I was reviewing my notes this morning. I found a folder full of saved links about LLM Wiki. Articles, Threads posts, and videos. A lot of captured material. None of it used.
I had months of saves. I could not tell you what any of them actually changed for me.
Writing is the mechanism for thinking. You find out fast what you actually understand when you have to put it into sentences.
When a topic catches my attention, I start a project around it straight away. It is an actual working draft, not a folder. I came across this idea reading about how Issac Asimov worked. He wrote to think, not after thinking.
Here is what that looks like in practice:
Open a blank doc and start writing what I already know. Even if that is one paragraph.
Research to fill the gaps. Read, watch, collect. Only to answer questions the draft has already raised.
Publish or file the draft as a post, newsletter, or note. The output forces the thinking to complete.
This sequence has one mechanical advantage: it removes the option to collect without processing. Every input has to earn its place by shaping the draft.
My LLM Wiki got finished this way. It took me a few sessions. I was stuck because the writing kept asking me what I actually understood.
I keep coming back to this question lately: what percentage of what I have saved have I actually used to build something?
Just another coffee thought from Gav
