Python Geospatial Development

GDAL and OGR

great circle distance

Geography Markup Language

Well-known Text (WKT)

WMS, WFS, and WCS

GML, KML, GeoRSS,

Open Geospatial Consortium, often abbreviated to OGC (http://www.opengeospatial.org)

PROJ.4, PostGIS, OGR, and Mapnik

that they directly support spatial data types, spatial queries, and spatial joins.
That spatial indexes generally make use of R-Tree data structures to represent nested hierarchies of bounding boxes.
That spatial indexes can be used to quickly find geometries based on their position in space, as well as for performing spatial comparisons between geometries based on their bounding boxes.
That MySQL, the world's most popular open source database, has spatial capabilities built-in, though with some limitations.
That PostGIS is considered to be the powerhouse of spatial databases, built on top of the PostgreSQL open source database engine.
That SpatiaLite is an extension to the SQLite serverless database, with a large number of spatial capabilities built-in.
That it is important to store the spatial reference for a feature, in the form of an SRID value, directly in the database.
That you need to choose the appropriate spatial reference for your data, depending on how much the Earth's surface your data will cover, and what you intend to do with it.
That you should avoid creating geometries and performing transformations on-the-fly as these kill the performance of your database queries.
That you need to be clever in the way you use spatial indexes to speed up your queries.
That you need to have an intimate understanding of your database's query optimizer, as well as any quirks or bugs in your database that may affect the way queries are executed.
That you can use the EXPLAIN command to see how your database will actually go about performing a given query, and to ensure that your query is structured correctly to make use of the available indexes.
That PostGIS needs to have the VACUUM ANALYZE command run periodically on the database so that the query optimizer can choose the best approach to take for a given query.
That SpatiaLite does not automatically use a spatial index, though it can be used explicitly in your queries to great effect.
That MySQL tends to be the slowest but easiest-to-use of the spatial databases, PostGIS is a workhorse that scales well, and SpatiaLite can be surprisingly fast, but is quirky and suffers from bugs.
That all three spatial databases are powerful enough to use in complex, real-world geo-spatial applications, and that the choice of which database to use often comes down to personal preference and availability