I'm trying to weight the differences between implementing a solution using DTO's between the presentation and the service layer or simply allowing each layer to have knowledge of LLBL Entities.
It seems that if I want to use DTO's (which I do for a future web service layer, there just so simple to serialize without all of the added features of an Entity) I'll need to implement most of the features of an entity (tracking new, dirty), create some type of UnitOfWork for DTO's, etc.
What is the consensus to this, do you think it is worth it to generate the DTO layer and then create the functionality for a DTO UnitOfWork and implement tracking on the DTO objects or just use the Entities provided in the framework?
-Kas