Software development is a weird place in some aspects. You can write a really long article like this one, telling people how to code, providing *zero* references to scientific studies to back up your claims.
Imagine someone writes a tutorial of how to build a skyscraper, providing no facts what so ever. How realistic is that? No, they research *everything*, right down to how thick a bolt needs to be to hold up the toilet door.
Software development guides in general, on the other hand, are mostly based on personal experiences. "During my time as a developer, I have come to this conclusion about X."
For example, you say that "dependency injection is not the best way to accomplish decoupling", without bothering to provide any evidence to backup that claim. No  and  notations. No need. You know this stuff. Don't use it.
Don't get me wrong, I agree with most of the stuff you have written in this article, but it would be great if our industry would lean on facts a bit more often.