* [2010-12-02 13:23:30 -0700] Wood, Jamey <[email protected]> écrit:
] Thanks for your response. I wonder if anyone could comment on how ] difficult the remaining work would be. Since the R-Tree indexing part has ] already been implemented, I wonder if the rest would largely just be a ] matter of integrating various higher-level functionality from an existing ] geospatial library (such as GEOS [1]--which I believe is what PostGIS ] relies on for much of its underlying functionality). I don't know about the implementation details for Virtuoso, but just some thoughts on how PostGIS implements search operations. Basically the operands are changed into bounding boxes -- both the query argument and the index are changed into rectangles. A first rough match is done on these rectangles and a list of candidate results is returned. The more computationally intensive overlap/intersection/etc of the more complicated shapes is then done against the candidate result set (which is almost always very much smaller). It seems to work pretty well. ] As for market interest, I would think there could be a lot of ] applicability in the energy space (where geometries are used to capture ] information about natural resources, transmission lines and other ] infrastructure, utility service areas, etc). I would think so. Also the outside plant management departments of communications infrastructure providers for some of the same reasons, emergency services (cf. the beautiful story from the Amsterdam fire department about how open data saves lives), land transport companies (especially rail which tends to have lots of rights of way and so collateral infrastructure to keep track of, e.g. fibre routes), in some places forestry and conservation (there's a reason PostGIS is Canadian), postal services and couriers, the list goes on... One challenge is that we still have no consensus on how to express this type of data and these operations in RDF, but that's a separate issue from support in the database itself. Cheers, -w -- William Waites http://eris.okfn.org/ww/foaf#i 9C7E F636 52F6 1004 E40A E565 98E3 BBF3 8320 7664
