i was reading on FME Form Advanced module, in Generic writing section
and it has this part that mentioned :
 

When you use the Generic Writer, it is sometimes helpful to pre-configure parameters specific to individual output formats. For example, you must specify a seed file to write a Bentley MicroStation Design file.

To do this, add a writer to a workspace. This step causes Workbench to think that there is more than one output format and forces Workbench to create a Multiwriter for the workspace. The first writer in the multi-writer will be the Generic writer, and this one should have all the output feature types. The second writer will have no feature types but will be used to communicate any format-specific writer settings to the Generic writer when FME users that format for the Generic writer.

Suppose you configure the Generic writer to write to a format requiring a filename as a dataset. In that case, the Generic writer will turn the output folder into a filename by appending a file extension to the last portion of the folder name. When format-specific settings are present, it takes this extension from the dataset specified for the specific format. It ignores the rest of the dataset.

as far as i understand from the explanation above, we can trigger “multi-writer” using 2 writer (in this case generic writer as the main, and another type of writer to set a specific-format parameter/value), and when we run the workspace, if that specific-format chosen it’ll use that writer format/value that have been set before on that particular writer, but the data will only go through the generic writer.

i want to test his “multi-writer” scenario where i simply translate a gdb into generic writer with 3 option for the output format (shp, gml, and geojson), and when i choose GML as the output i’ll have pre-configure value that i’ve set before (Target Namespace Prefix), i tried with this step
1. add the first generic writer connected with the source gdb
2. add another writer (GML) not connected with anything but configured the “Target Namespace Prefix”
3. Run the workspace with generic output formate select to GML

the output is a gml file and it goes through the Genereic Writer, but it doesn’t take the “Target Namespace Prefix” value i’ve set on the GML Writer.

what am i missing ? or it just that i have different understanding of what its actually written on the module.

thanks