Temporal Bracketing
An instrument for the comparative temporal bracketing of longitudinal qualitative material: one frozen register of evidence, a committed family of alternative periodizations, the same reading protocol applied under each, and a computed account of which findings survive re-bracketing, which depend on it, and what the contamination controls show. Entry is by invitation code.
What happens to material you bring
- Where it lives. Documents you add are stored in this instrument's database on a server in Singapore, with a nightly backup kept for ninety days. There are no analytics scripts and no tracking on this site.
- Who reads it. Extraction, readings, and matching send the text to two model providers, Anthropic and Google, on paid tiers whose terms, at the time of writing, do not use submitted material to train models. Every call is logged in your study's record.
- Who sees it. A study is visible only to the code that created it and to the instrument's operator.
- What you get back. The audit bundle — every layer of your study's record, your rulings verbatim, and a Word report — is yours to download at any time.
- Deletion. On request, the operator removes a study from the live database; its backups expire within ninety days. Write to the contact address in the footer.
- Confidential material. If your corpus is confidential or carries participant consent conditions, check that processing by the two providers named above is permitted under those conditions before you upload it.