NotesNudge the notes you never go back to

What Not to Save

Every archive has a sediment of material that should never have entered. It is not neutral — it dilutes search results, it dilutes anything surfaced at random, and it makes the collection something you avoid.

The filter has to sit at capture, because nothing downstream removes what should not have arrived. Review deletes three or four notes a session; an unfiltered pipeline adds forty a week.

For a workplace-oriented comparison with this kind of workflow, further details provides additional context.

Six categories

Anything you can find again in ten seconds. A well-known fact, a public definition, a link to a popular article. The internet is your reference layer. Saving what is trivially findable adds noise and buys nothing.

Imported highlights. Marking is not processing, and automatic pipelines from a reader into an archive are the single largest source of unusable material. If a passage matters, restate it.

Whole articles "to read later." These are a queue, not notes, and merging the two ruins both. Keep the queue somewhere that empties.

For an independent external reference related to this topic, see Todoist.

Things saved because they seemed important. The tell: you cannot say what you would use it for. This is the one that produces the most volume, because it feels responsible and costs nothing at the moment.

Aspirational material. The book list you will not read, the technique you will not adopt, the framework that impressed you once. An archive is not a plan.

And anything about a person that you would not want them to read. Not a filing consideration — a judgement one. Archives outlive their purpose, get synced, get exported, get inherited. This is the only item here that is not about usefulness.

The test at capture

One question, three seconds: what would make me want this back?

If you can name the occasion — a decision you expect to face, a piece of work in progress, a recurring question — save it, and write that occasion into the note, because it is what search will match later.

If you cannot name one, you are saving because it seemed important. That is the sediment.

And notice that this filter is nearly free. It is the same three seconds whether the answer is yes or no, which is why it survives where more elaborate capture rules do not.

The objection, taken seriously

But I cannot know in advance what I will need.

True, and it is the strongest argument for saving generously. Two responses.

Unpredictability cuts both ways. If you cannot predict what will matter, then indiscriminate saving does not raise your odds much — it raises the volume, and volume is what makes the useful item hard to find when the moment comes.

And the cheap version exists. Save it with a title and no processing, in a place you know is unprocessed. A holding area is not an archive, and keeping them separate means the sediment does not contaminate the part you rely on.

What does damage is treating everything captured as a note of equal standing.

What this leaves

An archive that is smaller than most people's and considerably more useful per item. Search results you can scan. A random selection that is usually worth reading. And no accumulated guilt about a backlog, because there is no backlog — there are notes, each of which had a reason.

That last part matters more than the efficiency. Most of the unhappiness in this subject comes from an archive that feels like a debt, and the debt is made of things nobody decided to save.

The short version