• venice_1808_landregister_geometries.geojson|csv: the vectorization of Venice's 1808 Napolean cadastral maps saved as GeoJSON polygons. Relate to the textual data entries through the geometry_id field.
• venice_1808_landregister_textual_entries.json|csv: the transcribed textual entries of the cadaster's manuscript (called "Sommarioni") with the field geometry_id relating the group of geometry from the geometries' GeoJSON file. Some fields have been standardized such as the parcels function and the identification of their owner.
• venice_1808_landregister_standardised_people.json|csv: disambiguation of the poeple mentionned in the "owner" field from the registries entries into its own data set.
• venice_1808_landregister_aggregated_data.json: an aggregated version of the aforementioned files in a single JSON file.
• venice_1808_landregister_landregister_merge_log.json|csv: during the disambiguation of people that resulted in the file venice_1808_landregister_standardised_people.json|csv, some specific heuristic were used in order to merge multiple mentions of a person into a single object. The file contains all non-unique people that were merged along with the reason that triggered the merge encoded in a bit-vector.

• geometry_id: a non-null unique integer linking the entry to the group of geometries in the corresponding GeoJSON.
• parcel_id: a non-null unique integer attributed to each registry entry as an operational identifier used to link the people disambiguated in the venice_1808_landregister_standardised_people.json file. It is unrelated to the parcel_number and sub_parcel_number which are identifier actually written on the source documents.
• district: a non-null standardized string representing the "sestiere" (district) in which the entry belong. Possible values are San Marco, Castello, Cannaregio, San Polo, Santa Croce and Dorsoduro.
• parcel_number: a nullable string linking the entry to the coressponding cadaster on the map. The name in the original document is Numero della Mappa.
• sub_parcel_number: a nullable string used to indicate sub-division of the entry. The name in the original document is subalterno.
• place: a non-null string representing the toponym (place) locating the entry. This attribute is written in the original document above the entries in the Denominazione dei Pezzi di Terra column.
• house_number: nullable string of the physical door number of the entry, can be multiple numbers in a chain separated by commas. The name in the original document is Denominazione dei Pezzi di Terra.
• owner_transcription: nullable string denoting the owner of the entry. The name in the original document is Possessori.
• owner_standardised: nullable string which is a standardised version of the owner_transcription.
• owner_standardized_class: nullable string representing the class of institution present in the owner in the case the current entity is an institution.
• owner_wd: nullable string, holds the entity code of the owner whenever a matching record was found on wikidata.org.
• owner_right_of_use: nullable string, represents the right of use attributed to the owner of the current parcel.
• owner_type: nullable string, represents the type of owner according to 3 possible values: RELIGIOUS, SECULAR, ISTITUZIONE PUBBLICA. An english-translated version of this field is also available under the name owner_type_en.
• old_entity: nullable string, whenever in the owner field there was some forme of transfer of ownership, this field holds information regarding the current parcel previous owner.
• old_entity_standardised: nullable string which is a standardised version of the old_entity field.
• old_entity_standardized_class: nullable string, similar to owner_standardized_class but in the case of the previous owner.
• old_entity_wd: nullable string, similar to owner_wd but in the case of the previous owner.
• old_religious_entity_type: nullable string, very often the previous owner of parcels happens to be a religious type, this field represent the different type of such religious institutions. An english-translated version of this field is also available under the name old_religious_entity_type_en.
• owner_supplementary: nullable string, holds supplementary information regarding ownership of the current parcel.
• old_owner_right_of_use: nullable string, similar to owner_right_of_use but in the case of the previous owner. An english-translated version of this field is also available under the name old_owner_right_of_use_en.
• old_owner_type: nullable string, similar to owner_type but in the case of the previous owner.
• quality: a nullable string of the quality of the entry. It describes the type of good from the entry in a more or less systematic way. The name in the original document is Qualità.
• citation: a non-null string containing the reference to the archival document page from which the transcription of the current entry has been made.
• ownership_types: a nullable list of standardized strings representing the types of ownership. The possible values are AFFITTO, PUBBLICO, COMUNE and PROPRIO. This standardized information stems from the quality field. An english-translated version of this field is also available under the name ownership_types_en
• qualities: a nullable list of standardized strings representing the possible qualities of a parcel. This standardized information stems from the quality field. An english-translated version of this field is also available under the name qualities_en

Only the fields stored in the "properties" sub-dict per GeoJSON object are described, as they are domain dependant metadata, whereas all other fields are standard GeoJSON properties.

• id: non-null string, operational ID attributed to each single geometry, in the format <type>/<id>. Multiple geometries may hold the same id, it generally means a single parcel with many different kind of geometries (building with a courtyard for instance)
• parcel_type: non-null standardized string, represent the type of the geometry, mainly for display purpose. Possibles values are building, courtyard, sottoportico, street, water.
• parcel_number: nullable string, refers to the parcel number associated to the current geometry. Matches with the corresponding value from the "sommarioni_text_data" JSON file.
• area: a non-null float value describing the parcel's area in square meters.
• parish_standardised: nullable string, refers to the adminstrative delimitation of the parish in which the geometries likely falls within. This is a derived data from an interpretation of the parish boundaries as they were thought to be organized following the Catastico edited in 1740. The appartenance of parcel was computed by checking under which parish boundaries most of a parcel's geometry fell under.
• geometry_id: nullable integer, an unique identifier grouping all the different geometries to the sommarioni entries they correspond. Matches the values from the corresponding field in the "sommarioni_text_data" JSON file and serve as the key to link both files. Entries without a value for this fields are urban objects drawn on the map that were vectorized without a corresponding description in the registry (for instance, streets and waterways).

• uid: non null unique integer, unique ID of the person in the dataset
• nucleus_uid: non-null integer, identifier of the family nucleus
• family: non-null string, lastname(s)
• name: non-null string, firstname(s)
• title: nullable string, person’s title
• father: nullable string, father’s name
• father_is_q: nullable boolean, True if the father is dead, False otherwise
• mother: nullable string, mother’s name
• siblings: nullable string, names of all siblings
• husband: nullable string, husband’s name
• husband_is_q: nullable boolean, True if husband is dead, False otherwise
• other_notes: nullable string, any other note
• parcel_ids: list of integer, all the unique parcel IDs (parcel_id in the venice_1808_landregister_textual_entries File) where this person appears as the owner or one of the owners.

Each of the entries listed in the file holds two fields:

• geometries: list all geometries that fall within the relation between the (potentially multiple) entries of the registries and the different vectorized parcel delimitations. The fields and contained data from the object in this listing are the same as the ones from venice_1808_landregister_geometries with the omission of geometries_id (used to do the join) and parcel_number (already present in the text entries).
• text: list all the entries from the landregister that are related to the geometries listed in the aforementioned file. As there exist vectorized geometries without textual entries (for instance street network, canals, or parcels that for historical reasons have not been described such as the military area of the Arsenal) this field can be empty. The fields and contained data from the object in this listing are the same as the ones from venice_1808_landregister_textual_entries with the omission of geometries_id.
• people: list all the disambiguated persons that are mentioned to be in possession of the current parcel group. As a single persone can be owner of a multitude of parcels, duplicates object occur in this file contrarily in venice_1808_landregister_standardised_people.json. The uid field uniquely identifies the person in the dataset. All fields in the people objects are the same as the ones from venice_1808_landregister_standardised_people.json with the omission of parcel_ids.

• uid this is the ID of the person’s copy. This ID is built on top of the original person uid in the people_Sommarioni_dataset.xlsx. The uid of a copy for ID 123 will always start with 123_: e.g. two copies of a Person with uid 123 will have IDs 123_1 and 123_2, respectively.
• owner_uid: this is the owner’s ID that was merged into the unique person object.
• parcel_id: this is the unique ID of the parcel(unique in the venice_1808_landregister_textual_entries) where this person appears as the owner or one of the owners.
• suddetto: if this column is true, this person was merged because of the mention of "SUDDETTO" (meaning "as above").
• same_father: if this column is true, mentions of this person were merged because of identical father being mentionned.
• same_mother: if this column is true, mentions of this person were merged because of identical mother being mentionned.
• same_siblings: if this column is true, mentions of this person were merged because of identical siblings being mentionned.
• same_husband: if this column is true, mentions of this person were merged because of identical husband being mentionned.
• same_title: if this column is true, mentions of this person were merged because of identical title being mentionned.
• same_other_note: if this column is true, mentions of this person were merged because of identical other notes.
• same_name_only: if this column is true, mentions of this person were merged only because of identical names.

A folder contains a set of scripts that were used to assist in the cleaning, standardization, and preparation of the final dataset. The data processing workflow involved extensive manual interventions and the use of intermediate working files. As a result, the notebooks included here are provided for documentation purposes only. They are intended to illustrate the types of data transformations applied during the project, and are not designed to be executed sequentially to reproduce the current version of the dataset.