How this site thinks

One test decides what gets written

A claim belongs on this site only if it would stop being true, or stop being sayable, for a market that published exactly one address. That test is applied to every study before it is written, and the study has to end by stating its own answer to it.

The test is unusually strict and it throws out most of what is normally written about WeTheNorth mirrors. Guidance on the Tor Browser, on payment, on choosing a vendor, on account security: all real, none of it about plurality, none of it here.

What this site never does

Opens an address

Never. No page here contacts any onion service. That is a fixed rule, not a limitation of the moment.

Publishes a figure it did not compute

Never. Where a number appears, it was measured from text on the page. There are no counts of anything this site cannot see.

Reports availability

Never. No uptime percentage, no status light, no time of checking, on any page, for any address.

Ranks the three

Never. No labels, no numbering, no principal, no fallback. The strings appear in the arrangement they were supplied in.

Names a date

Never in body copy. Periods are described in relative terms because a calendar date would be a claim about when something was true.

Quotes anybody

Never. There are no testimonials, no usernames, no reported conversations and no invented sources anywhere here.

Where the numbers on this site come from

Two kinds of number appear. The first is a count of things on this site, such as how many studies a section holds, and it is counted at the moment the page is built rather than typed into a sentence.

The second is a measurement taken from the three strings themselves: how long the shared head is, where each string first differs from the others, how many characters each one runs to. Those are computed by comparing the strings character by character when the page is built. If the set changed, the figures would change with it. That study is the head all three addresses share.

Why there is no status of any kind

A status light would be a claim about something this site has no information about. Filling the widget by inventing the information is the one thing a reference like this must not do.

There is a longer argument, which is that a reachability figure answers a question almost nobody actually has while appearing to answer the question they do have. It is set out in why there is no uptime number.

What this site is, plainly

An independent reference, written by somebody who is not the market and has no connection to it. It cannot reach an account, an order, a message or a dispute, and no page here should be read as speaking for anyone.

The three strings printed at the top of every page were supplied to this site and are published unchanged. Where they came from is the subject of nobody on this side decides what the set contains, which is also an account of this site's own position.

How to read a study quickly

1. Read the question. If it is not your question, leave.
2. Read the block headed if there were only one address. It tells you what the study is really about in three or four lines.
3. Read the line on what the study is not, which is usually where the boundary of the claim sits.
4. Read the prose only if the first three made you want to.