I tagged every blog post by cognitive mode. Most of mine are knowledge-telling — and that's fine.

I added a single optional field to my blog’s frontmatter. By the end of the afternoon, I had a meta page that classified my own posts into four groups, and the distribution was unflattering in a way I hadn’t predicted.

Here’s the field:

kt_mode: knowledge-transforming   # or knowledge-telling, or mixed

And here’s the schema change, in Zod for an Astro content collection:

kt_mode: z
  .enum(["knowledge-telling", "knowledge-transforming", "mixed"])
  .optional(),

That’s the whole new contract. Optional means existing posts don’t break; omission means “I haven’t classified this yet.” A new page at /lenses/ reads the field and groups every post under one of four headings: knowledge-transforming, mixed, knowledge-telling, unclassified.

The Astro snippet is unremarkable — getCollection, group by a property, render <Card> per post. I cribbed the structure from my own /archives/ page and replaced the “group by year” with “group by kt_mode ?? 'unclassified'.” About forty lines, including the intro paragraph.

What’s worth talking about is what the field is for.

The categories aren’t mine

They come from Carl Bereiter and Marlene Scardamalia, The Psychology of Written Composition (1987). Their central claim, simplified:

Two distinct cognitive processes underlie all written composition, and most writers default to the simpler one without noticing.

Knowledge-telling is the retrieve-and-write loop. You locate what you know about a topic, locate the rough shape of the genre, write down what comes up, and stop when nothing new surfaces. The output is coherent. Your beliefs don’t change. Cognitive cost is low. Most tutorials, most how-tos, most “here’s a thing I built” posts are knowledge-telling, and that’s appropriate — they’re not arguing anything.

Knowledge-transforming runs two problem spaces in tension. A content space — what do I actually believe? what’s the real shape of this claim? — and a rhetorical space — how should this be said, to whom, toward what end? Each space’s tentative answer constrains the other; the writer iterates between them. Beliefs shift during the writing. Cognitive cost is high. If you can predict the post at the start, you weren’t doing this — you were doing the first thing, well.

B&S’s empirical claim is the uncomfortable one. Most writing, including by competent practitioners, is knowledge-telling. Knowledge-transforming requires deliberate effort, slower pacing, and explicit problem-framing. It doesn’t happen by accident.

The reveal

I started with five pilot classifications — two I felt confident were knowledge-transforming, one mixed, two knowledge-telling. The remaining twenty-four sit under “unclassified” because I haven’t gone through them yet.

That ratio is itself diagnostic. I have written more than I have deliberately transformed. If I’m honest about which posts to classify which way, the unclassified pile is probably mostly knowledge-telling — explainers, tool intros, technique summaries — and that fraction is going to grow.

This is not a moral failing. Knowledge-telling is what most published technical writing should be. The failure mode B&S names is not “writing knowledge-telling posts.” It is mistaking one for the other — treating a knowledge-telling artifact as if it were substrate-bearing thinking. Or, conversely, treating an opinionated essay as if it were a how-to, and being annoyed when it doesn’t tell you what to type.

The categories are useful precisely because they expose this confusion. You can’t tell from the surface. Both kinds of post can be well-written, well-structured, fully formed. The difference is in the cognitive process behind them — and that process is invisible to the reader unless the writer marks it.

So I’m marking it.

Why this matters in 2026

Here is the version of the framework that matters most to me right now.

Large language models perform knowledge-telling natively. That’s what next-token prediction structurally is — retrieve plausible content that fits a topic and a genre, write it down, repeat. An LLM has no content problem space because it has no beliefs to revise. It can produce well-formed text indefinitely, on any subject, without anything changing for the model in the process.

This is fine when knowledge-telling is what you wanted. It’s fine for boilerplate, summaries, drafts you’ll rewrite, restatements of well-trodden material. The technology matches the task.

The failure mode is when knowledge-telling masquerades as the other thing. An LLM-written essay can have all the surface markers of knowledge-transforming — caveats, opposing views considered, conclusions that update mid-piece — without anyone’s beliefs having changed. The surface is dialectic; the substrate is empty. It reads like thinking happened; nothing was thought.

The load-bearing human contribution under LLM co-authorship is the part LLMs don’t do. The content-problem-space work. The “do I actually believe this?” question. The willingness to be wrong on the record. That’s what kt_mode is trying to make legible — not as a quality grade, but as a process disclosure.

Try it

If this resonates, here’s the small thing I’d suggest:

Add one optional frontmatter field to your blog or notes system. Any classification axis that names something invisible from the surface — cognitive mode, co-authoring posture, genre, audience, exigence, your own taxonomy. The constraint is that the categories should be hard to fake and hard to derive from skimming. The act of choosing the value should require an honest look at what you actually wrote.

Then build a meta page that displays the corpus through that lens. It doesn’t have to be pretty. The point is that the classification becomes part of the artifact, not commentary about the artifact.

The first useful thing this gave me was not the categories themselves — those were always there. It was the visible unclassified count, growing every time I declined to look at a post and decide. The blank space is its own diagnostic.

What’s next

I’m building this as an extensible structure, not a one-off. The /lenses/ page is designed to accept additional frames — co-authoring posture, genre per Carolyn Miller’s Genre as Social Action, others I haven’t picked yet. Each new lens is one optional frontmatter field and one section on the page. No refactor, no migration.

The pattern is the point. Most blogs surface their content along one axis at a time — chronology, tag, title — and call that the index. But a single corpus can be viewed through many frames, and the frame is part of the meaning. Frontmatter is the lightest possible vehicle for that work.

I’ll write up the second lens when I’m sure it earns its keep. In the meantime, the existing one is enough to surface what I needed to see: I write more knowledge-telling than I admit, and the discomfort of admitting that is itself the start of writing the other kind.