[MetaCRS] Non-Programming Collaboration

[Landon Blake]

Frank wrote:

" I'm imagining a CSV file with the following columns:

Projected Coordinate System ID: (possibly WKT or an EPSG id)
proj_easting proj_northing proj_z same_datum_longitude
same_datum_latitude same_datum_z wgs84_longitude wgs84_latitude wgs84_z

Then a test run could use a particular engine, transform the point from
projected to lat/long (on the same datum), check, and then back to
projected coordinates and check again."

I suspected something like this, but wanted to confirm.

Frank wrote: "As long as is a common way of describing coordinate
systems we should be in good shape.  I am hopefully that OGC WKT could
be a common exchange format for the different systems.  If nothing else,
we should at least all support EPSG's dictionary."

I haven't done a lot of work with the OGC WKT, but GeoTools already uses
the EPSG Dictionary. Since I hope to contribute to the maintenance of
this code in GeoTools I would personally lean in this direction.

Besides, EPSG is maintained by a bunch of surveyors. :]

Frank wrote: " I'm not sure how critical this is.  State plane
coordinate systems are
based on a few fairly well known projection methods with parameters
available in the EPSG dictionary (as well as elsewhere).   There are
already
algorithm descriptions for a variety of projection methods from various
places.  One of them is the EPSG dictionary itself."

Maybe this was a premature move on my part. I know that the EPSG
database has a lot of this information in it. I'll wait to do anything
more with the pseudocode until I have a chance to become more familiar
with the EPSG database.

Frank wrote: " BTW, I have in the past (poorly) maintained the following
pages which
I hope will become part of the MetaCRS project and better maintained.

   http://geotiff.maptools.org/proj_list

They attempt to describe the projection methods, and relate them back
to EPSG codes, and various other coordinate system formats."

Let me know how I can assist. In the meantime, I'm going to bounce this
idea off the GeoTools list.

Landon




-----Original Message-----
From: Frank Warmerdam [mailto:warmerdam at pobox.com] 
Sent: Monday, January 21, 2008 1:51 PM
To: Landon Blake
Cc: metacrs at lists.osgeo.org
Subject: Re: [MetaCRS] Non-Programming Collaboration

Landon Blake wrote:
> I'm glad that I did not step on your toes. :]
> 
> I've also added an item for developer articles.
> 
> I think that "Common Spatial Reference System or Coordinate Reference
> System Names and Descriptions" and " Coordinate System (and CRS
related
> object) dictionaries. Stuff like the EPSG dictionary" might be the
same
> thing. Maybe I should delete the first one to avoid confusion?

Landon,

I think they are closely related but not necessarily the same thing.

> How would we have a test suite that was not specific to a programming
> language?

I'm imagining a CSV file with the following columns:

Projected Coordinate System ID: (possibly WKT or an EPSG id)
proj_easting
proj_northing
proj_z
same_datum_longitude
same_datum_latitude
same_datum_z
wgs84_longitude
wgs84_latitude
wgs84_z

Then a test run could use a particular engine, transform the point from
projected to lat/long (on the same datum), check, and then back to
projected coordinates and check again.

The test could also transform to WGS84 and back checking at both ends.

Some sort of tolerance might also be appropriate in the test records.

As long as is a common way of describing coordinate systems we should
be in good shape.  I am hopefully that OGC WKT could be a common
exchange format for the different systems.  If nothing else, we should
at least all support EPSG's dictionary.

> P.S. - I'm going to dig around for information about a pseudo code
> standard. It would be nice if we could describe transformations,
> calculations, and algorithms in a standard pseudo code. I could start
us
> off with some pseudo code for California State Plane Coordinates,
since
> I've become way more familiar with the system than any man should as a
> result of studying for my licensed surveyor exam. :]

I'm not sure how critical this is.  State plane coordinate systems are
based on a few fairly well known projection methods with parameters
available in the EPSG dictionary (as well as elsewhere).   There are
already
algorithm descriptions for a variety of projection methods from various
places.  One of them is the EPSG dictionary itself.

BTW, I have in the past (poorly) maintained the following pages which
I hope will become part of the MetaCRS project and better maintained.

   http://geotiff.maptools.org/proj_list

They attempt to describe the projection methods, and relate them back
to EPSG codes, and various other coordinate system formats.

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    | President OSGeo,
http://osgeo.org



Warning:
Information provided via electronic media is not guaranteed against defects including translation and transmission errors. If the reader is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this information in error, please notify the sender immediately.