I spent the week building a knowledge tool, and the rule I keep coming back to is the one I almost never see anyone follow. Summaries restate. Insights connect. Those are not the same move, and most of what passes for thinking online is the first one wearing the second one's clothes.
The framing comes from Norman Webb, an education researcher at Wisconsin who in 1997 built a ladder called Depth of Knowledge. Four rungs. Recall. Skills and concepts. Strategic thinking. Extended thinking. It was an alignment tool for teachers, not a self-help thing, and it sat in education research for almost three decades without much crossover. A version of it got rediscovered recently by a school in Texas that turned it into a personal practice. The relabeled rungs are easier to hold in your head: facts, summaries, insights, new knowledge.
The ladder is useful because it draws a hard line between things people usually blur. A fact is a thing that's true. A summary is a few facts put into your own words. An insight is something that wasn't visible until you put two summaries beside each other. New knowledge is the rare thing that nobody else has put together yet, the kind of position that cuts against the consensus in a field.
Most knowledge work, including most of what gets called "research" or "thinking," lives on the second rung. Read three articles. Restate them in plain English. File the restatement somewhere. Done. That's a summary. It feels productive because there is now a document where there wasn't one. But nothing has been connected. Nothing has been put under tension. You ended the session with more files and the same number of ideas.
Here is the test I now use, which I borrowed from the ladder. If your note could have been generated by handing the source to anyone competent, it's a summary. If it depends on something else you've already filed, it's an insight. Insights have a "rests on" pointer. Summaries don't.
The reason this matters is not academic. It's about where time actually goes. When you sit down to think about a topic you care about, the easy work is summarizing. It feels like progress because the page fills up. The hard work is the move from summary to insight, which requires holding two or three different summaries in your head at once and asking what they have in common that isn't obvious yet. That move takes time. It feels uncomfortable. You can be at a desk for an hour and produce one sentence. Most people will not stay at the desk for that hour. They'll go produce another summary instead.
There's a third move that's even harder, and the ladder is right that almost nobody makes it. The move from insight to new knowledge. The first three rungs are about understanding what's already known. The fourth is about producing a position that wasn't known. By definition, an insight that everyone already shares isn't on the fourth rung. The position has to disagree with the field, or extend it, or notice a thing nobody noticed yet. It has to be something you would defend in a room where people who knew the field disagreed with you.
What I think is true, and what the tool I built this week is trying to make easier, is that the bottom rungs are getting cheaper in a way the top rung isn't. You can hand a stack of papers to a model and get a summary in seconds. You can ask it to compare three summaries and get a candidate insight in a minute. The first three rungs are increasingly automatable. The fourth one isn't, because the fourth one is the part where you bet on a position no model can generate from the prior alone. You have to have lived with the material long enough to know where the consensus is wrong.
The implication for how I want to spend time is uncomfortable but clean. Less reading and restating, more connecting. Less collecting, more pressure-testing. If I read three articles on a topic this week, I shouldn't end the week with three new summaries. I should end the week with one paragraph that says, here is what these three sources have in common, here is what I think they're getting wrong, here is the move nobody seems to be making. The summaries are scaffolding. They are not the building.
This isn't a writing tip. It's the same thing in every long-form discipline. Painters who copy other painters' canvases for years are doing rung-two work. Painters who notice what nobody is painting yet, and go paint it, are on rung four. Researchers who write literature reviews are doing rung-two work. Researchers who notice the gap nobody has put their hands on, and publish a paper on it, are on rung four. Writers who can summarize a debate well are doing useful work. Writers who write the essay the debate hasn't gotten to yet are doing the work that lasts.
The discipline that separates the two isn't talent. It's a willingness to sit with the discomfort of not having anything new to say yet, and to refuse to call the summary a finished thought. The notebook fills up either way. The question is whether it fills up with restatements of what you've read, or with the harder, rarer move that connects two things that were sitting in separate folders.
Most people stop at rung two and call it expertise. The best people I know are the ones who can tell, on a given day, which rung they're actually on, and refuse to count the lower one as the higher one.