John Legere Just Can't Stop The Misleading Bullshit About BingeOn
from the give-it-up,-john,-it's-over dept
The big story of last week was T-Mobile CEO John Legere's meltdown over people calling out the bullshit claims about BingeOn "optimizing" mobile video when the truth is that it was simply throttling all video traffic (partners and non partners alike). Things got even worse when Legere decided to attack EFF and suggest that it was being paid to discredit BingeOn. The simple fact remains, however, that T-Mobile is throttling video streams (and downloads).Legere briefly went quiet about all of this, but on Monday came out again with yet another statement in the form of an "Open Letter to Consumers about Binge On" which is at least a little more honest, but is still mostly misleading bullshit -- the very thing T-Mobile has built its recent reputation on avoiding.
We invented Binge On to provide customers with an easy and effective way to stretch their data bucket. Knowing that the number one (and climbing) use of data out there is video, it was obviously the natural place to focus. Binge On is like an economy button built into a new car to save gas, and it’s a benefit that customers got the minute we launched, to use it as much as they want to. Period.Again that sounds good but is totally misleading. First of all, it's T-Mobile that sets the data buckets in the first place. So relieving consumers of the burden that T-Mobile itself placed on consumers is not a consumer-friendly move. It's punching someone and then claiming you're being nice by offering them a hand to get them off the ground. If you start the anti-consumer practice, it's not pro-consumer to roll back a tiny part of it.
Binge On is a FREE benefit given to all T-Mobile customers. It is and always has been a feature that helps you stretch your data bucket by optimizing ALL of your video for your mobile devices.If this were truly a "benefit" then why does it also apply to unlimited accounts (like mine)? Unlimited account holders don't need or want this "benefit" (and it's not really much of a benefit as we'll get to).
We use our proprietary techniques to attempt to detect all video, determine its source, identify whether it should be FREE and finally adjust all streams for a smaller/handheld device. (Most video streams come in at incredibly high resolution rates that are barely detectable by the human eye on small device screens and this is where the data in plans is wasted). The result is that the data in your bucket is stretched by delivering streamed video in DVD quality - 480p or better (whether you have a 2GB, 6GB or 10GB plan etc.) so your data lasts longer. Putting aside the 38+ services for which we provide FREE data for video through Binge On, as discussed below – this “stretching” of your data bucket is estimated to allow you to watch UP TO 3X MORE VIDEO from your data plan than before. This is a huge step forward.Again, it's worth remembering that when T-Mobile launched this supposedly consumer-friendly offering, they completely hid the fact that it applied to all video, implying strongly that it only applied to partners. In fact, the company's CTO argued that it was not even possible to identify many YouTube videos -- a claim that turned out to be one of the many lies T-Mobile has spread over this mess.
Second, T-Mobile keeps claiming that most users can't tell the difference between 480p videos and higher quality HD videos, but that's bullshit. In many cases the difference in quality is quite obvious. And, again, if this was all about having your data "last longer" there would be no reason at all to turn it on for unlimited account holders.
Also note that T-Mobile is being a bit misleading here, as its original marketing on BingeOn noted that the free video streaming did not apply to accounts that had less than 3GB on their caps:
As with virtually all of our Un-carrier benefits, we immediately gave it to everyone! First we reached out to all of our customers via email and SMS message, and told them all about the new functionality that was coming their way. Then we turned it on, for everyone! So if you are a T-Mobile customer – you already have Binge On!Again, this makes absolutely no sense for unlimited accounts, and the fact that it's not opt-in is just silly:
We strive to default all of our customer benefits to “ON.” We don’t like to make customers dig around to find great new benefits -- that is something a traditional carrier would do when they really hope you, the consumer, won’t take any action. Can you imagine the disappointment, if people saw our TV commercials about Binge On, then went to watch 10 hours of video expecting it to be free, and only THEN learned that they needed to go into their settings to activate this new benefit? That’s how the Carriers would do it, but not T-Mobile. Everyone has it from day 1, period.So instead of making customers dig around to find this (which is not a "great new benefit"), they make customers dig around to find how to turn it off because they don't want it. That's exactly how the big carriers do things. And, once again, there's simply no reason why it should ever be turned on for unlimited data users.
But here’s the thing, and this is one of the reasons that Binge On is a VERY “pro” net neutrality capability -- you can turn it on and off in your MyTMobile account – whenever you want. Turn it on and off at will. Customers are in control. Not T-Mobile. Not content providers. Customers. At all times.This is what T-Mobile is banking on as the reason why it's not violating the bright line rule against throttling in the FCC's net neutrality rules -- because there's a small "out" in the rules, saying that the no throttling rule doesn't apply to choices made by the end user to throttle traffic. Of course, that's assuming a situation where the end user proactively decides to slow down certain types of traffic, not where it's forced upon them, and there's a convoluted process to opt-out of it.
Either way none of this addresses the actual concerns raised by many T-Mobile subscribers: T-Mobile lied. It said that it was "optimizing" the video when the truth is that it was just slowing down the video. It doesn't change the fact that T-Mobile was far from transparent in explaining that this throttling (not optimizing) applied to all video, even with non-partner video. Finally, T-Mobile lied in insisting that this "optimization" would make videos load faster, when the reality is that for many video services it neither saves any data (the full file is downloaded, just slower), nor does it speed things up. Instead, it makes it buffer when there's plenty of available bandwidth.
That's what people are complaining about and T-Mobile ignores all of it, continuing to insist that BingeOn is nothing but a consumer friendly offering.
In the end, Legere gives a weak apology to the EFF that again fails to recognize why so many people took issue with his characterization of the EFF ("who the fuck are you? and who pays you?") and pretends that it's just about a difference of opinion:
I will however apologize for offending EFF and its supporters. Just because we don’t completely agree on all aspects of Binge On doesn’t mean I don’t see how they fight for consumers. We both agree that it is important to protect consumers' rights and to give consumers value. We have that in common, so more power to them. As I mentioned last week, we look forward to sitting down and talking with the EFF and that is a step we will definitely take. Unfortunately, my color commentary from last week is now drowning out the real value of Binge On – so hopefully this letter will help make that clear again.The problem wasn't "offending" EFF, it was that EFF did a good job exposing what T-Mobile is actually doing, and rather than responding to them, you freaked out, attacked them and their credibility and acted like they were some nobody shills. That's not offensive, it's stupid and raises serious questions about T-Mobile's intentions.
Again, what is the "value" of BingeOn, other than throttling video down? Legere still keeps insisting things that aren't really true at all. It's too bad, because Legere had built up T-Mobile to be customer friendly and his reaction to this whole situation has done serious damage to that reputation.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: bingeon, john legere, net neutrality, throttling
Companies: eff, t-mobile
Reader Comments
Subscribe: RSS
View by: Time | Thread
publicity
[ link to this | view in chronology ]
IT's suxch a shame that's he's literally frothed all of that away over this.
[ link to this | view in chronology ]
Actually that's probably the most legitimat thing he said in the whole piece. People like to say how much better high-resolution video looks, but actual tests show time and time again that they can't actually tell the difference reliably. In fact, most people will consistently rate a lower-res video as higher quality than a higher resolution one, if the low-res one is displayed on a larger screen.
Sad, but true.
[ link to this | view in chronology ]
Re:
on my iphone 5 i can easily tell the difference between 480p and 720p youtube videos.
[ link to this | view in chronology ]
Re: Re:
[ link to this | view in chronology ]
Re:
In one case it was someone who was watching the HD service they were paying for, on an SD TV. In a couple cases it was people using an "upscaling" DVD player and, gullibly believing that they were now seeing HD, not seeing a difference.
[ link to this | view in chronology ]
Re: Re:
Anything requiring 1080p or above probably shouldn't be watched on a 4" screen -- because even if the screen can display it faithfully, my eyes sure can't see that clearly anymore.
[ link to this | view in chronology ]
Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re:
This may shock you, but not everyone watches video solely for entertainment. If you need a screencast (which makes up about 50% of my online video viewing), 480p vs. 720p is the difference between a blurry mess and readable text.
[ link to this | view in chronology ]
Re: video quality
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re: Re:
[ link to this | view in chronology ]
Re:
Screen size can be more important than the resolution, especially if your eyes aren't the best and the screen is only a few inches. That's not the same as saying that the better resolution can't be noticed when you're actually comparing apples to apples.
[ link to this | view in chronology ]
Another bad analogy
The only way any button in a car will save gas is if the button causes the engine to use fewer cylinders. Look up 8-6-4 engines to see how that works, and worked out. (Hint: not too well.)
[ link to this | view in chronology ]
Re: Another bad analogy
Limiting how fast you accelerate saves gas
Adjusting when the car shifts can save gas
The difference might be rather small but you can indeed save gas through more methods than simply reducing the number of cylinders.
You can also save on gas and brake wear by not racing idiots to red lights or stopped traffic.
[ link to this | view in chronology ]
Re: Another bad analogy
Please go see the current line of Dodge HEMI engines and most of the V8 offerings from GM.
It actually seems to have worked out really well.
[ link to this | view in chronology ]
Re: Another bad analogy
On my Mercedes, the button changes the shift pattern. In Economy mode (the default--OMG!), the car gets up to speed by never exceeding 2300 RPM.
If you are in performance mode, it will go to 5000 RPM or beyond.
Obviously, getting to the same speed in half the RPMs will use half the gas.
[ link to this | view in chronology ]
Re: Another bad analogy
The PCM can change the cylinders around so this is transparent to the driver.
It isn't the same as the old GM 4/6/8 but it traces its roots there.
That aside, as mentioned, many cars have economy buttons which not only change the shift strategies but also the spark and fuel mappings.
And yes they work really well.
[ link to this | view in chronology ]
Detecting video in HTTPS is probably impossible
However, they can see you're connecting to Youtube, and they can just limit all Youtube traffic to a specific speed. Then the Google speed detector picks the 480p quality. You just don't notice being limited in speed when looking at lists etc.
Alternatively, they could use connection tracking to possibly see when a specific connection has lasted long enough that it's either a huge file download or a streamed video.
Really this is a difference without a distinction. It doesn't matter if they can detect video traffic or not if they can make a good enough guess to limit it.
[ link to this | view in chronology ]
Re: Detecting video in HTTPS is probably impossible
No. They can't. Nor can they reliably tell when you're connecting to anyone else.
Why?
Because the contemporary Internet uses an incredibly complex system of content delivery networks, caches, cloud providers, and other moving parts to serve up things like video. Thus video streams from YouTube and J.Random.Service may actually share large chunks of the same infrastructure. And since the stream is delivered (we hope) via HTTPS, it's opaque to T-Mobile. AND on top of all of that, this infrastructure isn't static: it changes on-the-fly (in response to load and outages) and it changes over time (as services are turned up or turned off).
This results in a crazy-quilt pattern of traffic that is often difficult to understand even for those of us who built it. So if T-Mobile is claiming that they can differentiate it, that their engineers have a handle on this, that they have somehow figured out how to solve this in software, then my response will be a resounding "bullshit".
Which is why, as I pointed out last week, they're throttling my (large) tcpdump downloads -- which aren't even video in the first place. They can't tell.
[ link to this | view in chronology ]
Re: Re: Detecting video in HTTPS is probably impossible
If DownloadSize = Big Then Goto ThrottleConnection
All fixed.
Damn engineers always trying to complicate things.
[ link to this | view in chronology ]
Re: Re: Detecting video in HTTPS is probably impossible
If you have YouTube.com followed by a large stream, it's pretty obvious what's happening. Plus, any changes are going to be noticed pretty quickly by viewing hosts and the sizes coming from them for all users.
For example, if YouTube started using Akamai servers, it would become apparent within 10 users. YouTube followed by Akamai = throttle that Akamai connection.
[ link to this | view in chronology ]
Re: Re: Re: Detecting video in HTTPS is probably impossible
The net's plumbing is not in the same simple, easy-to-understand state that it was 20 years ago. What you're thinking of would have worked then -- but it has very little chance of working now.
[ link to this | view in chronology ]
Re: Re: Detecting video in HTTPS is probably impossible
Unless someone along the way is caching youtube and responding to t-mobile's requests, your comment doesn't make much sense.
[ link to this | view in chronology ]
Re: Re: Re: Detecting video in HTTPS is probably impossible
Because the source of the packet is actually some CDN, and not YouTube.
[ link to this | view in chronology ]
Re: Re: Re: Detecting video in HTTPS is probably impossible
I suggest that both of you try learning a little bit about how the various streaming (video, audio, etc.) services actually work before presuming that this is an easily-solved problem. It's not.
[ link to this | view in chronology ]
Re: Detecting video in HTTPS is probably impossible
[ link to this | view in chronology ]
Re: Re: Detecting video in HTTPS is probably impossible
[ link to this | view in chronology ]
Re: Re: Re: Detecting video in HTTPS is probably impossible
Also, the metadata of the packets is *not* encrypted.
Also, you're the moron if you really think all big ISPs are not doing DPI constantly.
[ link to this | view in chronology ]
Re: Re: Re: Re: Detecting video in HTTPS is probably impossible
[ link to this | view in chronology ]
But you made me dig like a fucking West Virginian coal miner to find out how to turn the damn thing off. T-mobile: You owe me one canary!
[ link to this | view in chronology ]
Re:
There's a giant magenta button that says ON in big white letters. Clicking the button turns if off.
It's not like they buried it three screens deep in Settings > Advanced Settings > Video Settings > Advanced Video Settings or something.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
"adjust all streams for a smaller/handheld device"
Is this at all true?
Do they actually adjust the stream or do they just throttle the connection hope the stream is adjusted by someone else to compensate?
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Data Cap Punching Analogy
Way to steal my analogy, Mike.
Clearly I have the copyright on punching based analogies as applied to Binge-On.
You'll be hearing from my lawyers.
</sarcasm>
(It's a good analogy, not surprising that parallel construction happens.)
[ link to this | view in chronology ]
Re: Data Cap Punching Analogy
I'm pretty sure you need to patent the process of gaining favor with someone by only punching them once and explaining that you could have punched them twice...on a computer.
[ link to this | view in chronology ]
Re: Re: Data Cap Punching Analogy
</sarcasm> (Because a depressing number of people actually think like that...)
[ link to this | view in chronology ]
Re: Re: Re: Data Cap Punching Analogy
[ link to this | view in chronology ]
So he's selling this idea it's great but not wanting to say anything about throttling. It's a typical dodge when you don't want to address or answer questions. You see this time and again with the politicians. It's called change the subject.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
T mobile
[ link to this | view in chronology ]
Q.
[ link to this | view in chronology ]
Re: Q.
[ link to this | view in chronology ]