Privacy, written as a list of what is recorded rather than a list of reassurances

1

What one request for a page writes down, and why

Four fields go into a server log every time a page here is fetched, and they go in because of how the web works rather than because anybody decided to gather them. The address asked for. The instant of the asking. Where on the network it arrived from. And the browser string riding along with it, which describes a device and a version without naming a person. A referring address joins the four where the previous page elected to send one.

Serving the page is the first purpose and keeping it working is the second. A log is how a broken address, a missing drawing or a flood of requests from one place becomes visible. There is no third purpose, because this publication has no account, no sign-in, no newsletter, no comment field and no basket, so there is nothing those fields could be joined to.

Where a lawful basis has to be recorded for keeping a log of that kind, the category it normally falls into is the publisher’s own interest in serving and securing the pages. Where anything is set on or read from a reader’s device beyond what the page needs to render, the category that applies is consent. This page names the categories and does not assert a specific basis against each purpose, because the register of those decisions belongs with the publisher and is not something to improvise in prose.

2

Measurement, and why no supplier is named in this box

Counting readers is the one further thing a publication like this ordinarily does, and a measurement tool is how it is usually done. Such a tool works by setting or reading an identifier in the browser, or by processing the request fields already described, and reporting totals: how many people arrived, which addresses they arrived at, what sent them and whether they came on a handset or something larger.

No supplier is named on this page. Naming a company that is not actually in use would be a worse fault than naming none, and it is the precise fault the pages this site replaces were full of. Where a measurement tool is in use, the publisher prints its name and its category here, along with whether it sets anything on a device, and does so before this page is next reviewed.

What a reader should expect in the meantime is this. A publication that earns money by sending visits onward has an obvious commercial reason to count arrivals and departures, so assume counting of some kind is happening. Assume the counts are aggregate rather than about any identified person. And assume that anything setting an identifier in a browser requires asking first, which is a rule about consent and not a courtesy.

3

The moment a reader leaves by a link

A link out of here carries a parameter identifying this publisher, and that parameter is the mechanism by which the payment described at how this site is paid is attributed. It marks who sent the visit. It does not carry a name, an address or anything else about the reader, because this publication holds none of those to put in it.

What comes back is a count rather than a list. A destination reports how many arrivals from this publisher went on to open an account, which is what the payment is calculated against, and the reporting is a tally rather than a roll of people. Beyond the moment of the click this publisher sees nothing: not what was deposited, not what was staked, not how long anybody stayed.

The destination’s own notice governs from the instant of arrival, and it is a different document written by a different party under a different regime. A reader who follows a link has left this publication’s reach in the same instant, and the sensible habit is to read the notice at the other end before typing anything into it.

4

What can be switched off from the reader’s own side

Four controls sit with the reader, and not one of them needs permission from anybody. Storage a browser keeps for a site can be emptied and can be refused outright. A browser can be told to send no referring address, which deletes one of the four fields in part one. Tracking preferences and content blockers stop a measurement tool loading at all. And every word published here can be read anonymously, because there is nothing to identify yourself to.

Writing in is the single case where a reader hands over something deliberately. A message sent by the route described at write to us holds whatever its sender decided to put into it, and the sensible habit is to include the page address, the sentence objected to and the document behind it, while leaving out everything about an account anywhere. An account number, a username or a photographed document is of no use at this desk and cannot be acted on.

Nothing published here is directed at anybody under eighteen, and no part of the site invites a child to supply anything, because it invites nobody to supply anything at all.

5

How long anything is held, stated as far as it is known

Retention is a setting rather than a fact about this publication, and the honest position is that the figures are not on file for this page to print. What follows names the category in each case and leaves the number to the publisher.

Corrections are the one record kept on purpose and the reason is accountability rather than filing. Where a figure on this site is changed because somebody wrote in, the revision number on the page moves and the reviewed date moves with it, and those two marks at the head and foot of every page are the only part of this publisher’s record-keeping a reader can see from outside.

6

A different order of collection at the far end

Three items get verified before a first press wherever a British licence applies: the person’s name, the address they live at, and their date of birth. All three are taken and checked before any gambling may happen, and that floor stretches over a bonus and over the no-stake versions of the games too. Leaving the document request until a withdrawal is asked for, when it could reasonably have come earlier, is the practice the condition exists to stop.

The destination at the far end of such a link is, in the shapes this publication prices, a business with no British licence, and what goes missing on a page about data is the order in which documents change hands. Out there, under the rulebook consulted while these pages were assembled, registering turns on a date of birth being confirmed and a box being ticked to say the person is over eighteen, with the identity check proper fastened to later milestones in that business’s own anti-money-laundering process. A check arriving later is not a lighter check. It means photographs of a passport or a driving licence change hands after money has already moved rather than before the first press.

Whose hands they change into is the part that matters. The British regulator has put its own view of unlicensed sites in writing, and four hazards come out of it: controls on the money-laundering side and on the social-responsibility side that may not be adequate, customers left open to unfair practice, exposure to fraud, and trouble with the privacy of data, with identity theft named among the exposures in a separate post of its own. Four fields in a log at this end, photographed documents at the other: those are the two collections a reader is choosing between, and only the first of them belongs to this publisher to describe.