Quantcast

Boundingbox issue for discussion

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate
star

Boundingbox issue for discussion

kliu4
Dear Developers,

We are from George Mason University; and currently we are using GeoNetwork for our GEOSS Clearinghouse program. We found a boundingbox issue when we use OGC templete for CSW getRecords request as below:

Geonetwork defines boudingbox in /trunk/web/src/main/webapp/xml/schemas/iso19139/present/csw/ogc-full.xsl and /trunk/web/src/main/webapp/xml/schemas/iso19139/present/csw/ogc-brief.xsl

                                 <ows:LowerCorner>
                                         <xsl:value-of select="concat(gmd:eastBoundLongitude/gco:Decimal, ' ', gmd:southBoundLatitude/gco:Decimal)"/>
                                 </ows:LowerCorner>

                                 <ows:UpperCorner>
                                         <xsl:value-of select="concat(gmd:westBoundLongitude/gco:Decimal, ' ', gmd:northBoundLatitude/gco:Decimal)"/>
                                 </ows:UpperCorner>

But we checked the OGC:GML Encoding Standard(http://portal.opengeospatial.org/files/?artifact_id=20509), we think that the LowerCorner should be(west,south) and the UpperCorner should be(East, North)

"The first direct position is the "lower corner" (a coordinate position consisting of all the minimal ordinates for each dimension for all points within the envelope), the second one the "upper corner" (a coordinate position consisting of all the maximal ordinates for each dimension for all points within the envelope)." Page60.

Could you please look into this? Thanks and best regards,


Kai
Joint Center for Intelligent Spatial Computing
703-395-2337

------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev
_______________________________________________
GeoNetwork-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate
star

Re: Boundingbox issue for discussion

ByronCinNZ
Forwarding comments from New Zealand Geospatial Office in respond to these questions:


Hi Byron (and others)
Thanks for your question, I've had a look into the question (with help from Stephen - thanks) and have the following comments:
• ISO 19115 defines EX_GeographicBoundingBox as
o   + westBoundLongitude : Angle
o   + eastBoundLongitude : Angle
o   + southBoundLatitude : Angle
o   + northBoundLatitude : Angle
    i.e. The ISO standard defines a bounding box be described as two sets of lines of longitude and lines of latitude.
• In contrast, as quoted below the OGC defines a bounding box differently. The GML Standard (OGC 07-036) as quoted below defines the Envelop as a set of points the minimal ordinates followed by the maximum ordinates. I.e. the lower x-, y-values followed by the upper x-, y-values - corresponding to south, west then east, north - as suggested by Kai  :)
• In contrast (yet again), GeoNetwork as quoted below seems to define a bounding box differently.
So overall GeoNetwork seems to define the bounding box differently to ISO and OGC. Note however, all these definitions provide alternative and equally appropriate ways of defining a bounding box. And I gather that in practice it may well not matter what order the ordinates are provided in... parsing should be able to make sense of the coordinates, regardless of order provided...
 
In addition, I note that the OGC standard (07-036) seems a bit misleading. In particular, the standard refers p.60 to "minimal ordinates" and "maximal ordinates"... without defining ordinate. I suggest that it may make (slightly) more sense 07-036 was revised such that "coordinates" was used in place of "ordinates". Coordinates being a pair of ordinate (y-axis) and abscissa (x-axis) values.
 
I'll have to hand over to Richard to comment regarding the what if anything needs to be done re the "standards issue(s)" this raises and how we might contribute to the discussion...
 
Please feel free to post this answer on the GeoNetwork user forum.
 
Cheers,
John
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate
star

Re: Boundingbox issue for discussion [SEC=UNCLASSIFIED]

john.hockaday-2
On  Thursday, 1 September 2011 6:16 AM ByronCinNZ wrote:

<snip>

> In addition, I note that the OGC standard (07-036) seems a
> bit misleading.
> In particular, the standard refers p.60 to "minimal
> ordinates" and "maximal
> ordinates"... without defining ordinate. I suggest that it may make
> (slightly) more sense 07-036 was revised such that
> "coordinates" was used in
> place of "ordinates". Coordinates being a pair of ordinate
> (y-axis) and
> abscissa (x-axis) values.
>  

"ordinate" is a cartographic or surveying term that means one element in a coordinate system. Most people are aware that "coordinate" means a pair of values defining a location in a two dimensional spatial system but, as an ex surveyor, I understand "ordinate" to be one element of that pair or a distance along one axis. Hence a latitude or a longitude would be an "ordinate". A latitude and a longitude stated together would be a "coordinate".

Hence OGC's use of "minimal ordinates" and "maximal ordinates" makes sense for a bounding box. The "minimal ordinates" would be most Southerly latitude and the most Westerly longitude. The "maximal ordinates" would be the most Northerly latitude and the most Easterly longitude.  Hence the bounding box would be the North, South, East and West ordinates.

I hope this helps.


John

<snip>

> Cheers,
> John
>
>
> --
> View this message in context:
> http://osgeo-org.1803224.n2.nabble.com/Boundingbox-issue-for-d
iscussion-tp6742832p6748000.html

> Sent from the GeoNetwork developer mailing list archive at Nabble.com.
>
> --------------------------------------------------------------
> ----------------
> Special Offer -- Download ArcSight Logger for FREE!
> Finally, a world-class log management solution at an even better
> price-free! And you'll get a free "Love Thy Logs" t-shirt when you
> download Logger. Secure your free ArcSight Logger TODAY!
> http://p.sf.net/sfu/arcsisghtdev2dev
> _______________________________________________
> GeoNetwork-devel mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
> GeoNetwork OpenSource is maintained at
> http://sourceforge.net/projects/geonetwork
------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev
_______________________________________________
GeoNetwork-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork
Loading...