pattern-of-life-from-socials

Extract every post timestamp into a table normalized to UTC, then plot hour-of-day and day-of-week histograms. A contiguous seven-to-nine-hour gap indicates the sleep window, whose position gives a UTC offset. Verify posting is manual first, since scheduling tools post at fixed times regardless of location.

Some platforms show creation dates directly; others encode them in snowflake-style 64-bit IDs whose high bits are a millisecond timestamp from a platform epoch. Plain sequential IDs give registration order, which you can bracket against accounts of known age.

Yes. Early followers are typically family and close friends, mutual-follow clusters map real-world communities like employers or schools, and inbound tags from less careful relatives often leak details the subject withheld. Build the result as a graph, not a list.

It depends on the platform and surface. Ephemeral story views are almost always reported to the poster, professional networks often report profile visits, and ordinary feed posts generally do not. Liking, following, or commenting is always attributable, so use a dedicated research account.

Automated scraping breaches most platforms' terms of service, while manual viewing of public content generally does not. Under GDPR, publicly available data still needs a lawful basis, and sustained monitoring of an individual's location and routine can meet the statutory definition of stalking.

Scheduling tools post at fixed wall-clock times regardless of the human's location, platforms may render timestamps in the viewer's locale, and shared or team accounts blend multiple people into one histogram. Check replies, which are hard to schedule, against top-level posts before trusting the distribution.