nyc_license now uses the October 2026 release of the NYC Dog Licensing Dataset. The table has 819,323 rows and includes the 2026 extract year.
nyc_license$breed_name now has standard whitespace and capitalization. For example, "Beagle Mix", "BEAGLE MIX", and "beagle mix" are now one breed.
nyc_license$breed_rc has a new, more complete recode:
Inverted names are in their usual order. For example, "Bull Dog, French" is "French Bulldog".
Each variety has the name of its breed, such as "Dachshund", "Poodle", or "Collie".
Long, abbreviated, and misspelled names have one standard form. For example, "American Pit Bull Terrier/Pit Bull" is "Pit Bull".
Each crossbreed is distinct from its breed and has the label "<Breed> Crossbreed".
The "Pit Bull (or Mix)" and "Labrador (or Crossbreed)" categories no longer exist.
nycdogs now depends on nycmaps for zip code tables and maps. It no longer depends on nyczips. R still attaches nyc_zips and nyc_zip_sf with nycdogs.
nyc_bites now uses the October 2026 release of the DOHMH Dog Bite Data. The table has 39,082 rows for 2015 to 2025. The previous table had 10,280 rows for 2015 to 2017.
nyc_bites$breed_rc now uses the same codes as nyc_license$breed_rc, so you can compare breeds between the two tables. nyc_bites$breed now has standard whitespace and capitalization.
nyc_bites has these breaking changes:
zip_code is now character, not integer.
unique_id and year are now integer.
unique_id is not unique in the table, because the source starts it again from 1 in each release. The combination of year and unique_id is unique.
The columns have a new order. year is after date_of_bite, and breed_rc is after breed.
nyc_bites has a new zip column. It is identical to zip_code and agrees with the zip column in nyc_license.