I happened to read this in one of the recent import issue threads… I can sort off understand the why but it’s really backward… you guys dropped TVDB only to make it a partial requirement again!? The IDs aren’t even shown to the user so it’s a hidden mapping even. Neither the import dialogue or help topics have been updated to reflect this requirement either (in addition to not being communicated in general). TVDB also gets duplicates so it’s not really a protection either, and what gets recreated most frequently in my experience are edge cases people can still circumvent like that anyways or by typing in bullshit numbers.
It would be bothering more if I were still creating new things as frequently as I used to but is that really the best way of handling things?
Hi,
The idea is mostly to block the spam pages and duplicates, as it is less likely that a page with at least 2 IDs is bad, than a page that only exists at TMDB. This decision was made based on the data we gattered from the pages that we end up having to merge or delete.
A lot of times we just ended up with 2 pages, one using TVDB as it’s datasource (back when we imported from them) and another using TMDB. And this would not have happened if the IDs were properly added at TMDB.
We check if the show has either the IMDB ID or the TVDB ID to have at least 2 options (based on the IDs that can be added at TMDB), to make sure no valid pages are skipped.
We are currently testing to see if this method improves the quality of the pages we end up having, and since it is a test, we haven’t changed the text on some places, but we will, if we decide this method works.
Thanks for the feedback!
So most of the merges are from the initial migration is what you are saying? That would mean even less reason to really require this. Caused by deciding to switch metadata sources one day and taking away the tools before it was finished…
Are you gonna update the help pages and tooltip though?
Yeah, I’ll discuss with the developers about updating that, if we are keeping that check.
To give some more data, more than half of the pages we have to delete are spam pages with very low data that don’t have any other IDs, they aren’t even duplicates, just spam.
The ones that actually have data, are usually duplicates and some are already listed as pages here using only the TVDB ID, so we need to merge them and migrate. If the IDs had been added at TMDB before importing, they wouldn’t even have been imported, since it would say the page already exists in the system.
So we can at least import a lot less bad pages. Even if it doesn’t resolve all the issues with duplicates, it certainly minimizes them.