Privacy
Nothing here requires an account, and this site runs no analytics and sets no cookie of its own unless you sign in with Patreon. That is not the whole of it, because the pictures and 3D scans are loaded from the museums and archives that hold them, and some of those set cookies of their own. The second half of this page is about that half of the page.
What this site does not do
No analytics, no advertising, no profileThere is no analytics script, no advertising, no advertising identifier and no profile of you. Nothing about you is sold, shared or handed to anyone, because nothing about you is gathered to hand over. There is no account system: the one optional sign-in is with Patreon, described below, and nothing on this site is behind it.
What is stored, and where
Your settings and anything you save
Stored in your own browser, on your own device, and never sent anywhere. That covers the theme, the display options, and any search or taxon you save.
Which means clearing site data clears it, and it does not follow you to another device. That is the trade for having no accounts. The Saved panel offers a download so you can keep a copy.
If you sign in with Patreon
Stored in one cookie in your browser, holding your Patreon user id, the display name Patreon gives, and the names of any tiers you support this site at. It is signed so it cannot be edited, and it expires after seven days. Two further cookies exist only for the few minutes of the sign-in itself and are deleted the moment you come back.
Which means signing in tells Patreon that you visited, because you go to their site to approve it. No email address is requested. No access token is kept, so nothing here can act on your Patreon account afterwards. Nothing on this site is behind the sign-in: it exists so that something can be later, and today it puts a name in the corner and does nothing else. Signing out deletes the cookie.
If you sign up to test the Android app
Stored in one row of one table, holding the email address you typed and the date. The same address is emailed to the inbox at the bottom of this page, because adding somebody to a closed test is done by hand in Google Play and a person has to read it. Nothing else about you is recorded: not which page you were on, not your address on the network.
Which means it is used to send you the app and for nothing else. It is not shared, not sold, and not added to any mailing list, because there is no mailing list. It reaches Google, because the inbox that receives it is a Gmail one and because the invitation itself has to be sent to your Google account. The list exists for the closed test and goes when the test does. Ask at the address below and yours comes off it sooner.
If you write to us from the contact form
Stored in nothing. The form in the footer sends your address, your subject and your message to the inbox at the bottom of this page, and keeps no copy. There is no ticket system behind it and no row in any table.
Which means your address is used to answer you and for nothing else, and it is not added to any list, because there is no list. It is an ordinary email once it arrives, so it sits in that inbox until it is deleted, the same as if you had written directly — which you are welcome to do instead, and the form says so.
Answers to searches, on the server
Stored as the answer to a question, keyed by the question. A period and a place, and the fossils recorded from them. There are two such tables and neither has a column for a person.
Which means it holds nothing about who asked. It exists so the next person asking the same thing does not cost the public databases a second request.
Your network address, briefly
Held in memory for a few minutes by the rate limiter, to count requests and stop one caller overwhelming the databases behind this site.
Which means this site does not write it to disk, does not log it, and does not tie it to anything you looked at. Separately from that, the company that hosts the site keeps ordinary web server logs of requests, in the way every host does. Those are theirs, are kept for a short window, and are not read for anything but faults.
What the server prints while it works
Written to the server log: which database was asked about which taxon, whether it answered, and why a picture was rejected. These lines are how a wrong picture or a silent failure gets found.
Which means they name fossils, not readers. The line for a cached search used to print the search itself; it now prints only how many taxa came back, so a period, a place and a group asked for are no longer written down there. There is one exception, and it is the next entry.
What broke, on the rare occasion something does
Sent to a private channel the people who build this read: the error, the stack trace that came with it, the address you were on including whatever you had searched for, and the line your browser uses to describe itself.
Which means it happens only when the page has actually failed, never on an ordinary visit, and it is the one place a search you did gets written down. Nothing in it identifies you: no network address, no account, nothing that would join two reports together or say they came from the same person. The Android app is a shell around this site, so a fault here is invisible to the crash reporting Google does for ordinary apps — without this, a page that breaks for you reaches nobody unless you stop and describe it yourself.
Other people’s services
Where the pictures and scans come fromPictures and 3D scans are not copied onto this server. Your browser fetches them from the institution that holds them, which means those services see the request the way they would see you visiting their own site, and some of them set a cookie while answering. This site cannot prevent that. What it can do is not start the request until it is needed, and say plainly which services are involved.
Wikimedia Commons, for most photographs
Loaded from upload.wikimedia.org, whenever a card that has a photograph is on screen.
Which means the response carries a cookie of theirs, WMF-Uniq, a random identifier that lasts about a year. It is set by Wikimedia, not by this site, and it falls under the Wikimedia Foundation’s own privacy policy. Article text is fetched by this server rather than by your browser, so reading a card’s description involves no request from you.
Smithsonian and Sketchfab, for 3D scans
Loaded from 3d-api.si.edu or sketchfab.com, inside a frame, and only after you press Rotate in 3D.
Which means a viewer is their page, running their code. The Smithsonian’s loads Google Tag Manager and Google Analytics. Sketchfab sets a cookie and records that the model was viewed, and its embed is asked not to track you, which is a request rather than a guarantee. Neither starts on its own: opening a card shows a still thumbnail and a button, and the frame is created when you press it. If you would rather they load with the card, Options, then Behaviour, then Load 3D scans automatically turns that on. It is off until you do.
MorphoSource, GBIF and iDigBio
Loaded from their own servers, as thumbnails and specimen photographs. MorphoSource’s viewer cannot be framed, so those are always a picture and a link out.
Which means the same as above: they see a request for an image, and whatever they set with it is theirs. Following a link out, to any of them or to Discord or Patreon, takes you under that service’s own terms.
Typefaces are downloaded when the site is built and served from this domain, so no request goes to Google Fonts when a page loads. The databases behind the fossil data are named on the Sources page.
The Android app
The app is a Trusted Web Activity: a thin shell that opens this same site full screen using the browser already on your phone. It adds no data collection of its own, and everything above applies to it unchanged, including the parts about other people’s services.
It requests no device permissions. It cannot see your location, contacts, camera, microphone or files, because it never asks for any of them. It does ask Android which apps can open an https link, which is how it finds a browser to run in, and that is the only thing it can see about the rest of your phone. Its own data is excluded from Android’s automatic cloud backup, so saved searches and settings stay on the device. The Patreon sign-in above is reachable inside it, and behaves there exactly as it does in a browser.
Corrections
Until September 2026 this page said that no cookie was set and that there was no tracking of any kind. That was true of this server and untrue of the page you were given, because the 3D viewers and the borrowed images bring their own. It is corrected above rather than quietly rewritten, and the viewers were changed to wait for a click at the same time. If you find something else here that does not match what the site does, that is a bug and worth telling me about.
In September 2026 the site began asking for Android testers and grew a contact form, which together are the first time it has ever handled an email address. What that form does with the address is described above, in the same list as everything else, rather than in a separate notice that nobody reads.
Questions, or a request to have something explained: EpochExplorer.Earth@gmail.com