Second Brain

← Blog

How a Notes App Can Answer Questions About Your Life

· 4 min read

A person on a balcony in morning light holding a phone in one hand and a coffee in the other

A notes app can answer questions about your life when it stores what your notes mean, not just what they say — and when it can read across many notes to assemble one answer. Ask “when did I last service the car?” and instead of a search-results list, you get: “March 14, at Özkan Oto — you noted they recommended new front tires by autumn.” That shift, from searching to asking, is the biggest change in personal notes since they went digital.

From searching your notes to asking them

Search and asking sound similar; they’re different jobs. Search hands you documents and leaves the reading to you. Asking hands you the answer — with the source notes attached so you can check it.

The difference compounds with archive size. At fifty notes, search is fine. At three thousand notes spanning four years, a keyword query returns a haystack, and you wanted the needle: one date, one number, one name. The whole reason you wrote things down was so a future question could be answered — the notes were always means, never ends.

How it works, without the jargon

Under the hood, two capabilities make a notes app answerable.

First, meaning-based search: every note is indexed by what it’s about, so “oil change” matches “car service” and “that fish place in Lisbon” matches a note that never contains the word “fish.” Your memory stores gist, not strings; this search style finally matches it. (It’s the honest meaning of the phrase “smart notes” — the intelligence in the finding, not the formatting.)

Second, reading across notes: an AI takes your question, pulls the handful of notes that might matter, reads them together, and writes the answer. The vet note from January and the one from May, combined: “Boncuk’s vaccine is due in July — the January note says the clinic moved to the new address.”

Questions a personal archive can actually answer

From real life, the kind of thing this is for:

  • “What time is Ela’s pediatrician appointment?” — buried in a voice note from the school run.
  • “What did the plumber quote in March, and was it valid until June?”
  • “Which wine did we like in Vienna?”
  • “What’s the hallway paint color called?”
  • “When did the headaches start?” — answerable because your journal entries live in the same memory.

Notice what these have in common: you wouldn’t remember which note holds the answer, when you wrote it, or what words you used. With askable notes, none of that matters — that’s the whole point for people with bad memories.

What you have to change about how you capture: nothing

The quietly radical part: this works on messy notes. No tagging discipline, no titles, no structure — the AI reads your “veli banyo 9k haziran sonuna kadar” shorthand the way a person would. Capture stays as fast and sloppy as life requires, and the intelligence applies at retrieval time, where it belongs. Every previous attempt to make notes useful demanded better input habits; that’s exactly why they failed. This one demands nothing.

Accuracy and limits — the honest part

Three truths to hold any answering app to:

  1. It can only answer from what you captured. No note about the parking spot, no answer about the parking spot. The system’s ceiling is your capture habit.
  2. It must show its sources. An answer about your own life should come with the notes it was built from, one tap away. Trust-but-verify isn’t optional when the subject is your memory.
  3. “I don’t know” must be an answer. The failure mode to fear isn’t a missing answer — it’s a confident invented one. An app that would rather guess than say “nothing in your notes about that” hasn’t earned questions that matter.

What to look for — and what I built

If you’re evaluating apps for this, the checklist is short: capture has to be instant (the answers are only as good as what gets in), voice and photos should count as notes, answers must cite sources and admit ignorance, and the privacy posture has to survive five minutes of policy reading.

I’ll end with the disclosure-slash-pitch: I built Second Brain because I wanted exactly this and couldn’t find it — and answering questions about your own life is the one job I’d claim it’s the best at. Everything goes into one chat — text, voice, photos — and asking your memory is just typing a question into the same thread, with the source notes attached to every answer. Judge that claim against the checklist above, on your own notes. That test is the only review that matters.