A quick lightning style talk to present to a small community - not with a solved approach but rather attempting to debate the role that an Effect System could play in making architectural layers constrained in the code base directly. E.g. How can we encode the constraint "You shouldn't be able to write to the database directly from a request handler". The audience has very little haskell experience, and the presenter is not a full-time nor expert haskell programmer - but of course (as we all know) haskell is categorically the best language to experiment with these ideas ;)

Obviously Effect Systems are not perfect, and the talk was not meant to be some sort of tutorial - but rather to try and build an intuition of why they exist, and a very simplified model for how they work - with the hope that it sparks some interest and that individuals see them as being something worthwhile to look at when attempting to surface architectural boundaries within a code base, and MAYBE this can keep technical debt lower over time?