Privacy

Nothing to sign up for

This journal has no reader accounts, no comments, no newsletter and no forms.

No cookies or trackers

Reading this journal sets no cookies in your browser.

Its pages load no advertising, analytics or other outside scripts, and no fonts or images from other sites.

Reading plans

Some essays offer reading plans with checklists. If you choose to keep your ticks, your own browser stores them on this device. They are never sent to the journal, and the plan itself lets you remove them.

Photographs

The journal keeps the photographs its editors add to essays. Before you see one, the journal removes hidden data from the file, such as the place it was taken.

Visits from answer services

When you arrive by following a link in an answer service, such as ChatGPT, Perplexity, Gemini, Copilot or Claude, the journal adds one to a daily count for that service and that page.

The count keeps only the date, the name of the service and the page address. It keeps no IP address, no browser details, no cookie and no full referring address. Counts older than 400 days are deleted.

Where readers come from

When you open a page of the journal, it adds one to a daily count for that page and for where your visit came from: a known search engine such as Google or Bing, a known social site such as Facebook or Reddit, an answer service, another website, or no link at all. Following a link from one page of the journal to another adds nothing. Going back to a page, or reloading it, can add one again: your browser may load the page afresh and say again where you first came from.

The journal reads the address of the page you came from only to tell which of these it was, and your browser details only to leave out programs that read pages automatically. A count keeps only the date, the page and that source. It keeps no IP address, no browser details, no cookie and no referring address. Counts older than 400 days are deleted.

Requests from search and answer-service crawlers

Search engines and answer services send programs, called crawlers, to read the journal. When a request names a known crawler, such as Googlebot, Bingbot, GPTBot or ClaudeBot, the journal adds one to a daily count for that crawler, the kind of page asked for, and whether the journal served it, sent it elsewhere or answered with an error.

The journal reads the browser details of each request only to recognise a crawler by name; a request that names none adds nothing to the counts. A count keeps only the date, the crawler's name, the kind of page and the kind of answer. It keeps no IP address, no browser details and no page address. Counts older than 90 days are deleted.

Addresses that moved or do not exist

When an essay moves to a new address, its old address sends you on to the new one. The journal's editing software counts how often each old address is used and notes when it was last used.

When you ask for an address that does not exist here, the editing software notes that address, when it was first and last asked for, and how many times, so the editors can mend broken links. It keeps up to 10,000 such addresses and deletes those not asked for the longest first.

Neither record keeps your network address, your browser details or the page you came from.

Hosting and records

The journal runs on Cloudflare's network, which delivers each page to you. The journal's own code keeps no record of who reads it.

Writing to the editors

If you write to the editors, your message reaches their mailbox outside this site. The site stores nothing you send by email.

Reviewers

For each of its reviewers, the journal keeps their name, their qualification, the body that issued it, an address where it can be checked, the review areas they cover, the day they agreed to be named and the editor account they sign in with. It also keeps who added the reviewer and when, who checked the qualification and when, and who deactivated the reviewer and when.

Readers see only a reviewer's name, qualification and review areas: below the essays they reviewed and on the reviewers page.

For each review, the journal keeps the version of the essay, its address and the subject and review areas it covers, the decision to approve or reject, the confirmation the reviewer gave with an approval, the reviewer's statement, their name and qualification at the time, the account that made the decision and the time. For each essay, it keeps which editor accounts saved its text and when each first did, so that no reviewer approves an essay they edited.

The journal never changes or deletes these records, and keeps them for at least seven years. Each night it copies every review to its private archive repository, which never removes one, and a nightly encrypted backup of the whole site, kept for 90 days, holds all of these records too.

Writing to the grievance officer

If you write to the grievance officer, your message reaches their mailbox outside this site. The site stores nothing you send by email.

Editors and reviewers

Editors and reviewers sign in to a private editing area. Only they receive a sign-in cookie.