On 09.04.2012, at 18:14, Iván Sánchez Ortega wrote: > On Lunes, 9 de abril de 2012 15:56:28 Stefan Keller escribió: >> And even for webservices: When asked which vector geodata file format >> to choose for the response part to make a project interoperable.... > > > If you ask that to a gov't agency person, he'll say "OGC standards", because > any gov't agency understands OGC. > > If you ask a wev developer, he'll say "GeoJSON", because any web dev worth > his > salt understands JSON. > > If you ask a operator using ArcMap, he'll say "Shapefiles", because any > person > using ESRI stuff likes shapefiles.
Nowadays ESRI tells not to use Shapefiles, but Geodatabase. They may have http://resources.arcgis.com/content/geodatabases/10.0/file-gdb-api, but who uses it really? > If you ask hipster neogeographers, they'll say "Spatialite", because all > hipster neogeographers liked spatialite before it was mainstream. > > If you ask a DBA, he'll say "Gimme the database dump", 'cause all DBAs > understand database dumps. Unless it is neogeographer DBA who wants PostGIS. > You ask about interoperability. And I dare to ask back: interoperability > between *who*? The only de facto interoperable geodata standard is Shapefile. With all its issues. There is nothing similar in webservice space. In fact all standards there describe geographical primitives only, and data around it remains application-defined. For interoperability you need full interface conformance, but this is already outside geographers specific interests. OCG/GML, GeoJSON, WKT are not complete data formats like Shapefile, just describe point/line/polygon elements in it. There is KML which is remotely similar and quite web-friendly, but for geographers it is piece of joke. Jaak _______________________________________________ Geowanking mailing list [email protected] http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
