Similar recommendation to blog post writers: Try to include only relatively important links, since littering your post with links will increase effective reading time for many readers. Which will cause fewer people to read the (whole) post.
This is similar to post length: There is an urge to talk about everything somewhat relevant to the topic, respond to all possible objections and the like. But longer posts will, on average, be read by fewer people. There is a trade-off between being concise and being thorough.
Yeah, I view links the same way as I view the reference section in an academic article - they're there if you want or need to read up on a concept in order to understand better. But if you feel like you understand what's being said already, and don't have a particular need to dive into the details, you can just read without opening a single link.
(FWIW after I saw the exercise, I went back and clicked on all the links, because I prefer to open all the links, and don't want to practice the opposite mental move.)
I would encourage people intending to succeed at integrative research to practice getting through all the sublinks from an article in a limited amount of time - eg, for practice purposes, set ten 10-minute timers, read one seed article every ten minutes, try to get through enough sublinks to be sure you mentally loaded a syllabus of the sublinks. I'm sure there's a lot more good work out there in academia about how to get through a literature search for different purposes and how to practice being good at it; I might try to do some literature searching on that very topic myself, but whether I do or not, I'd love to hear any seed links folks have about how to learn to do a slow and solid literature search and when and how to instead do a fast and time-boxed lit search.
I find this approach very useful when attempting to understand a new subfield (eg neural implicit representations) entirely by reading the most recent papers in the subfield, something I do often.
There are a lot of blog posts that have >5 links to other blog posts. I used to click on all of these, and upon seeing that I hadn't read them, I would embark on a several hour long journey of reading until exhaustion, often times without even finishing the original post I started.
Recently, I had the simple realization that I could usually read a blog straight through without ever reading one of the linked posts while still understanding the content of the blog.
I wanted to share this with other lurkers in hopes that I could make reading on LessWrong more enjoyable and efficient.
Exercise for the reader: don't click the links!