Start with the question
Saved material is not automatically an obligation to read or document. Begin by stating the question the source should answer and the decision, design, or capability that might change if the answer is useful.
If there is no concrete question, the source may still be worth keeping as searchable reference, but it should not quietly become scheduled work.
Triage the source
Use four outcomes:
- Deep read — read the source firsthand because it can answer a concrete question or improve a current decision.
- Summary sufficient — capture the useful claim and decide whether it changes anything; a full read is unlikely to repay the time.
- Reference — keep the source searchable for a future need without scheduling it as active work.
- Review or release — remove duplicates, stale material, inaccessible sources, or material below the current learning level.
The classification is a time-management decision, not a judgment about the source’s absolute quality.
Make a deep read produce a durable result
For a deep read:
- State the question before opening more sources.
- Read for trade-offs, failure modes, and boundary conditions.
- Connect one idea to a real problem, decision, or practice context.
- Create or update a durable note only when that connection is concrete.
- Close or release the source after the result—or an explicit no-value decision—exists.
This keeps the public note, design choice, experiment, or changed practice as the result of reading instead of treating the collected source itself as progress.