--- license: cc-by-4.0 language: - en pretty_name: 'Florida Local Codes of Ordinances: Publication and Currency, 2026' size_categories: - n<1K tags: - legal - law - empirical-legal-studies - municipal-law - ordinances - local-government - legal-research - Florida - code-of-ordinances configs: - config_name: default data_files: - split: train path: fl-local-codes.csv --- # Florida Local Codes of Ordinances: Publication and Currency, 2026 Where each of Florida's 478 local governments publishes its code of ordinances, who publishes it, and how current that code is. One row per county and per incorporated municipality, covering all 67 counties and all 411 cities, towns and villages. Municipal law is the hardest layer of American law to locate: there is no master index, and each commercial codifier lists only its own clients, so a reader who does not already know who publishes a town's code has nowhere to start. This dataset is that missing index for one state, and it records the jurisdictions no codifier carries, which is the part no vendor list can show. Currency is measured from the codifier's own stated codification date, so a code that has not been supplemented in years is visible as such. **478 rows.** Licence CC BY 4.0. Not legal advice. ## Where this comes from | | | |---|---| | Canonical record | https://doi.org/10.5281/zenodo.22247377 | | Code and methodology | https://github.com/stepuplaw/legal-empirics | | Research page | https://stepuplaw.com/research/ | | Author | Kevin D. Klagge, [ORCID 0009-0002-1385-8498](https://orcid.org/0009-0002-1385-8498) | | Source | US Census Bureau subcounty population estimates (vintage 2024) for the jurisdiction frame, joined to the public metadata APIs of the commercial code publishers. No court opinions are involved. | The DOI above identifies the **code**, which is a different object from this dataset. Cite the code when you are describing the method and cite this dataset when you are using the numbers. ## Columns | Column | Type | Meaning | |---|---|---| | `geoid` | string | Census GEOID: 12 followed by the county FIPS code, or 12 followed by the place FIPS code | | `name` | string | jurisdiction name without its type suffix | | `type` | string | county | city | town | village, taken from the Census, never from the codifier | | `county` | string | the county the jurisdiction sits in; for a county row, itself | | `counties` | string | every county the place falls in, pipe separated, largest population share first | | `population` | integer | Census population estimate, vintage 2024 | | `code_status` | string | online-html | online-pdf-only | no-online-code | unknown | | `publisher` | string | municode | american-legal | general-code | code-publishing | municipal-code-online | elaws | self-hosted | county-code | planning-council | none | unknown | | `code_url` | string | the verified URL of the code of ordinances; empty where no code is online | | `publisher_client_id` | string | the codifier's own identifier for this jurisdiction, where it has one | | `code_product_id` | string | Municode product id for the code of ordinances; the key to its table of contents | | `codified_through` | string | date of the most recent ordinance included in the code, as the publisher states it | | `code_last_updated` | string | date the publisher last posted an update online | | `pdf_available` | integer | 1 where the publisher offers the code as a downloadable PDF | | `pending_ordinances_count` | integer | adopted ordinances awaiting codification, per the publisher; null outside Municode | | `official_status` | string | official | unofficial | unstated -- recorded only where the page says so, never inferred | | `source` | string | municode-api | browser | manual -- how this row was resolved | | `source_checked_date` | string | the date this row was checked; codifier contracts move, so the date is part of the claim | | `notes` | string | free text, used mainly for jurisdictions no codifier carries | | `statement` | string | the row written as one self-contained English sentence, so it can be retrieved, quoted and checked on its own | Every row carries a `statement` column, which is the row written as one self-contained English sentence. A row of codes can be downloaded but not retrieved or quoted, and the sentence is what makes each row usable on its own. ## How it was built The jurisdiction list is the Census Bureau subcounty population file, which is independent of every codifier and yields exactly 67 counties and 411 municipalities, matching the Florida League of Cities directory. Each publisher's public metadata was joined onto that list by normalised name, and every jurisdiction no publisher carried was opened by hand in a browser. Two traps had to be handled: the largest publisher's classification field does not encode jurisdiction type, so type comes from the Census, and its library returns HTTP 200 with an empty JavaScript shell for addresses that do not exist, so every URL is verified against the resolver the library's own page calls rather than against a status code. Metadata only. No ordinance text is retrieved or republished, and the dataset must not be described as ordinance text. `datapackage.json` carries the Frictionless schema, `croissant.json` the MLCommons Croissant description, and `dataset.jsonld` the schema.org form. ## Limits **This is a point in time.** Codifier contracts move. Every row carries the date it was checked, and that date is part of the claim rather than a footnote to it. **Currency is the publisher's word.** `codified_through` reports when the code was last compiled, not that the compilation is correct or complete. **Official status is usually unstated.** Most Florida codes do not say on the page whether the online version is the official one, and the column records what the page says rather than inferring it. **This says where the law is, never what it says.** There is no ordinance text here, by design. **Read the study's own limitations section** in the repository before quoting a number. Each one names the specific threats to its own validity, including the ones that are unflattering.