Anita Graser 🇪🇺🇺🇦🇬🇪Appreciate the nod that spatial data viz is hard, @motherduck!
The motivation part about expensive GIS tools is a little off though 💸 
https://motherduck.com/blog/geospatial-for-beginner-duckdb-spatial-motherduck/
brandon
Sean Gillies@underdarkGIS I feel like I've read this same blog post a dozen times from Mapbox, Carto, Mapzen, etc. Every four or five years the modern GIS stack gets reinvented and we never really go anywhere new.
Kevin Smith@sgillies @underdarkGIS I think it's an example of the COBOL fallacy, in this case combined with an unwritten assumption of familiarity with relational databases (Although it is the subject of the blog so in context that may be reasonable).
What makes GIS hard is not the tools, it's that it involves complex and unintuitive fundamental concepts, just like programming. Making the tools easier is at best, a small but real improvement and at worst obscuring important aspects of the work.
@sgillies @underdarkGIS To be clear, I'm not accusing this of being anything like COBOL other than it feels like it's rather overselling itself. An RDBMS with spatial capabilities is better than one without all else being equal but It's still an RDBMS (with all the complexity and special knowledge required to use one) and still involves geospatial data and processes (with all the complexity and special knowledge required for that).
Paul Norman@smithkm @sgillies @underdarkGIS In theory you have the complexity of both, but I find most uses are complex spatial with limited use of relational and transactional complexity, or complex relations and transactions with simple spatial queries.