The simple example at src/pages/basic-pipeline.md (visible at https://www.nextflow.io/example1.html) says that:
Line 42: The outputs of splitSequences are passed as inputs to the reverse process, which processes each split file in parallel.
But this doesn't happen, because the workflow is lacking a "flatten" step, so the "reverse" process runs just once and receives all the files as input.
So I think either the workflow or the description should be fixed!