UNFORGIVEN’s release date is shown here and does not match that shown in IMDB or the Movie Database. Where is this coming from? I frequently see odd choices showing up in the refreshed metadata but this one is off by almost a year and doesn’t make sense.
(2 cents more: As a Plex enthusiast, I’d like to be able to rely on a metadata field that tells me the “year the movie came out”. This should not be a fuzzy value. But often the Originally Available field seems to contain a date that, when not wholly made up, matches the first film festival appearance that may have been over a year before any official release. This doesn’t help me any, as a user, because it doesn’t match how anyone talks about movies and time. UNFORGIVEN came out in 1992. Others may feel differently.)
(1 more cent: I know Plex doesn’t manage the metadata. As a long time user, I’ve just become a little confused about where it’s coming from right now and how much control I still have over that choice.)
I then went into Agents > Movies > The Movie Database and unchecked “Plex Movie (Legacy)”, refreshed metadata and the date chanced to the correct date.
We edited the date on the back end last night. It shows the 1991 date was gotten from IMDB at some point but that may have been edited recently on IMDB which currently has correct and we just did not pick it up.
Seeing a completely inaccurate release date for RENFIELD as well, not matching either database. A Plex Employee says above that they “edit the date on the back end”.
What does that mean? Does Plex obtain movie release dates from IMDB or MovieDB.org or does it obtain movie release dates from a secret database manually managed by anonymous people for strange reasons?
I don’t mean to be nitpicky, but the problem is the YEAR of the release of many films is wrong. Getting this right seems to be fundamental to a movie metadata management application.
RENFIELD retrieves 11/30/22 as a release date, which is over six months earlier than any release date I can find for the film. Why?
Are you telling us that the Plex application uses the IMDB API exclusively to obtain the release date for films? If you are, are you also telling us that the result you get is essentially a random date with no relationship to the actual release dates listed on the site itself? And are you saying Plex staff is satisfied with this?
Whether you like/believe it or not every dataset on planet earth has some incorrect data. If you find something wrong you can tell us an we will do our best to fix it and find out what caused it to be wrong; and/or you can edit that on your server itself if you want.
I thought that was what I was doing. My other goal is to simply understand how the application is intended to work, as the docs and forum responses from Plex Employees on this issue remain unclear.
Release date here almost a full year before any known release date for the film. How. Why. Who is doing this and what kind of phrase is “Originally Available”?
Thanks for looking into some of my examples. I realize Ronin was “corrected” at some point. I should have refreshed before I posted it as an example.
I’ve now done a full refresh of the 5000 or so films in the library. It took me 0 seconds to find another one wrong, a film that was released 60 years ago. The year of this film’s release is… known. It is 1963.
The process by which Plex assigns “Originally Available” dates is poor, opaque, and looks arbitrary. IMO the release year of a film is the third most important piece of metadata next to title and director. Maybe others don’t agree. Manually correcting hundreds of films might seem silly to some. Try putting a ten best list together for 1972 and seeing that half the movies Plex lists weren’t even released that year. Then learn that the wrong dates appear nowhere but in Plex’s metadata. Then sit and wonder.
We just use the earliest theatrical release that we have (not film festivals and the like). As I have already stated dates can be wrong in a sites API even if the site page that you can see does not show it that way.
Is there a particular reason why Plex should “use” anything at all rather than deferring to a single known and canonical source (either IMDB or TheMovieDB)? It seems like a tremendous waste of resources to be handling metadata internally and for apparently not very good or transparent results.
I’m really just trying to determine how this piece of metadata is obtained when it is assigned to media because I have observed that a large fraction of release date metadata does not match the documented source of the information, which used to be IMDB and perhaps is now TMDB. Plex’s documentation is unclear.
Plex Employees’ posts on this topic have not clarified matters, but I’m gleaning through what they’ve said and common sense that the Plex back-end, so to speak, pulled release data from a source (IMDB? TMDB?) and cached it, where it is accessed by Plex server installations when metadata agents seek the information for local assignment to media. This may continue to occur periodically to get metadata as new media is released to the world.
Of course it wouldn’t make sense that when we scan our libraries a load of API calls go out to TMDB to get their metadata on whatever movies we’re scanning. We don’t get cart blanche to use those APIs. Instead, Plex HQ’s cached data is fed to us.
Plex employees’ mention making manual corrections to this data based on some of the errors I’ve pointed out, which does seem to be a Sisyphean practice unless one assumes Plex doesn’t refresh old metadata from the canonical sources such as IMDB or TMDB any longer, likely due to the costs.
Whatever the case, release dates have gone too wrong to trust without verifying if one cares enough about a particular datapoint. As I said, the year the movie came out being wrong was a big deal and too common and had cascading effects on how I manage and present my collection. Hence my harping on it.
Okay, but users can fix the “wrong” data by using those sites – and even if they don’t, tens of thousands of other users do – whereas with the new Plex setup, users need to report things in threads, hope staff see it. If staff don’t see it, hope volunteer moderators see it and are convinced to report to the staff. Then the staff report it to other staff who aren’t on the forums and handle metadata, then after an investigation and depending on resources the problem will or won’t be fixed. Isn’t it a little bit obvious why this seems like a worse outcome for users?
Like just to put my cards on the table here, it seems like either metadata sources have changed their business model in a way that Plex feels they need to counter (perhaps large programs are being charged more for metadata syndication?), or else someone up the Plex hierarchy has decided internally handling metadata is an important goal for completely inscrutable reasons, or maybe it has something to do with aspirations for the free ad-supported streaming service – the cause of almost everything that seems to degrade or frustrate long-time paying customers of the core product – but either way to a user it feels like data is being migrated away from systems that work towards systems that work less well.
