Here are a set of tools to try to identify and count the number of movies in the Internet Movie Database it is legal to distribute on the internet, either because they are no longer copyright protected (aka public domain) or because they are licenced using a license allowing such distribution (for example with a Creative Commons license).
The latest version of these tools and data sets can be found in https://github.com/petterreinholdtsen/public-domain-free-imdb .
Various sources are consulted, and here is a year histogram for the unique IMDB title IDs identified so far:
For each consulted source, a JSON file is generated. The JSON format look like this:
{
"http://www.imdb.com/title/tt0000005/" : {
"status" : "free",
"freenessurl" : "https://archive.org/details/blacksmith1893",
"wp" : "https://en.wikipedia.org/wiki/Blacksmith_Scene",
"wdurl": "https://www.wikidata.org/wiki/Q470705",
"title" : "Blacksmith",
"year" : 1893
},
"http://www.imdb.com/title/tt4044364/": {
"status": "free",
"freenessurl": "https://archive.org/details/Citizen.Four2014",
"wp" : "https://en.wikipedia.org/wiki/Citizenfour",
"wdurl": "http://www.wikidata.org/entity/Q18218448",
"title": "Citizenfour",
"year": "2014"
}
}
The key is the IMDB title ID of the movie, or some other unique and idenfitying URL if such IMDB ID is not known. Note the structure of the IMDB URL with trailing slash and www.imdb.com as the hostname. URLs must be normalized to this format.
The JSON fields are as follows:
-
title - the movie title, preferably the title registered in IMDB, but any title is OK. This is used to manually check if the mapping betweeen freenessurl and IMDB ID is correct.
-
year - the publication / screening year of the movie, preferably the year registered in IMDB, but any relevant year is OK. This is used to manually check if the mapping betweeen freenessurl and IMDB ID is correct.
-
status - indicate the status of the movie. Use 'free' for those that are distributable, 'unknown' if the status is unverified and 'non-free YYYY-MM-DD' if the status is verified and it is not legal to distribute the movie at the moment.
-
freenessurl - URL to a web page to back up the claim that the movie status is 'free'.
-
wp - link to a (preferably English) Wikipedia page about the movie. This normally form the basis for a wikidata page about the same movie.
-
wdurl - link to the wikidata page about the movie.
-
imdblookup - indicate the IMDB title ID is from a IMDB search, not a link found in the source. The content of the field is the search term used.
Only the 'status' field is required, and the 'freenessurl' field if the status is 'free'. One should try to fill in as many fields as possible for easier validation of the field.
The file free-movies-manual.json is manually generated, to have a
place to store the status of free movies that are currently missing in
the automatically generated lists.
A complete list of IMDb title IDs can be downloaded from IMDb. Details about the format is available on the IMDb interface page.
As the generated lists consist of statements of facts, as well as I know them, I consider the content to be non-copyrightable / public domain / Creative Commons CC0 (pick whatever suit your legislation :).
The program code used to generate the list is licensed under the GPL v2 or later versions.