The record: people, theses and works
The record is a graph. Its things are people, theses and other works; its links are roles a record gives a person on a thesis or a work. Supervision makes the lineage; examining, sitting on a committee or opposing a defence sit beside it; coauthorship, citation and teaching are drawn and never counted as lineage.
People, theses and works
The record holds three kinds of thing. A thesis is one record of a thesis, as a repository or a register describes it: a title, an author, a year, a school, a degree, a page to read it from, and the people it names. A work is any other record a repository holds, an article, a chapter, a report, kept with its authors as the repository lists them. A person is everyone those records name, in whatever role.
Each has a page of its own. A person's is at /person/ followed by the id the search lists them under: a Wikidata item, an OpenAlex author, or, for someone only a repository names, an id made from their name as that repository spells it. It shows their advisors with the theses that make each link, their students, what Wikidata states of them, apart, their own theses and works, who they sat beside, and the registers they are linked to. A thesis's page is at /thesis/, then the repository it comes from, then the identifier the repository gave the record; it shows everyone the record names, each in their role and with the kind of evidence behind them, and says where the record stops when it names no supervisor. A work's page, under /work/ in the same shape, shows its authors in the order its record lists them.
The links between them are roles. A record that names a person as a thesis's supervisor makes a link from that person to that thesis, and through it to the thesis's author: that link is a supervision. The schema above draws the whole: who can stand in what relation to what, and which relations make the family.
The roles a record names
Repositories describe their theses in many dialects, and each names its people in its own words. The record turns those words into a small set of roles, the same in every country, and keeps every role a record gives, not only the supervisor's.
- Author
- The person who wrote the thesis: for a doctoral thesis, the student.
- Supervisor and co-supervisor
- Who directed the work, under whatever name the record uses: advisor, promotor, directeur de thèse, orientador, Betreuer. A co-supervisor shares that role, and both make the lineage.
- Committee chair and committee member
- The dissertation committee of North American practice. Where a record names a chair and no advisor, the chair stands as the supervisor.
- Examiner
- A scholar the record names as having examined the thesis and judged it.
- Referee
- A reader who reports on the thesis in writing before it is defended: a German Gutachter, a French rapporteur, a reviewer in Poland or Czechia.
- Opponent
- At a public defence in Sweden or Finland, the scholar from outside the university appointed to question the candidate on the thesis, in public and at length.
- Custos
- At a Finnish defence, the professor who presides over it, often the candidate's own supervisor. The word is Latin for guardian.
- Jury
- The panel that hears a defence and awards the degree, as in France or Brazil, with its president and its members.
One link is suggested rather than recorded. In some German repositories a record names a thesis's referees and no supervisor, and there the first referee is so often the supervisor that the record draws that person as a likely supervisor. Such a link is marked inferred, drawn in its own style, and never counted as one a record cites. Where every link comes from sets out each kind of evidence.
Levels, and why the doctorate makes the lineage
A thesis has a level: a doctorate, a master's, a licentiate (in Sweden and Finland a degree on the way to the doctorate), a bachelor's, or unknown where its record does not say. The level is read from the record's own words for its degree or its kind; where those words do not settle it, the thesis is of unknown level, never guessed to be a doctorate.
The lineage follows doctoral supervision alone. That is the relation academic genealogies have always traced, and the one recorded in nearly every country: a long apprenticeship at whose end the student is an independent scholar, often with students of their own. Below the doctorate, supervision ranges from years of shared research to a single term's essay, and a family drawn through it would mean something different in every country. So the theses below the doctorate are kept and shown, on a person's page and on the sheet when its box for master's and bachelor's theses is ticked, but only doctoral supervisions are counted as lineage, and a thesis of unknown level is not counted as a doctorate.
Drawn beside, kept apart
The roles beside the supervisor come from the same thesis records and are cited as the supervision is. They are drawn on the sheet when its box for committees and examiners is ticked, beside the student whose thesis they sat on, and are never followed as lineage: an examiner is not an ancestor.
Wikidata states doctoral advisors and students for many scholars. Those statements are part of the lineage, marked as stated by Wikidata so they are never mistaken for a repository's record, and a person's page lists them in a part of their own.
Four more kinds of relation are drawn from other sources and never counted as family. Coauthors are the people OpenAlex, an open index of scholarly works, lists as sharing the most works with a person; they are suggested by shared papers, not recorded as training. Citations are the authors a person cites most and is cited by most, from the same index, drawn when the box for them is ticked. Teachers and mentors are Wikidata's statements that one person was another's student or teacher, a relation wider than supervising a thesis, with a few pairs read by hand and shown as likely, not confirmed; they stand in rows of their own above and below the person. And for a person known only by their papers, a card may show a stage of medical training, the degree, school and year that Medicare's national list of clinicians in the United States holds under their name; since a name is not proof of a person, it is said to be unconfirmed, it draws no link, and it is in no count.
A person with no advisor and no student in the record opens with their coauthors and teachers shown, so their page is not empty. Each can be turned off, and none of them changes who is family.
Why a graph
The record could be a table, a row for each supervision. It is a graph instead, people and theses joined by their roles, because the questions a scholar asks of it are walks. A lineage is a walk: from a person along their supervisors' links upward, a generation a step, and along their students' downward. A person's academic siblings are a walk up one step and down one. How two people are tied is a walk too, the shortest routes from one to the other.
The routes page finds those routes, between two people or among several, over the record's doctoral supervisions, Wikidata's advisors and OpenAlex's coauthors, each step saying which it went by. School to school follows doctors from the school of their doctorate to the schools where they later supervised. Neither is a question a table answers in one look.
On a real page
Marie Curie's page shows the shape on a person whose family Wikidata states: her doctoral advisor and her doctoral students, each marked as stated by Wikidata and shown apart from thesis records, and the registers she is linked to. A person a repository names shows the same parts, with the thesis record behind each link one press away. Who a person is, and when two records are one person, is the subject of who is who.