I feel the blog argument is more towards any structured data assimilation/curation itself. The set of problems/hurdles posed by the authors also exist when you are creating a schema (or a set of schemas) for structuring your data in RDBMS. Actually even more. EKG as a solution is a step towards that.