Thank you for your thoughts. The MVP concept remains a kind of basket of meta-ideas that often lack proper rigor and definition, so anyone trying to shed light on the concept is doing us all a service.
This seems like an argument for what might be called “proper design,” which has nothing to do with viability unless you define viable as being properly designed.
Even that is a contentious claim as surely design can serve any purpose that it is possible to achieve with design. It must depend on what viability means.
In my experience, the problem many teams struggle with is how to define what viable means. Actually, and especially inside of existing companies (versus start-ups) many of them do what you suggest. They assume that viable includes meeting all the quality and reliability attributes of mature products and so spend too much time and investment on the product without, as the MVP advocates like to say, testing the core hypothesis.
There is an understanding that “Early Adopters” are tolerant of certain inadequacies as a trade-off for early access. This is a plausible and well-tested principle, so I think we can still rely upon it.
Perhaps one of the challenges is when early adopters don’t know that they’re early adopters. Many of the signals are perhaps missing in the IoT field because there are so many products that it seems unlikely (to consumers) that the tech is still immature, which it clearly is.
As it happens, you have highlighted a particularly difficult category that is dogged by potentially systemic reliability problems, like using unlicensed RF spectrum and fragmented technology stacks.
Your elucidation of classic versus romantic seems a useful metaphor and gives much food for thought. However, there is a reason for the imbalance, which is that achieving the romantic requires significant, almost disproportionate, effort.
p.s. My “video door bell” of the type you photographed never worked reliably :(