When an architect is given a brief they first need to consider the appropriateness of the materials available to them. They should understand the functional and aesthetic benefits, and limitations, given the context they’re placing it into. For example, you wouldn’t apply brutalist techniques when designing a treehouse. The tree would buckle under the weight of concrete, a material best left for the Barbicans of the world.
As designers for screen we face similar choices around structure. The recent debate surrounding the hamburger menu and its place within a shared visual language comes to mind. Like brutalism it has a place, but like all solutions they’re designed to answer specific problems.
No amounts of A/B test results are needed to show that a collapsed menu which contains only three items might not be appropriate. While it shouldn’t stop you from trying, you don’t need to build a concrete treehouse to know that it might fail. Initially applying a solution out of context can help form new ideas, however, the results of such an experiment don’t render the original solution useless in all situations. To think that they do would be irresponsible.
As designers in a relatively young field we should be hesitant to reject ideas. Maybe the hamburger icon’s time will pass but as a community we should be willing to experiment. It’s the only way to expand our design vocabulary to be suitable for treehouses and Barbicans alike.