Say Your Name
One person, written down by six different clerks in six different languages. Type a given name and see every form the same human being would have been entered under.
A Josip in a Croatian register, a Giuseppe in an Italian one and a Josephus in a Latin one are the same name. An index that treats them as three people will never join a family up — and neither will a searcher who only types one of them.
firstname=josip returns records
firstname=joseph returns records
firstname=josephus 0
firstname=giuseppe 0
And the index's own «include name variants» box is on by default and still returns nothing for Josephus, so whatever it expands, it does not expand across the languages a Croatian parish was written in. Where the searching is recorded →
A name you have looked up is a link you can send. The address bar follows the search — ?n=Mihovil — so a card can go to a cousin or an archivist and open on exactly what you were looking at.
Two kinds of cell. A form printed plainly is one this archive has actually met in one of its own records. A form marked like this° is a lookup aid — the correct form in that language and the right thing to type into a Hungarian or German index tomorrow, but not something written down here yet. 174 of the 343 forms below are attested — 51%. Nothing decides that by hand: tools/names.py reads it out of the records, and the build refuses if a form the records contain is missing from the table. The first time that gate ran it found two men called Mathæus filed apart from the twenty-one Matijas they belong to, separated by a single t.
40 names · 343 forms · 174 met in these records, 169 as lookup aids.
The table was built from the men
Every name group here that has a Latin feminine — Joanna, Josepha, Georgia, Jacoba, Paulina — had nowhere to put her until 17 September 2026. Eight Joannas, two Josephas, two Georgias and a Jacoba fell out of the fold entirely, and so out of every variant list built from it.
That is the worst possible place for a hole. The women of this tree are already its weakest part — seven direct ancestors carry no surname at all, because a woman was entered under her husband's — and the one tool that would have helped somebody find them had been built from the men. Five feminine groups were added and the fold now holds 40 names.
Latin is one language with six endings
Michaelem is not another form of the name. It is Michael in the accusative, because the priest wrote baptizavi Michaelem — «I baptised Michael». Nobody was ever called it, and nobody was ever indexed under it.
And the sex is counted, not asserted. The fold declares matija a man's name. Eight of the forty people written that way here are women — one of them ahnentafel 13, Matia Pilipić, whose marriage is row 4 of the work list. Croatian Matija is unisex and the table had one column for it, so this page counts the bearers instead.
This matters more than it sounds. The family tree is built from what the register says, so the register's grammar walks straight into it: ten people in the emigration target list are named in an oblique case — Gregorium, Conradum, Joannem, Helenam, Georgium, Mathiam, Elisabetham, Joannam — and a search for any of them returns nothing for ever. tools/latin_cases.py already knew this about 47 people in this archive; what was missing was anything that used it before typing a name into a search box. The names, counted →