With Didgets, tags are an integral part of the system. They don't get lost or forgotten when you copy a file from one place to another. Searches are a native part of the system as well so you aren't relying on a separate indexing service that has its own database somewhere else. BTW, managing file data using folders and tags are just a few of the features of the system. I found out the columnar stores I used for tagging, were easily used to also form relational tables. I can load in a 100 million row, 40 column table and do queries against it much faster than the same data loaded into Postgres, MySQL or SQL server.