I agree, but I would put it differently. It is a lot easier to advocate a conclusion with full control over the assumptions, than discovering the boundary assum

I agree, but I would put it differently. It is a lot easier to advocate a conclusion with full control over the assumptions, than discovering the boundary assumptions where the opinion no longer holds. For example, it is easy to argue for "HTTPS is more secure than HTTP". But what I am often interested in -- as a curious and skilled IT person, and not a lambda user -- is what attack models does HTTPS defend against. Remove assumptions one by one, when does HTTPS offer equivalent security to HTTP. This is not just nitpicking, but extremely relevant for our profession. TLS has gotten under quite a lot of fire lately, so sooner or later, someone somewhere will discover those hidden assumption. I rather it happens on a public forum.