Because the business is required to coordinate with T-mobile on the technical side (source IP blocks, video streaming protocol).
You're accepting a video stream degraded to a bitrate that's still acceptable on a mobile device for unlimited video consumption (which requires technical intervention).
If you read the EFF's report on this, they show that T-Mobile is not cooperating with Google, et. al. in order to provide this service. All they're doing is artificially throttling the connection their own end, and letting the video service's automatic bitrate selection take effect.
https://www.eff.org/deeplinks/2016/01/eff-confirms-t-mobiles...
"The first result of our test confirms that when Binge On is enabled, T-Mobile throttles all HTML5 video streams to around 1.5Mps, even when the phone is capable of downloading at higher speeds, and regardless of whether or not the video provider enrolled in Binge On."
I support T-mobile's zero rating. You do not. We can argue the technical details all day long. I no longer donate to the EFF after their campaign against T-Mobile. I also would donate to whatever SuperPAC attempts to lean on the FCC to ensure zero rating is permitted to continue (as it behooves me as a mobile customer; I don't care about your startup, in the general sense).
Too often, Hacker News gets weighed down in the minutia, the little bits that don't matter to the non-technical. That's fine, of course. We love engineering for engineering's sake. But if you want to effect real change, remember that there aren't any strings to pull here to do that. Its simply an echo chamber of like minded folks with internet points to giveth or taketh away.
They are throttling a service because it does not participate in their BingeOn program. That is horribly anti-competitive and bad for users. Why do you want T-Mobile to slow down your connection to YouTube and give you absolutely no discernible benefit from it? It absolutely boggles my mind.
Because I consume almost all of my contact through BingeOn partners, and don't watch YouTube, so it behooves me? Does that truly boggle your mind?
Why do I care if YouTube is throttled if I don't use it?
That seems exceptionally short sighted, but to each their own. As you say, I doubt we're going to change each other's minds here.
There is, but you're fighting the wrong fight.
Twiddling bits for network management = okay.
Twiddling bits because you own a network and a content provider = not okay.
Get back to me when you're fighting the right fight, and I'm more than happy to help.
If this was twiddling bits for network management, then they would be done at the point where they throttle all video (which they are doing). But they are going past that, and also giving it away 'for free' to end users, but only if a service signs up.
So in either case (whether a service is signed up for BingeOn or not), a streaming service is having their videos throttled. So what benefit does T-Mobile get by giving away it's 'twiddled bits'? One answer is 'to try to make network neutrality violations more palatible'. However no answer could be 'traffic management' because they are already doing it regardless of the BingeOn status of the provider.
That's not correct. Non-participants are treated the same as participants as far as throttling goes--both get throttled when BingeOn is enabled. What BingeOn does is make it so that the throttled video from participants is not included in your data usage.
ISPs have been performing traffic management since their inception two decades ago. I've done network engineering for about 1/3 of my 15 year career. To be blunt with you, its definitely scalable. Could you explain to me why it isn't?
Imagine a world in which every ISP in the country had this system. Let's say that number is 1000 ISPs (with a typical 80/20 distribution).
As a video streaming start up, that's 200 ISPs that would need to be contacted and technical measures put in to place. It's likely that at least some of them are going to have incompatible requirements. So now the bar for a new (possibly better) video service is not just maintaining browser compatibility, but also signing contracts and meeting technical limitations for 10-200 ISPs.
Sure it works for exactly one company, but it doesn't work if the system is expanded to everyone. In other words, it's not scalable.
If you can't meet the technical requirements that provide T-Mobile the network benefit to support BingeOn, you don't get included in the program. Its pretty simple.
If you're a video streaming startup, you're not going to be serving content to ISPs. Your CDN is. Your CDN would have a process in place to integrate with whatever ISPs wanted to support this. If you're attempting to service content directly from your own servers, you're either enormous, or stupid.
Only it's not. If you need Internet transit, all you have to do is buy transit from two to three transit providers and you are all set. You'll get as big and as fast pipes as you like and at a price cheaper than doing it yourself. There is no need to make bespoke deals with 200 ISPs are there would be if everybody started with this BingeOn crap.
> If you're a video streaming startup, you're not going to be serving content to ISPs. Your CDN is.
Only if you are made out of money and/or don't know what you are doing.
> If you're attempting to service content directly from your own servers, you're either enormous, or stupid.
Neither enormous not stupid, just competent. You on the other hand have obviously never done video streaming and don't know what you are talking about.
My last gig was managing the video streaming infrastructure for a major US broadcaster. I think I've got a grasp on how its done.
Your CDN would have a process in place to integrate
So rates through CDNs go through the roof (even more than they are for streamed video content already) because now they have to peer every end user ISP, and as a video streaming service, those costs are passed on to you. So something that was previously "free" (so as not to be a net neutrality violation) has the same effect of increasing costs and reducing competition.
And then on the flip side, if a startup is trying to build a new (possibly better) CDN company, itstill has the same problem.
You didn't remove the problem, you just moved it, and it's still just as unscalable.