Difference between revisions of "GeoJSON Features"
From GeoJSON
(New page: Based on an initial outline at [http://icon.stoa.org/trac/pleiades/wiki/GeoJSON http://icon.stoa.org/trac/pleiades/wiki/GeoJSON] { "features": [ { "id": "alesia", "title": "A...) |
(No difference)
|
Revision as of 01:27, 13 March 2007
Based on an initial outline at http://icon.stoa.org/trac/pleiades/wiki/GeoJSON
{ "features": [
{ "id": "alesia",
"title": "Alesia",
"classname": "settlement",
"geometryType": "point",
"spatialCoordinates": [[47.535, 4.478, 0.0]],
"srs": "EPSG:4326",
"center": [47.535, 4.478, 0.0]
}
]
}
Possible Changes
- Should
geometryTypeandspatialCoordinatesbe rolled together? For example:
{ "features": [
{ "id": "point_example",
"title": "Point Example",
"point": [[0.0, 0.0, 0.0]],
"srs": "EPSG:4326"
}
{ "id": "linestring_example",
"title": "LineString Example",
"linestring": [[0.0, 0.0, 0.0] [1.0, 1.0, 0.0]],
"srs": "EPSG:4326"
}
{ "id": "polygon_example",
"title": "Polygon Example",
"polygon": [[0.0, 0.0, 0.0] [1.0, 0.0, 0.0] [1.0, 1.0, 0.0] [0.0, 1.0, 0.0] [0.0, 0.0, 0.0]],
"srs": "EPSG:4326"
}
]
}
- Should the GML (RDF-like) Object-Property model be used instead?
- Should the
srsbe at thefeatureslevel, instead of repeated for each feature? This would also force homogeneoussrs-s. - Should heirarchical (non-spatial) properties be allowed, or only flat "shapefile-like" structure?