Pair Search
What we collect
Three kinds of thing, and nothing beyond them. Account details: the email address and display name you give us at registration, plus a hashed password. Declared activity: things you knowingly do on the Journal, meaning reviews and ratings you submit, Bias Meter runs, Theel coins earned and spent, drafts you write, and the pairs you share. Technical basics: device type, browser, approximate region derived from your IP, and request logs, kept so the site works and so we can tell a reader from a bot.
Writer profile visits: when you open a writer profile while signed in, we record your account id so the writer can receive one daily visit summary. Signed-out profile visits are counted but never identified. We record no other page visit for this feature.
Your reading diet
The Journal can draw a map of your reading: where the articles you open sit on the two lenses, and the gaps between them. Read plainly, that is a picture of political exposure, so we treat it as sensitive by design: the map and article-by-article history are computed and stored on your device. If you select Make public, we receive and display only a summary: counts across the lenses, reading regions and article archetypes. We receive no list of the articles you opened. Switching Make public off removes that summary from your public profile and our stored profile data. Slide to your next read scores candidates against it locally. It measures what you read, not what you believe, and it is never reduced to a label: a distribution and its gaps, nothing more. Clearing your browser data erases your local history. If you previously made its summary public, switch Make public off to remove that separately published summary.
Why we collect it
To keep your account and Theel balance intact across devices; to show your reviews back to you; to credit contributors and pay out where payment is due; to show writers aggregate profile interest and seat conversion; and to publish aggregate, de-identified statistics about how opinion on the Journal is distributed. The bias reading is computed on the text of articles, never on the person reading them.
Cookies and local storage
Essential cookies keep you signed in and remember which lens you last used. Local storage holds in-progress reviews, slider positions, reading position, and the reading-diet map itself, which lives here rather than on our servers; clearing your browser data erases the map for good. Analytics is one privacy-respecting tool with truncated IPs and no cross-site identifiers. There are no advertising cookies on the Journal.
Who else sees it
Only the service providers we need to operate: hosting, email delivery, payment processing for donations, and error monitoring. Each is bound by contract to use the data solely on our instructions. We disclose data to authorities only against a valid legal order, and we will tell you unless the order forbids it.
Your rights
Access & export
Correct
Delete
Object
How long we keep it
Account data for as long as the account is open. Declared activity (reviews, meter runs) for two years, then aggregated and stripped of identifiers. Writer profile visit rows and technical logs for 90 days, never joined to any diet map. Donation records for as long as tax law requires.
Contact
Write to [email protected] with any request under this policy. We answer within 30 days, usually much sooner.