As I write posts on this blog, I'm slowly familiarizing myself with the challenges of writing. I'm quite happy about having learned a while ago, i think in some YouTube video, that one can just hammer out the words that come to mind without stopping. This is an incredibly fun kind of writing.
I can imagine it being very difficult when perfection is the objective. Sitting there in front of an empty editor buffer seems stressful.
Recently I got some confirmation for this strategy of simply hammering out the thoughts at speed by Richard E. Nisbett, in his book "Book: Mindware by Richard Nisbett". He quotes someone saying that for any work to start happening on writing one must have something written down, even if it is the most ridiculous rough draft. The subconscious needs to get primed or triggered to be able to tackle the task of forming a coherent structure out of the disconnected thoughts that initially flow to the page.
Up to this point I've been only hammering out stuff. So everything that can be read here until mid 2025 is something of a direct readout of my brain. Now for high quality and persuasive writing I'm sure we can agree that something more needs to happen.
Today I tried to move on to this second step of refining the jumble that comes out in the initial hammering phase. It was quite hard. I think I spent like a good half hour on a tiny paragraph in the Trying to synthesize learnings on life design post. And the result wasn't even any good. So I'll have quite a bit of learning and practicing to do on this. Let's think about what this second step of refinement entails. For that it might be useful to zoom out a bit and look at the writing process as a whole.
Why do I write in the first place? To get a better understanding of things I learn, be able to make connections between different topics and ideas, and to arrive at new insights that help me better understand the world. It's also about improving my memory. Not by providing a written record (although that's useful too) but by consciously processing the stuff that's floating around in my head. Clive Thompson also talks about how writing can help organize ones thinking. More on this book here: "Book: Smarter Than You Think". Another aspect is that writing like this is just a lot of fun.
So that's my motivation as a hobby blogger. What might a person who would eventually like to make money by writing or someone like a software engineer who relies on good writing skills on the job need to get from his writing practice? A coherent structure would be nice to have. Also persuasiveness, to get the reader to buy into the ideas I might want to advocate.
What I think I need as a next step is to improve on the practice to make my arguments clearer to myself, and to question assumptions to get closer to a truth. I think Charlie Munger is attributed as saying something like "I don't allow myself an opinion on something unless I've thought all the thoughts that can be thought against it." I'm very sure what he said is different but the idea was something like that. I'll have to look it up in Shane Parrish's book where I think I stumbled across that one.
Notice how I am now linking to other posts in this blog? That's because now I've set up a separate vimwiki for my blog. See this post for the process Improve blogging.
I guess that thus far my writing has been useful in somewhat organizing my learnings better and to improve my recollection by performing active recall on the books I've read and ideas I collected elsewhere. And it has been fun.
While writing this I notice that I haven't got a large store of knowledge about writing. I might not have read anything specifically focused on writing. That's a gap I'll have to close. To start with, I'll get me a nice overview from GenAI.
Now let's get to the challenge I had while editing the brain dump in an earlier post. First of all I had to recollect what my purpose was in that specific piece. I had to know the context in which that paragraph stood and how it related to the rest of the post. I think a good way to go about this would be to get a high level mental representation of the post by creating clear separations and chunking parts. These chunks can then be labeled with headings to make it easier for the mind to get a handle on the post. In this step it might already become obvious that some parts are duplicated or would fit in at another place in the post.
Here we might already be able to move around chunks. But this would be a destructive action since it might break dependencies to other parts that we're previously below the moved part and now no longer make sense. So I'd like to separate the conceptualization step from the action step where I actually make changes. I like how there's a connection coming up to programming.
There seems to be some value here: The writing itself is what went incredibly quickly when creating the first draft but slowly when editing. So I'd like to move the actual writing to the end of the editing process.
We now have a two step process of hammering and editing. Thinking about it like this makes it seem that we we get only one pass at hammering. Editing is brownfield work. We'll see if we can get us a process that allows us to go back and forth between the two steps.
Let's break down the edit step: First I'll look at the purpose of the post. I'll get an overview of what's written and break it down into sensible chunks. This happens by inserting line breaks. Then I'll look into the resulting chunks in more detail and find appropriate headings. By this point I will already have decent mental map of the post.
Perhaps a mode of diverging and contracting like in the double diamonds method from design thinking can come in handy here. After contracting in the chunking and labeling step, I could diverge and check for any missing ideas or thoughts that would fit in with the post.
Maybe the other way around is better, to contract at this stage and find the true purpose of the post and close in on the main idea that needs to be transported. To lend a term from Mortimer J. Adler's Book, "How to read a book", I could try to find the "unity" of the post. The unity is a description of a book only a few sentences long.
I'm trying to get to the point where I can start writing again after the conceptualization phase but am struggling to find a segue there. The place I want to be after the conceptualization phase is to be as close as possible to be able to hammer again. But this time with better structure and stronger arguments.
Maybe an iterative approach could work, where I repeat hammering and conceptualization phases until the result is a strong written piece. That would entail a lot of hammering. My first thought was that this sucks and is a lot of work but I'm now kind of excited. This is a grand example of a drill as specified in Ultralearning by Scott H. Young (See "Book: Ultralearning"). So maybe that's the way to go.
To fantasize a little, wouldn't it be nice that at some point after extensive practice I would have the conceptualization phase so ingrained that it would happen subconsciously at the time of the first hammering? So I would end up with a polished post after half an hour of hammering out and enjoying myself.