Currently, the plots for workshops are made from the raw CSV file, while the data feed used in our website is from a modified version that filters out workshops that don't include an URL. As much as possible these 2 data sources should be identical.
I think our query is also doing an inner join of workshops & instructors (and therefore excludes workshops where instructors have not yet been recorded) while the AMY feed includes those workshops.
Currently, the plots for workshops are made from the raw CSV file, while the data feed used in our website is from a modified version that filters out workshops that don't include an URL. As much as possible these 2 data sources should be identical.