Random Resurfacing

A tool picks a note from your archive and shows it to you. No schedule based on difficulty, no attempt to recall, no selection by relevance — just something you wrote, arriving.

It is the lowest-effort intervention in this subject and the most commonly sold. Worth being precise about what it is good for, because the usual justification borrows research it does not fit.

A related workplace concept is cognitive offloading, which offers a useful contrast with the personal knowledge-work problem discussed here.

What it does not do

It does not strengthen memory. The evidence for spaced review is about effortful retrieval, and random display involves no attempt. What you are doing is closer to restudy, which is the condition retrieval practice consistently beats.

It does not follow the spacing research. Spacing means intervals calibrated so recall is effortful but successful. Random has no interval structure. A note may arrive twice in a month or once in three years, and neither is a schedule.

It does not tell you what you have forgotten. It tells you about one note.

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

And it does not scale with the archive. With two thousand notes and one a day, you see each about once every five and a half years. The larger your collection, the less any individual note is likely to return — which is precisely backwards from what an archive owner wants.

What it does do, and it is not nothing

Rediscovery. You cannot use an idea you have forgotten exists. Random resurfacing is a mechanism for reminding you that things exist, and no amount of search does that, because search requires knowing what to look for.

Recombination. An old note arriving beside today's problem sometimes connects. This is the strongest informal case for the practice, it is reported consistently by people who do it, and there is essentially no research on it because it is very hard to study.

Almost no cost. No habit, no discipline, no willpower. Given that most people do nothing at all with their notes, an intervention that requires nothing has a genuine advantage over a better one that requires something. A weak practice that happens beats a strong one that does not.

And it keeps the archive alive. Seeing your own old writing occasionally affects whether you keep writing, which is a real effect even if it is not a memory effect.

How to make it worth more

Three changes, all cheap, and each moves it toward something better supported.

Read the title first and try to recall the note. Fifteen seconds. This single change converts resurfacing into retrieval and it is free.

Give every surfaced note a decision. Keep, delete, or promote to a small set you actually review deliberately. Without a decision the practice is entertainment; with one it slowly improves the archive.

Weight the pool. Fully random is worse than random-within-a-subset. Restrict it to notes you marked at capture time as worth returning to, and the hit rate rises sharply. Most tools will not do this and a folder plus a shuffled list will.

Where it fits

Good: as a background practice, on a curated pool, alongside something deliberate. Cheap, occasionally excellent, and it costs you thirty seconds a day.

Bad: as your entire relationship with the archive. That is how people end up with a daily email they stop opening in week three, and a lingering sense that the system failed them when the system was never doing what they were told it was doing.

And it is not a substitute for deletion. An archive that only grows becomes less useful at a rate proportional to its size, and random display does nothing about that. If anything it disguises the problem, because a stream of interesting old notes makes a bloated archive feel productive.

The honest recommendation

Use it. Expect rediscovery rather than retention. Add the fifteen-second attempt, add a decision, and narrow the pool. The ceiling on what it can do is worth knowing precisely.

And do not believe anyone who tells you it works because of the spacing effect. It might work; that is not why.

The short version

• Random resurfacing is display, not retrieval, so the memory research does not apply to it
• It has no interval structure, so citations of the spacing effect are borrowed
• It scales badly: two thousand notes at one a day means each returns about every five years
• What it genuinely does is rediscovery and recombination — real benefits with almost no research behind them
• Its decisive advantage is costing nothing, and a practice that happens beats one that does not
• Improve it with fifteen seconds of attempted recall, a decision per note, and a narrowed pool