tag:blogger.com,1999:blog-4741970181714760187.post8529971250433521106..comments2008-04-11T13:33:22.272+01:00Bjørn Sandvikhttp://www.blogger.com/profile/16080335362672606377noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-4741970181714760187.post-55957102495090756752008-04-11T13:33:00.000+01:002008-04-11T13:33:00.000+01:002008-04-11T13:33:00.000+01:00Hei Michael,<BR/><BR/>I've not yet had time to look at a SLD/WMS solution. I will do at a later stage of my thesis. <BR/><BR/>You might have a look at European Interactive Atlas Explorer: http://karlinapp.ethz.ch/wms_client/<BR/><BR/>The top layers shows diagrams generated using SLD (look at the source code). <BR/><BR/>The diagrams are generated server side by the open source QGIS map server: http://karlinapp.ethz.ch/qgis_wms/<BR/><BR/>Maybe this tool will improve the performance? Please let me now if you have any success.Bjørn Sandvikhttp://www.blogger.com/profile/16080335362672606377noreply@blogger.comtag:blogger.com,1999:blog-4741970181714760187.post-44143605489548059362008-04-11T11:00:00.000+01:002008-04-11T11:00:00.000+01:002008-04-11T11:00:00.000+01:00Hi,<BR/>I have also done some experiments with OpenLayers vector-rendering for making Thematic maps but I'm having problems with performance when I have more than 100 polygons. I'm now working on a solution based on a WMS-service where a SLD-file is dynamicly generated on the server based on the request from the client.<BR/><BR/>Do you have any experiences on this?<BR/><BR/>The filtering technique in SLD is though rather limited. When doing the maps I first select eg 200 polygons in a layer that should be vizualized. Then on these 200 polygons I use a specific thematic rendering-scheme. When doing this in SLD I have to repeat the filter for each thematic class which I guess will degrade performance.<BR/><BR/>Michael ÖstlingAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-4741970181714760187.post-91797435952887433492008-04-02T18:49:00.000+01:002008-04-02T18:49:00.000+01:002008-04-02T18:49:00.000+01:00I agree with this clarification!Bjørn Sandvikhttp://www.blogger.com/profile/16080335362672606377noreply@blogger.comtag:blogger.com,1999:blog-4741970181714760187.post-57642891632305585552008-04-02T18:44:00.000+01:002008-04-02T18:44:00.000+01:002008-04-02T18:44:00.000+01:00"""<BR/>Conclusion: GeoJSON has a lot of potential, but is currently not suitable for world maps due to browser restrictions.<BR/>"""<BR/><BR/>I disagree with this conclusion. GeoJSON is no worse than any other vector format. This is purely a browser limitation.sgillieshttp://zcologia.com/sgillies/noreply@blogger.com