Questions are the to-dos of learning

What Things gets right

Things, the to-do app from Cultured Code, is admired less for what it does than for how little it asks of you. It has a small, fixed vocabulary — to-dos, projects, areas, headings — and you can’t invent new kinds of thing. Projects can’t be nested inside other projects. People list that as a limitation. It is the design.

Three ideas in it are worth stealing for learning.

An Inbox. Everything you want to do ends up in one place first, before you decide what it is or where it goes. Cultured Code calls it a temporary staging ground for unprocessed thoughts (Things guide). Capture is cheap and fast; sorting comes later.

A short list of what to do now. Today shows a handful of things you have chosen, and hides everything else so it doesn’t compete for attention (Things’ lists). The point isn’t the list. It is not having to decide, every time you sit down, what to do next.

Something you can tick off. A to-do is a unit of work with a clear end. You know when it is done, and ticking it is satisfying for a reason: it marks real progress.

All three come from David Allen’s Getting Things Done (2001), whose five steps — capture, clarify, organise, reflect, engage — describe how loose commitments become finished work. Things is that method made into a calm piece of software.

Why notes apps have no unit of work

Notes apps have plenty of Allen’s first step and almost none of the rest. You can capture endlessly: clip articles, jot ideas, highlight passages. What you can’t do is finish anything, because a note has no end. A note about attention is never done. It can only get longer.

That is fine for an archive. It is a problem for learning, because learning is work, and work without a unit has no shape. You can’t tell what to do next, you can’t tell whether you made progress today, and the notebook slowly fills with material you once meant to understand.

Learning does have a natural unit of work. It is the question.

A question has everything a to-do has. You can capture it in a second. It has a clear subject — it is about something. And it has a clear end: you can answer it, or you can’t. Why does attention’s cost grow with the square of the input? is either something you can explain now, or it isn’t.

Writing good questions

Not all questions are equally good to-dos. What is attention? is easy to write and easy to “answer” by pasting a definition, which teaches you nothing. The questions that do the most work ask why and how:

• Why does attention’s cost grow with the square of the input?
• How does a central bank raising rates slow down prices?
• Why don’t the teeth of a zip come apart when you pull the fabric?

Asking why and how something is true — elaborative interrogation — is one of the study techniques that held up best in a large review of the evidence, rated well ahead of rereading, highlighting and summarising (Dunlosky et al., 2013). A why-question forces you to connect a fact to the mechanism behind it, which is the difference between knowing a fact and understanding it.

A few habits make questions better:

• End it with a question mark. It sounds trivial. It keeps you from writing topics (attention costs) instead of questions.
• Make it about one thing. How do transformers work and why are they expensive? is two questions. Split it.
• Attach it to what it’s about. A question floating in a general list is easy to lose. A question written on the note for the concept it concerns turns up exactly when you return to that concept.
• Keep the questions you couldn’t answer when you tried to explain something. Those are the most valuable questions you will ever write, because they mark the precise edge of your understanding.

When a question is done

A to-do is done when the thing is done. A question is done when you can answer it in your own words, with the notes closed.

Not when you have found an answer. Not when you have saved a good explanation next to it. Finding and saving feel like progress, and are the learning equivalent of filing an email. The tick belongs to the moment you can produce the answer yourself, because that moment is the only reliable evidence that you understand it.

So when you tick a question, write your answer underneath — a sentence or two, in your words. It will be useful later, both as a record and as something to check yourself against.

Questions that come back

Ticking a question isn’t the end of its usefulness. A question you answered once is the perfect thing to try again later.

Retrieval — producing an answer from memory — strengthens memory far more than looking the answer up again. In a well-known study, students who read a passage and were then tested on it remembered much more a week later than students who reread it, even though the rereaders were more confident they would do well (Roediger & Karpicke, 2006). And spreading those attempts out over days and weeks, rather than repeating them in one sitting, helps further; the spacing effect is one of the oldest and most reliable findings in the psychology of memory.

You don’t need a system for this to work. Every week or so, look back at a few questions you ticked and answer them again without looking. If you can’t, untick it. That isn’t a setback; it is the notebook doing its job.

A short list of what’s next

The last thing to borrow from Things is the short list. Learning has more threads than time, and an open list of everything you could study is the notebook version of an overflowing inbox.

Keep a list of no more than three things to do next, one for each kind of work:

• Understand: the next concept to learn, usually one that something you already know depends on.
• Connect: two things you’ve learned that you haven’t yet said how they relate.
• Teach: an idea you think you understand, ready to be explained aloud.

Three lines is enough. When you sit down, you start at the top. When one is done, you choose the next. The rest of your questions stay where they belong, attached to the concepts they are about, until their turn comes.

How Graphite does this

Graphite is built around this. Everything you write down lands in the Inbox first — a line, a question, a link — and stays there until you move it to a project or a folder, or, once you turn on File my Inbox, Graphite files it there, writing each act in the Logbook, with Undo. A question is a plain - [ ] …? item inside the note it’s about, and a project’s page gathers every open one across its notes under What’s missing, to tick where it is; the gaps you find when you teach something back go there too. When you want to know what stands between your notes and something you could publish, What’s missing asks the questions a curious reader would ask that your notes don’t answer yet, each one sitting’s work, kept in the note only when you choose. You tick a question when you can answer it in your own words. You can try it in your browser.