Privacy, corrections and removal
last changed 2026-10-08
Who runs this
Academic Ancestry is a personal, non-commercial project. Everything about this notice, and every request under it, goes to academicancestry@proton.me.
What the site shows about people
For a person: their name, who supervised their thesis and whose theses they supervised, and who examined, refereed or sat on the committee or jury of a thesis. For a thesis: its title, year, institution and level, doctoral and, where a repository's records carry them, master's and bachelor's, with the department, the language and the file where its record gives them. A birth year is shown only where Wikidata records one. Where Wikidata holds a person, their card also shows the portrait Wikidata names, served by Wikimedia Commons and linked to the page that credits it, and says what Wikidata states of them: a short description, the year of death, where they studied and worked, their fields, awards and memberships, and the identifiers other registers know them by. A list of a person's papers comes from OpenAlex, and only under an ORCID or an OpenAlex identifier that Wikidata or their own thesis record gives, never by matching a name; a paper's card names its authors as OpenAlex does. Coauthors, when asked for, come from OpenAlex.
The site holds nothing else about the people it names: no contact details, no addresses, and no pictures of its own; a portrait is Wikidata's and leaves with it.
Where it comes from
Only from sources already public: Wikidata's statements about a person, their doctoral advisors first, the thesis records universities and national libraries publish in their open repositories, and OpenAlex. Each entry names the record it was read from, and the sources lists every repository read and every one known and not read. Nobody is asked for information about themselves or about anyone else; a reader who reports a missing advisor tells the project where to read it, and the report itself is never shown.
Because the entries are read from public records and not from the people they name, this notice is how the project tells them. Writing to each person is not possible: the records carry no contact details.
Why, and on what basis
The purpose is to make the public record of who taught whom easier to see and to follow: the history of scholarship as the theses themselves record it. Under the GDPR the basis is legitimate interest (Article 6(1)(f)). The records are already public, they concern a person's academic work and not their private life, and anyone named can have their entry removed, as below, without giving a reason.
Having an entry corrected or removed
If the site names you, you can ask what it holds about you, have a mistake corrected, object to being shown, or have your entry removed. Write to academicancestry@proton.me with your name and a link to the sheet that shows you; every person's card on the sheet has a link that starts that message with the entry named. You may be asked for enough to show that the entry is yours.
A removed person is taken off every page, list, search result and answer of this site, and stays off when the records are read again. Requests are answered within a month, usually much sooner. The site keeps one thing about a removed entry: its identifier, on a list that is never published, so that it stays removed.
Removal here does not change the source. The thesis record stays with the repository that published it and a Wikidata statement stays on Wikidata; to change those, ask the repository or edit Wikidata.
You also have the right to complain to the data protection authority where you live.
Search engines
Search engines are asked not to read the pages of people: the site's robots.txt closes those pages and the routes they are drawn from, and each page says it is not to be indexed, so a person's sheet is not offered to a search engine. When someone posts a link to a person's page, the service it is posted on (X, Facebook, LinkedIn, Discord, Telegram, WhatsApp) may read that one page to draw the link's preview; the page tells it too that it is not to be indexed.
What the site records about its readers
No account, no cookies, no advertising, and no outside service that counts or follows readers. Your browser keeps, on your own device, the sheet's settings (the theme, the form and bounds, the tips already shown) and the last five people you opened; none of it is sent anywhere.
The server keeps no access log of readers' addresses; an address reaches its logs only when a request fails. To stop one address from flooding it, it counts requests for one minute under a code scrambled from the address with the key described below, deletes the count when the minute is over and keeps no address; it logs which page was asked for and how long it took, without the address.
The site counts its own use, on its own server, to see how many people read it and what they look for. For each answer it sends it keeps the time, what was asked for (a person's sheet, a name searched, a card opened), whether the answer came, the site the reader came from when the browser names one, and whether a phone, a computer or a program asked. To count a day's readers it also keeps a code made by scrambling the reader's address and the browser's name with a random key. The key is kept only on the server, in its own database, and is deleted when the day turns at midnight UTC: after that the code cannot be turned back into an address, and the same reader has a different code the next day, so nothing follows a reader from one day to the next. The address itself is not kept, and neither is a question typed into the ask dock. The record of each answer stays on the server for a year and is then deleted. Only totals leave it (how many readers, pages, sheets and searches in a day), sent to the monitoring of the Amazon Web Services account the server runs in, where the project reads them as charts.
The tools the site offers to assistants are the one place an account exists: a reader's assistant signs in, with ORCID, Google or GitHub, to ask on the reader's behalf what the pages answer. Reading the site needs none. An account keeps which provider it was signed in with, the identifier that provider gives, the email address where the provider shares one, when the account was made, and its plan; it holds no password, which the provider keeps. Each call made under it keeps the time, the tool, what was asked and whether the answer came, tied to the account for a year at most, so that its daily allowance can be counted; the address is not kept, as for every other answer. The provider learns that you signed in here. The terms for accounts say what an account may do and how much it may ask. To have an account and its calls deleted, write to academicancestry@proton.me.
A question typed into the ask dock is sent, with the sheet on screen, to Anthropic's API to be answered; the site does not keep it.
A report that an advisor is missing keeps what was typed into it (the student's and the advisor's names, where it is written, and the person whose card it was sent from) with the time it came, in a file on the server that is never published. It is kept so that the project can find that record and read it in; it keeps no address, and it is never drawn on a sheet or counted as a link. The names are checked against Wikidata and the repositories' records when it is sent, and the answer is shown at once. To have a report deleted, write to academicancestry@proton.me.
The typefaces are served by the site itself. The server is hosted with Amazon Web Services in the United States. Its pages and answers reach a reader through Amazon CloudFront, a network of servers nearer to readers that keeps a copy of each page and each answer for an hour and serves the next reader from it; a name searched and a question to the ask dock are never kept there, it passes the reader's address on to the server, which counts as described above, and it keeps no log of its own.
Changes
When this notice changes, the date at its top changes with it.