Robin Gower

Just to expand upon this, the [locating metadata](https://w3c.github.io/csvw/syntax/#locating-metadata) section of the spec defines the precedence: 1. overriding metadata (`csv2rdf -u`) 2. find via link header 3. find via [defaults/ configuration](https://w3c.github.io/csvw/syntax/#default-locations-and-site-wide-location-configuration)...

... and deploy to clojars?

Have you tried this in standard mode? I'm not sure why you're trying to label the `csvw:Column` why not use `csvw:titles`? If you want an RDF representation of the annotation...

I agree this would be nice to have. The complication is that we would need to handle all types, not just strings. We'd might want a list column whose values...

Yes the spec lets you specify a [`csvw:TableGroup`](https://w3c.github.io/csvw/syntax/#table-groups) which is effectively just: ``` { "tables": [{"url": "1.csv"}, {"url": "2.csv"}] } ``` The CSVW tests provide an example: ``` > countries...