There are situations where it would be useful to have more than one ETL be able to modify an existing dataset. For example, I sometimes need to make a one-time fix via a query to a dataset. As it stands today, this is nearly impossible; the only way to do it is to hijack the existing ETL, comment out all of the code, put in the temporary code for the one time-fix, and then change everything back again once it's finished. It would be great if I could specify an existing dataset directly within the ETL.
