I do not understand the backend reason for this (and did not when I saw the announcements about this two years ago): while I can see not wanting to provide a full movie metadata API, having "is this or is this not available on Netflix right now" seems like a really important API to have, one that allows websites people use to browse and find movies to funnel users into Netflix (and seems a natural fit for their very popular affiliate program).
Apple actually provides massive database dumps of everything available in iTunes, for free, to anyone who asks, which is why virtually every music discovery site or app has a "buy this on iTunes" button, not just "a small set of developers whose applications have proven to be the most valuable for many of our members" (which is a quote from the Netflix developer blog earlier this year; you have to pull this from Google Cache now, sadly).
I don't see how "stream/rent this on Netflix" is a bad thing for Netflix (but am totally willing to believe I am not "seeing far enough", because this is not my area of expertise, hence why I am asking this question: I hope to be enlightened ;P). (edit:) I guess maybe for the same reason Uber and Lyft don't want APIs? Because they want to dominate the end user searching and discovery experience as well? But like, even Walmart and Amazon offer APIs... ;P.
Netflix is great, don't get me wrong... I wouldn't want to do without them. But at the same time, a large portion of their business model is devoted to making sure that nobody can tell just how little content they have to offer. There is no other explanation for deliberately nerfing their search engine and killing off third-party apps.
If and when Netflix is able to reach fair and reasonable streaming contracts from all of the major studios, I believe the API will come back. Right now, though, it's more of a liability than an asset to them.
>But at the same time, a large portion of their business model is devoted to making sure that nobody can tell just how little content they have to offer.
You nailed it.
Netflix is the best thing out there for legal streaming TV and movie content; it's that "legal" part which makes things complicated. Objectively speaking, Netflix clones that pop up but have no qualms about distributing pirated content provide a much better experience for users, because just about everything they want to watch is a click away.
In fact, Netflix's entire UI was designed with this in mind. All the "main" views try to give the impression that there is a lot of content, yet the search functionality is not very visible or easy to find and takes 2 steps to actually use. This seems to be a rule that applies to many applications: when your home page is just a search bar, that often means 80% or more of what the majority of users want is available. When your home page is a directory, or just a flattened spattering of content, you know that user's options are constrained.
Netflix is trying to push customers to use their application as more of a feature listing rather than mislead them into thinking it's a universal "watch whatever you want", because they know people will be disappointed if they think it's the latter (and I would assume A/B testing and metrics have demonstrated this to them empirically). So they go with a model of "here's a bunch of stuff you can choose from, you may not have heard of this show but it's good!" instead of "your own personal streaming media hub".
Obviously one can't fault them for this, since they have to rely on B2B relationships that don't always necessarily provide an excellent value proposition for the partner business and/or are very expensive for Netflix, and they have to obey the law. I'm also hoping they can work out more contracts with more studios.
> "But at the same time, a large portion of their business model is devoted to making sure that nobody can tell just how little content they have to offer. "
But this doesn't quite scan. Other services with plenty of content such as YouTube also nerf sort by ranking functionality.
If I have 100 things to show you as a subscriber and I know that you will only really enjoy 10 of them, than the argument is showing you all 10 up front means that you burn through those and don't stay around for the other 90. Of course there would be a margin of a few shows until you figured out that nothing was left of value, so say 10 you truly enjoy (the "Value Set"), and 10 mediocre ones you have to watch before you decide there's nothing else worth staying subscribed for (the "Transition Count").
How is this better than showing you only 2 of the 10 you'll like from the Value Set, and than making you hunt around for the other 8 until you decide there isn't any more valuable content even though there are 8 more items you would like. It seems like it just starts the Transition Count sooner.
Hmm, maybe it's a gambling phenomenon, in that inconstant rewards keeps the gambler playing? The odds of getting a show from the Value Set before completing the Transition Count is high enough that it just keeps resetting the Transition Count to 0?
Yeah, I think it's an inconsistent-reward thing. Netflix wants you to be pleasantly surprised when you stumble across something you enjoy, not frustrated because there's 57,000 channels with nothing on. Apps like abetterqueue had the effect of letting the user peek behind the curtain, so they had to go.
When a site like Craigslist that has total control over its content pulls this kind of user-unfriendly trick, I have nothing but contempt for them. But I'm more sympathetic to Netflix. In negotiating deals with Hollywood, they are dealing with some of the most primitive beetle-browed troglodytes in the business world so I don't have to.
It is probably a business decision to 'own the data' and control the storefront/display for margins. However not realizing that the data/api part is what helped them along in the first place and all this means is they lose control of the data.
Cutting off the API does nothing to stop scraping which will happen more now and not funneled into a controlled/throttled/tracked usage of that data.
Worst for Netflix is that other competing services will own more and more data instead of relying on theirs. Another streaming service could come along with that type of developer love again and catch Netflix high centered.
This is a very business over technology/open decision here for a Netflix that has moved away from developer support with things like the Netflix Prize, that time is done and over.
The argument of public vs private api usage that Netflix states how public api is only 0.3% is not valid because most public api usage just pulls data to cache and rarely hits the endpoint where private apis they use constantly track progress/settings/state etc.
while I can see not wanting to provide a full movie metadata API, having "is this or is this not available on Netflix right now" seems like a really important API to have, one that allows websites people use to browse and find movies to funnel users into Netflix (and seems a natural fit for their very popular affiliate program).
Wouldn't that API use would also funnel users to its competitors in the case that Netflix doesn't offer the movie you were looking for?
I see their API closing as a natural business decision — anticipating a time when streaming services are a commodity and margins are thin on streaming licensed content they don't own. They'll have to compete with something else: their core service, their own content, and their recommendation engine (which I find to be fantastic). A public API fights against the recommendation engine. They want you to sit down and watch Netflix, not a movie in particular.
> Wouldn't that API use would also funnel users to its competitors in the case that Netflix doesn't offer the movie you were looking for?
I don't know what you are envisioning :(. What I am talking about is when you find yourself on some random movie review website, and there's a "buy/rent on iTunes" button, and a "stream from Amazon Prime" button, but no Netflix button, that sounds like it would be bad for Netflix. Especially so if the only reason the person added the "stream from Amazon Prime" button is because they wanted to have a "stream" button, were sufficiently lazy to only put one such button, and Netflix (their first choice as having a longer-term brand) didn't have an API available. Like, I can't imagine Apple ever deciding "no, we would rather have you put a buy from Walmart button on your website than a buy on iTunes button, so we are going to stop giving you database dumps".
Apple actually provides massive database dumps of everything available in iTunes, for free, to anyone who asks, which is why virtually every music discovery site or app has a "buy this on iTunes" button, not just "a small set of developers whose applications have proven to be the most valuable for many of our members" (which is a quote from the Netflix developer blog earlier this year; you have to pull this from Google Cache now, sadly).
I don't see how "stream/rent this on Netflix" is a bad thing for Netflix (but am totally willing to believe I am not "seeing far enough", because this is not my area of expertise, hence why I am asking this question: I hope to be enlightened ;P). (edit:) I guess maybe for the same reason Uber and Lyft don't want APIs? Because they want to dominate the end user searching and discovery experience as well? But like, even Walmart and Amazon offer APIs... ;P.