[x3d-public] [geospatial] Latitude/Longitude units

Roy Walmsley roy.walmsley at ntlworld.com
Wed Apr 13 10:47:42 PDT 2016


Andreas,

 

Thanks for clearing that up. I wasn’t hinting you should suggest words, although I very much appreciate them. Rather, I was wanting to fill a gap in my understanding.

 

But, I also agree that they should be reviewed by the geospatial working group.

 

Roy

 

From: Andreas Plesch [mailto:andreasplesch at gmail.com] 
Sent: 13 April 2016 16:09
To: Roy Walmsley
Cc: Holger Seelig; geospatial at web3d org; X3D Graphics public mailing list
Subject: Re: [geospatial] Latitude/Longitude units

 

Hi Roy,

it would be nice to get input from other members of the geospatial working group.

If I would remain the only voice in the geospatial working group to have an opinion on the angle units question, should the proposed change then be made ? At first glance, it breaks backward compatibility but in effect the text change would maintain it, as de facto use is units of degrees even in 3.3.

 

I agree that  the specification of the *_first options should be moved to section 25.2.3 as they relate to the geoSystem field. The choice of ellipsoid is completely independent.

What about this ?

Section 25.2.3 second paragraph:

 

Add after Geodetic spatial reference frame (latitude/longitude):


The order of latitude and longitude in geographic coordinates is controlled by an optional argument. If "latitude_first" is specified, the order is latitude then longitude. If "longitude_first" is specified, the order is longitude then latitude. If neither is specified, "latitude_first" is the default.

 

Section 25.2.4 second paragraph:

remove:
"If "latitude_first" is specified, the order is latitude then longitude. If "longitude_first" is specified, the order is longitude then latitude. If neither is specified, "latitude_first" is the default."

Perhaps link the word geoSystem in first sentence to Section 25.2.3.

 

and the exact equivalent for UTM (third paragraphs).



Andreas

 












 

 

On Wed, Apr 13, 2016 at 10:33 AM, Roy Walmsley <roy.walmsley at ntlworld.com> wrote:

Thanks for this great discussion. I can see that text modifications are likely to be required here.

 

In addition, I would like to raise a related question. In 25.2.4 Specifying geospatial coordinates, the ‘GD’ option allows either “latitude_first” or “longitude_first”, the order being controlled by the geosystem field. There is no other specific mention of an option within the geosystem field to specify either of these. Should there be? Or is it automatically determined by the choice of ellipsoid from Table 25.3?

 

Under the UTM option the option of “easting_first” or “northing_first” is an analogous case. What is the answer for this case?

 

Roy

 

From: geospatial [mailto:geospatial-bounces at web3d.org] On Behalf Of Andreas Plesch
Sent: 13 April 2016 14:30
To: Holger Seelig
Cc: geospatial at web3d org; X3D Graphics public mailing list
Subject: Re: [geospatial] Latitude/Longitude units

 

Good point. Section 25.2.4 could be read that way as it does point out that the angle unit is a choice.

The problem with using angle standard units is that authors would want to use radians for orientations and rotations but at the same time degrees for latitude/longitude. In my experience radians is very rarely used as a unit for geodetic coordinates. The only example example I can think of is in equations.

Some options:

1) no changes to the spec., modify examples

2) prescribe degrees as unit for geodetic coordinates in section 25.2.4

3) add a standard unit for angular distance in Table 4.2 (http://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/concepts.html#t-Standardunits) with a default of degrees and refer to that in 25.2.4

My preference would be option 2), followed by 3).

Therefore I propose that in section 25.2.4, second paragraph:

" Latitude and longitude are given in units of angle base units. The following assumes an angle base unit of degrees. If a UNIT statement for angle base units has been provided, the following values for latitude and/or longitude should be suitable converted to that angle base units."

shall be replaced by:

" Latitude and longitude are given in units of degrees."


for simplicity.

Andreas

 

On Wed, Apr 13, 2016 at 6:21 AM, Holger Seelig <holger.seelig at yahoo.de> wrote:

My interpretation was that with version 3.3 the UNIT statement defines what unit a angle value has. And therefore on default the angles of geodetic coordinates should be in radians. The author has then with the UNIT statement the choice to define its preferred angle unit.

Holger

Am 13.04.2016 um 06:16 schrieb Andreas Plesch:

http://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/components/geodata.html#Specifyinggeospatialcoords

defines that coordinates in the GD geoSystem are given in angular base
units.

http://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/concepts.html#t-Standardunits

defines the standard angle unit to be radians.

Cobweb interpretes this to mean latitude/longitude in geoCoords are
given in radians which I think should not be the intention of the spec.
All geospatial example scenes use degrees. So I hope this is a
misinterpretation.

Although section 25.2.4 as linked goes on to assume that the base
angular unit is degrees, it is a bit unclear what unit is really expected.

If more precise language is needed, I suggest simply to add "angular
base units of degrees" to section 25.2.4 .

Andreas



-- 
Holger Seelig
Mediengestalter Digital – Digital Media Designer

Scheffelstraße 31a
04277 Leipzig
Germany

Cellular: +49 1577 147 26 11 <tel:%2B49%201577%20147%2026%2011> 
E-Mail:   holger.seelig at create3000.de
Web:      http://titania.create3000.de

Future to the fantasy ★ ★




-- 

Andreas Plesch
39 Barbara Rd.
Waltham, MA 02453




-- 

Andreas Plesch
39 Barbara Rd.
Waltham, MA 02453

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20160413/26f27b76/attachment-0001.html>


More information about the x3d-public mailing list