In an engineering context, people should convince others with evidence. A test case showing an objective improvement would be ideal. Pontification about the rel

In an engineering context, people should convince others with evidence. A test case showing an objective improvement would be ideal. Pontification about the relative merits of disparate approaches is irrelevant. This can be avoided by having an benevolent dictator / architect make those decisions. While step one is getting the job done (executing even a bad plan is superior to no working product), I have noticed that sometimes the most efficient way to get the job done is to defer it until later, when the problem space is better understood. Tradeoffs are the great beauty of engineering and perhaps why the field successfully holds the attention of so many minds.