I originally set out to write because “it’s valuable for me”. That was a mistake; such a hand-wavy ‘why’ makes writing a low priority. In practice, my most urgent driver has been the fear that without writing about a topic, I don’t fully understand it (1). LLMs make this kind of writing easier, but worse, unless we treat them like human reviewers.

Motivation

I didn’t expect that impostor syndrome would be a good trigger for writing. Reframing this fear can be as easy as starting from a genuine novice mindset, and getting a more objective assessment of my understanding from others.

For example, in some 1:1s, I was sometimes asked for advice on a topic I thought I knew well, but I felt like my answers were too shallow. Writing up what I wish I’d said, including sometimes with more research, really helped me and helped my team.

The problem with today’s LLMs

I wrote a lot more with LLMs, which increased word count, especially for internal docs. It helped cover my bases for template documents like PRFAQs. LLMs can ask good questions, and fill in some blanks to make documents easier to read.

I found talking to ChatGPT for first-drafts to be efficient, but dull. My intuition is that when you ask ChatGPT to help with a task, you will get worse, not better, at that task, which has been studied already (2).

Only once I have a fully hand-written shareable draft will I ask others for feedback. At that stage, it makes sense to ask an LLM, alongside the humans whose perspective I value.

The painful iterations of a good first draft

Once I’ve read around the subject matter and let the ideas percolate, the actual text evolves across 2-5 writing sessions, covering multiple aspects of the problem. If I’m inspired, it’s easier to explore a new facet. If I’m lower energy, it’s easier to criticise and edit.

After a few rounds of diverging/converging around an idea, I don’t feel like I’m done, but I feel confident enough about the topic to check in with people who know more about it than me. Ultimately, I’m inspired by the (‘based on a true story’) ceramics parable from Art and Fear (3), where students who worked on a new project every day for 30 days had much better results than students who worked on 1 project for 30. Fairly quickly, I’ll learn more from my next post, than from my iteration on the current post.

These perspectives are quite personal; your mileage may vary. If nothing else, there are some great essays about the benefit of writing, which I read or re-read in researching this post, that I can only recommend (4).

PS - in the interest of transparency and learning, here’s my ‘clear writing’ answers based on Shreyas’s advice (5)

What am I trying to say: Here’s a better “why” and “how” for me to write

Why do people care: Can helps someone like me, thinking about writing, write more, and write better

What’s the most important point: Good context (motivation, tools, mindset) makes writing easier and more helpful

What’s the clearest way to get that across: Describing my personal experience with writing recently

How should a reader feel: more knowledgeable. If they are thinking about writing or reflecting on their writing, they should feel inspired

What should the reader do next: Pick a topic they think they know and write it up; reflect on their own writing

---

(1) Words

(2) Your brain on chatGPT.

(3) Quoting the parable in full:

A ceramics teacher announced on opening day that he was dividing the class into two groups. All those on the left side of the studio, he said, would be graded solely on the quantity of work they produced, all those on the right solely on its quality. His procedure was simple: on the final day of class he would bring in his bathroom scales and weigh the work of the “quantity” group: fifty pound of pots rated an “A”, forty pounds a “B”, and so on. Those being graded on “quality”, however, needed to produce only one pot — albeit a perfect one — to get an “A”. Well, came grading time and a curious fact emerged: the works of highest quality were all produced by the group being graded for quantity. It seems that while the “quantity” group was busily churning out piles of work – and learning from their mistakes — the “quality” group had sat theorizing about perfection, and in the end had little more to show for their efforts than grandiose theories and a pile of dead clay.

Source

(4) Great articles about writing and thinking:

• Paul Graham; Writing advice.
• Paul Graham; Good writing matters especially when LLMs can approximate it;
• Julie Zhuo; Write better

(5) Clear Writing, a thread