All I know about writing — double square brackets.
I have sporadically published short fiction in various, mostly Australian, magazines for more than twenty years. I average less than a (published) story a year, however, so I would not call myself a writer. So I would not give writing advice.
Adding my few stories to my technical writing, though, has given me enough to experience to recommend one simple habit.
When the words are coming nicely, whether it’s fiction or a scientific argument, and I need to look up a reference, check whether I need to say ‘shroud’ or ‘stay’, or I have just realised that something I did three pages ago needs to be changed, I just [[make a note in double square brackets and]] keep going.
In a first draft of mine, gold melts at [[XX]] degrees, a minor character comes from [[some city near Tokyo]] and so on.
What works for me about this approach, and I am not pretending this is novel, is that as long as I am reasonably comfortable that there is an answer, I can get on with the job. Filling in the [[notes]] (like [[that plumber I mentioned in chapter 3]]) is something to do when the time for creative heavy-lifting is over.
Now sometimes the thing in the brackets is going to be a hinge for the rest of the article or story — in which case I just have to find out. But if it is not so crucial, I find this helps me get on with it. Sometimes it is useful just because I am in a mood to look for excuses to dawdle over the work, and sometimes because I am feeling more ‘in the flow’ and don’t want to stop.
I like the [[…]] notation because character combinations [[ and ]] are unlikely to occur in any document as a genuine part of the text, so I can search for any occurrences easily. Secondly, typing [[ is much quicker than blocking text in in red or bold or anything, and red and bold are not available on a Neo or in a text editor anyway. The notation stands out reasonably clearly to the eye, and works equally well within a LaTeX document or a Word document. I can search for it within my editor or I can grep for it, and so on.
That’s it. I don’t think I am qualified to talk about characterisation, plot, style, or pacing. I could perhaps write my own idiosyncratic guide to assembling a scientific paper, but I suspect that is highly redundant.