Privacy
There are no accounts, so there is almost nothing to say. Here is all of it.
What we do not have
- No accounts. No email address, no name, no password, no payment details. There is no signup, so there is no profile to attach anything to.
- No advertising. No ad networks, no retargeting pixels, no data brokers, no third-party trackers. Nothing here is sold or shared with anyone.
- No tracking of API calls. The pages you read and the API you call are
separate.
/v1/*requests carry no analytics of any kind.
What we do have
Server logs
Standard web server logs, needed to run the service: IP address, timestamp, requested path, response code, size, user agent. Used for capacity planning, abuse detection and debugging. Not sold, not shared. Rotated and deleted on a short cycle.
If you voluntarily send agent identification — see identify your agent — those values land in the same logs. That is the whole mechanism, and it is why the page tells you not to put personal data in them.
A site analytics counter
These documentation pages carry our own analytics counter, served from
statinside.com. It is our software on our infrastructure, not a third-party
analytics product, and the data does not leave us.
On a page view it records: the page, the referring URL, screen size and colour depth, browser language, connection type, and coarse interaction such as where on the page you clicked. It may set a cookie to recognise a returning visit, and it can capture page structure to help us find layout bugs.
To opt out, block statinside.com — any content blocker does it, and
the site works identically without the counter. There is nothing on these pages that
depends on it.
The API itself is untouched by this. The counter runs only on the HTML
documentation pages, which is where humans are. An agent calling
/v1/search executes no JavaScript and is measured by nothing beyond the
server log line.
Your search queries appear in those logs as part of the URL, as they do on any HTTP service. We do not analyse them per caller and we do not associate them with anything, because there is nothing to associate them with.
Traffic in front of us
Requests pass through a CDN for TLS termination and denial-of-service protection. It sees the same request data and applies its own logging and privacy terms.
Article content
The corpus is news articles from public sources. It may contain personal data about people mentioned in the news, exactly as their publishers published it. We do not augment, enrich or cross-reference it. Articles are deleted after 30 days automatically.
If you are named in an article and want it removed, contact the publisher — they control the source. If it is also indexed here after they remove it, tell us and we will delete our copy.
