FileBot and TheTVDB

[IFireflyl]

Hello! I just received an email from TheTVDB and I figured I'd create an account to see if this email would have any impact on FileBot users. This is the email I received:

Hello,
 
There are some big changes coming for TheTVDB that will affect both developers and our end users, most notably our brand new API (“v4”). With the launch of this new API fast approaching, we wanted to give an overview of all of these changes.
 
Two Models for API Access
There are now two ways to access TheTVDB API. Each company, platform, or project will have the ability to select their preferred method of access. This decision will ultimately be up to the developers of the projects themselves, so if you are an end user, you'll need to wait for the platform/software you use to announce any changes regarding your access.

Licensed

This is the way things work now. Companies and projects enter into a contract with us and are given an API key. License fees are based primarily on usage, company size, and how the data is used. 

Example: Mobile apps & websites

User-Supported
 
This is new. Some companies can’t or don’t want to license API access directly, and have requested that we pass along any cost to end users. We have determined an approach that will keep this affordable and accessible for all, detailed below.

Example: Synology Video Station

A note for Plex users: TheTVDB is in ongoing active discussions with Plex. Please stay tuned for more information if you are an end user of Plex.

User Subscriptions
User subscriptions are global, and not tied to a specific API key. This means that if, as an end user, you use three projects that make use of TheTVDB API, your subscription will grant access for any of them that choose the user-supported model.

Important Notes

• A subscription is required ONLY if the project/software you're using has indicated so — although we’d love for you to support the site anyway.
• Subscribing will grant you a unique PIN, which will be entered into the software you use.
• User subscriptions will be $11.99/year.
• Subscriptions will also include an ad-free site experience, a warm fuzzy feeling in your heart, and future incentives that we’re planning.

While we believe this to be a reasonable subscription fee, we understand that some of you may prefer another approach. With that in mind, we’ll also allow users to earn a free annual subscription by contributing quality data to the site, like missing IDs, translations, and artwork. We’ll provide additional details concerning this model in the coming months.[/list]

[rednoah]

FileBot has commercial license for TheTVDB since March 2020. However, that predates all the recent announcements, so we'll have to wait and see.


https://www.reddit.com/r/filebot/commen ... _the_plan/

[ByteShare]

Just as a note, if something falls through with TheTVDB and TheMovieDB doesn't step up.
I saw this and it seems to have some movement:
TheShowDB
https://forum.kodi.tv/showthread.php?tid=352588

[rednoah]

TheMovieDB works pretty well for both Movies and TV Shows as far as I can tell. Lots of users have already switched simply because TheTVDB servers are down rather frequently at times. FileBot does not plan to add additional databases at this time, but we are steadily improving support for --db TheMovieDB::TV to make things on par.


It's always good to have options though. TheShowDB might become a viable alternative. Either by running for a few years paying bills and proving themselves, or by explaining how they have a sustainable business model. FileBot is happy to pay to help with that.

[kim]

FYI:
https://www.reddit.com/r/kodi/comments/ ... iscussion/

TheTVDB Subscription Discussion

Two things to start:

I received permission from the mods to post this.

I'm aware this is an unpopular decision, but also want other users to see this post. I don't care about karma, but if you downvote the post itself then people won't see it. If you want to downvote all of my responses within, that's fine with me.

With that out of the way, the post itself. There are a lot of posts across a lot of subreddits, and many of the questions/comments are the same. So I figured I would put together responses for the most common ones. I'll also be responding to any comments or concerns within this thread to the best of my ability. Note that some discussions with companies and projects are covered by NDAs or are in active negotiations, which will limit some of my responses.

Why are you charging for something that has been free, with data contributed by users?

The site is extremely expensive to run. Storage is relatively inexpensive, but bandwidth and processing add up quickly. We have tens of millions of users creating tens of BILLIONS of API hits every month, with some months exceeding 1 Petabyte of bandwidth used. The result is infrastructure costs in the tens of thousands per month.

That cost excludes the staff it takes to keep the site running, and just as importantly, help improve the site moving forward. We’re aware of many of the past complaints (which I’ll address below), and we’ve been working to improve these things. Some of those areas, like increased reliability and the ability to have multiple orders for series, cost money to improve. And those improvements haven’t stopped; we’re still moving forward to improve the site and community for everyone.

What about the ads on the site?

Ads are only on the frontend of the site, not the API. Since the frontend of the site receives a fraction of the traffic that the backend does, ad revenue covers only around 10% of our infrastructure costs alone. We could drive more ad revenue by using more invasive ads, but we hate that approach as much as our users would.

Why not take donations instead?

We had a donation link on the site for over a decade, and our monthly donations averaged a couple hundred dollars per month. People just don’t donate voluntarily.

$12/year seems like a number picked by marketers, and is going to increase, right?

I picked out the amount, and although I’m a programmer I don’t mind marketing. Yes, the goal was to select an amount that people would find acceptable. For a very long time I paid fees for Usenet access and NZB indexer memberships. I felt like $1/month was an acceptable amount based on the cost of other services.

I won’t make empty promises about it never increasing, but we intentionally chose an amount that wouldn’t require it.

Why should we donate when the API has gone down so much or has been broken a number of times?

Valid point. Part of the reason we need to cover costs now is because we completely revamped our infrastructure to improve reliability. Part of the migration required reworking the old APIs to work with our new database. Other than an outage on AWS’ side, we’ve had very few outages over the past year. And those will be reduced even more once everyone is on the v4 API and we are able to turn off the old APIs.

Why does your API’s cache take so long to update?

This was necessary to help reign in costs with the old APIs, but the v4 API statically renders the content instead of dynamically generating and caching it. That allows us to reduce the update time for the API from 24 hours to around 5 minutes after a record is updated.

Why should I donate when your records are wrong and locked?

Our goal is always to improve the accuracy of the data. We actually have small internal team that certifies the validity of our records. We have access to other systems that end users don’t always have access to, plus relationships directly with some of the studios, production companies, and networks. Requests about contentious records are brought up in our weekly team meeting and we talk through the nuances to determine the right approach for each.

We understand that people will still disagree with our decisions on some of these records, which is why we’ve also implemented some features that allow both the official and unofficial/popular version of these records to be maintained.

A prime example of this is American Dad!. You can see the reasoning behind the official order in our knowledgebase (https://support.thetvdb.com/kb/faq.php?id=5). But since the community has demanded it, you can see on the American Dad! page itself (https://www.thetvdb.com/series/american-dad) that there are tabs in the season area and one is labeled “New Order FOX”. This approach is also used on The Chilling Adventures of Sabrina, Money Heist, and many others. This data is only available within the v4 API, but we feel like it will help resolve most of the contentious issues.

You’re still not storing sports, wrestling, etc though…

Actually, we’ve started opening these up. We have a mod dedicated to WWE and, as a result, have come up with an approach for this data. We also now allow UFC and F1. And we’ve brainstormed an approach that should allow us to handle every other type of sport, without the need to fit seasons and games into arbitrary orders that don’t really mean anything. Features like that cost money to implement, though.

You have toxic, power-hungry mods

This is the most common response, and I fully understand since I was the person that had to bring it up with them in the past. Most of the mods that people considered “toxic” have moved on. To be completely transparent, I fully understood their frustration. The toxicity ran in both directions, and I received a lot of it in my inbox as well. However, they needed to handle it better and didn’t.

The unrestricted and anonymous nature of the forums contributed to the toxicity, which helped drive our decision to shut them down. Now, we have a ticketing system with canned responses that not only makes things far easier for our mods, but also highly reduces the likelihood of an improper or misinterpreted response.

We also now have a full-time community manager that is working to improve our messaging, tools, and procedures. The combination of these changes with the previously mentioned site features (flexible seasons, especially) means that the “toxic mod” problem should be resolved.

We don’t expect everyone to trust our word for it. We know this is just something we’ll have to prove over time.

Why did you remove the forums?

In addition to the previous toxicity issue, they also lacked any sort of organization. There were dozens of pages of unhandled requests and it was impossible to tell what was happening with any of them unless we read every thread.

Now, we understand that the forums were useful for some things, though. The ticketing system, knowledgebase, and feature suggestion tools cover part of this. We’ll also be adding meta-discussions directly on the series and movie pages, which will help both mods and users understand the history and decisions related to records, and also give a place where these things can be discussed. I’ll be taking a thoughtful approach when implementing this to make it easy for users to report toxic behavior from other users and from mods.

While these other tools may not accomplish everything the forums did, I believe they’re a better starting point. We can always add other tools into the mix moving forward.

edit: I realized I didn't mention this, but we are implementing a data contribution model as well. Users that require a PIN can subscribe, or they can contribute data to the site. This will be done through an interface that allows us to validate the incoming information properly. The result is that we get better data for the community and users get free access if they need it. We've been trying to get the work time for a free PIN (valid for 1 year) around 2 hours, but that might go slightly up or down.

[rednoah]

I note that "TheTVDB API v4" still yields no results on Google or GitHub and no mention whatsoever on the apidocs page. FileBot is currently using battle-tested API v3 and will continue to do so until API v4 actually exists in a somewhat public and production-ready shape or form.