008 11:30:09 am Noel Zinn wrote:
> > Gerald,
> 	...
> > (theoretically) without empirical error.  Except - and this is my
> objection
> > in this thread heretofore - except when the ellipsoid is changed and a
> > datum transformation is implicitly coupled with a map conversion.
> 
> I concur and that is where I am also having the most problem.  I have always
> 
> viewed map conversions and datums as completely different entities.  This 
> goes back to the early days of several of the programs that tightly bound
> the 
> projections with the datum process and my pleas to more clearly separate the
> 
> processes were pretty much ignored.
> 
> Alas.  :-(
> ...
> 
> -- 
> The whole religious complexion of the modern world is due
> to the absence from Jerusalem of a lunatic asylum.
> -- Havelock Ellis (1859-1939) British psychologist
> _______________________________________________
> Proj mailing list
> Proj at lists.maptools.org
> http://lists.maptools.org/mailman/listinfo/proj
> 
> _______________________________________________
> Proj mailing list
> Proj at lists.maptools.org
> http://lists.maptools.org/mailman/listinfo/proj
>