I am relatively new to Microsoft Fabric. Overall very good. But I could find a lack of consistency in the user experience (which sometimes even has impacts on the organization of workspaces or the name of tables).
I will proceed from the most impactful to the most trivial (from my point of view):
1 - From a Dataflow, you cannot target a table in a specific schema of a DataLake. The notion of schema is brand new in DataLakes, so this lack is understandable. But suddenly it greatly limits the uses of schemas...
Note that the saveAsTable of a Notebook, manages the schemas well.
2 - A DataFlow can create a Delta table by respecting the indicated case. For example, the name "MyTable" will create in the DataLake, the table "MyTable". On the contrary in a Notebook, the saveAsTable creates the table with a name entirely in lowercase.
3 - The creation of a Data Pipe can be done from a folder. But creating a Dataflow in a folder is not possible, it must be created directly in the workspace. However, after creation, the Dataflow can be moved to a folder.
4 - When you are in Data Pipe, you can go directly to another element without any problem (left menu). But if you are in a Dataflow, if you do that you get an error message, because you must first close the Dataflow (with the cross at the top right). It is not homogeneous.
These are points of detail. Especially for the last two. But these are unnecessary disruptions for someone who is discovering the platform.
Best regards
