>People who want to learn about a subject at depth are still helped by simple and accessible descriptions.
I don't see this as being always true. For example, scientific papers are written in such a way that normally you need a great deal of understanding in a field to read them. Making them accessible would mean either removing or explaining all domain specific knowledge in each paper. Expecting the reader to acquire the knowledge elsewhere and then read the paper with the knowledge makes a group of papers more accessible than if each paper did either of the options making them simple.
In the same way, the wikipedia page on integration by parts does a better job of explaining what it means than if it took the time to explain what a function is, what a variable is, what an integral is, what a product is, what an antiderivative is, or what the common notation it uses. I bring these up because these are all assumptions made before you even get to the table of contents of the article. If the article was simple enough that someone who didn't know any of those concepts could understand the same information presented in the first paragraph, you would have a significant introduction to mathematics that would slow down those who have some basic calculus knowledge seeking to refresh or get a summary of what integration by parts is.
