Hi there,

Yes, you can cache data on your end.

Just remember you need to attribute TMDb as the source of whatever data and/or images you decide to use.

Cheers.

so with "You shall not ... 3. Cache any TMDb metadata or photos other than for reasonable periods in order to provide your service." from https://www.themoviedb.org/documentation/api/terms-of-use the scenario issued above is a legit use-case?

@JB0urne said:

so with "You shall not ... 3. Cache any TMDb metadata or photos other than for reasonable periods in order to provide your service." from https://www.themoviedb.org/documentation/api/terms-of-use the scenario issued above is a legit use-case?

I will credit the movie database, provide link back to TMDB with exact movie or series or serial back here link. Maybe they ban it globally but not in this case because I will provide link back and clearly say that those data are not my but TMDB.

so with "You shall not ... 3. Cache any TMDb metadata or photos other than for reasonable periods in order to provide your service."

Focus on the "for reasonable periods". The crux of this point is more around the idea that you don't own the data. You are free to cache the data, but that is all.

I will credit the movie database, provide link back to TMDB with exact movie or series or serial back here link.

As long as you attribute TMDb you are ok.

sorry for reopening, but I want to clarify this before I implement something wrong :)

so I can cache the movie (/poster) data in a database to reduce the needed API-calls as long as I attribute anything correctly - but how (long) exactly? Lets say we have a webside with lists of movies and one list contains Marvels Avengers with the related metadata and poster-image (during the creation of the entry the datas (received from tmdb) are saved in a service database). After some days a user of the service wants to create a new list, also including this movie / data - is it necessary to doublecheck the data with the api for changes or is it allowed to directly use the data from the service-database (as mentioned before TMDb is always attributed correctly, so there is no misunderstanding where the data come from)? Maybe I should save the date when the query was triggered last time and update it after a certain amount of time?

happy coding JB0urne

@JB0urne said:

Lets say we have a webside with lists of movies and one list contains Marvels Avengers with the related metadata and poster-image (during the creation of the entry the datas (received from tmdb) are saved in a service database). After some days a user of the service wants to create a new list, also including this movie / data - is it necessary to doublecheck the data with the api for changes or is it allowed to directly use the data from the service-database (as mentioned before TMDb is always attributed correctly, so there is no misunderstanding where the data come from)?

This is 100% your decision to make at the time of implementation. I would imagine most of the time, you will want to expire data/check for freshness at some kind of set interval or based on certain actions but that's entirely up to you.

Dear The Movie DB

I'm Bokwang in charge of contents business at ZUM Internet.

ZUM Internet (www.zum.com) is the third-largest search portal in Korea, where 8 million users visit per month. Using your api in search results I'd like to do a movie search service.

We want to leverage your api and links as the TMDB has high-quality worldwide movies, TV programs, and character data.

I would like to inquire about the following items regarding API use.

1. Is it possible to freely use images such as movie posters and portrait photographs available in your company's api? All your data and images used by ZUM Internet are assigned the TMDB as the source.

2. To provide search results optimized for Korean, movie data will be saved and changes will be updated. " You shall not ... 3. Cache any TMDb metadata or photos other than for reasonable periods in order to provide your service."(link) Here, it is necessary to ensure that the 'reasonable periods' is understood to be a period of service using the TMDB api data. Of course, data will be destroyed except for the period of service using TMDB api data.

Please answer these two questions and I hope that this will be an opportunity to inform Korean users of the advantages of TMDB through the provision of ZUM Internet.

Thank you very much

@travisbell said:

@JB0urne said:

Lets say we have a webside with lists of movies and one list contains Marvels Avengers with the related metadata and poster-image (during the creation of the entry the datas (received from tmdb) are saved in a service database). After some days a user of the service wants to create a new list, also including this movie / data - is it necessary to doublecheck the data with the api for changes or is it allowed to directly use the data from the service-database (as mentioned before TMDb is always attributed correctly, so there is no misunderstanding where the data come from)?

This is 100% your decision to make at the time of implementation. I would imagine most of the time, you will want to expire data/check for freshness at some kind of set interval or based on certain actions but that's entirely up to you.

@travisbell, Thank you for your clarification on Games and JB0urne's questions. I am working on my first web app so I'm still learning how all of this works. Just to be clear, would the following be permitted?

1. Attribute TMDb on the site
2. Server has a scheduled discover GET for all pages every 24 hours that caches every movie from TMDB in my database (~25,000 requests based on 500,000 movies)

EDIT: It looks like discover only fetches up to 10000 results, so I'm not event sure if it's possible to do what I'm thinking.

It sounds like Letterboxd does something similar every 30 hours, but I can't verify whether or not they cache every movie, or just the ones requested by users during usage.

I ask because I ran into a roadblock where I want my users to be able to save information about a movie, and in order to link that to TMDb data without TMDb having a feature to fetch multiple movies by ID at once, it seems like the ideal way is to grab all movie data. Let's say one of my users stored 800 movie ratings on their profile, the only other option seems to individually request all 800 movies individually when they access their library (or limit with pages), and that will likely lead to performance issues.

Thank you for your time, and thank you for creating such an awesome resource for developers.