What. If. Every. Thing. You. Did. On. The. Internet. Took. This. Long.
Give me a break. There were no “good guys” in this debate, just corporations scrambling to control the popular narrative to suit their business interests.
Of course Google and Netflix were going to pay for peering and priority. The question was if everyone else who wanted to get into the video business should have to do so. Comcast wanted to basically double-bill for traffic: they wanted their customers to pay, and then they wanted the services on the other end to also pay.
This is all a moot question nowadays, because the video market is dramatically different than it was a decade ago. "Free-to-air" content is basically all handled by YouTube and Twitch, both of which are services run by companies with the money to afford the storage costs. "Premium" content fragmented into a bunch of different exclusive providers that already have business relationships with cable companies. So there really isn't a competitive video market being stymied out of existence purely because Comcast wants them to pay in order to compete with cable. The barriers to entry in video got a lot higher than just getting double-billed on bandwidth.
(As an example of this: Floatplane, an early-access "premium" platform for people already on "free-to-air" video platforms, had far more problems getting onto iOS than paying for bandwidth. Apple was their existential threat more than Comcast.)
With distributed computing & peer-to-peer storage networks such as IPFS, this may not be moot.
Right now, using something like PeerTube is taking your life into your hands. Look at all of the copyright nonsense that hits otherwise legitimate YouTube channels, and now imagine that they get to sue everyone who watched the video, too.
Centralized services can rely on DMCA 512 for liability limitation; and individual viewers almost certainly do not have copyright liability for merely watching a video on a central service. At the very least, you'd have to subpoena the service to get an IP address, and then subpoena the ISP to get at DHCP logs, and then argue to a court that a user that the temporary copies involved with merely watching a video had infringed your copyright.
When you join a swarm that shares a file, you just broadcast that IP address directly to the person who's going to sue you. That's how these networks work. There's a cottage industry of firms that deliberately "leak" their content to BitTorrent and then sue whoever downloads it; that also applies with PeerTube, IPFS, etc. You don't get DMCA 512 liability limitation because it's your own copyright liability. Furthermore it's easier to argue that someone in a P2P swarm has infringed copyright as there's uploading - further copying - involved. So both practically and theoretically speaking it's far easier to get sued in a P2P system.
Also, P2P video is terrible for mobile use-cases. Mobile devices are leeches in a P2P system - they don't have access to unlimited power like a desktop does. So you either flatten the batteries of everyone on phones, or you detect the use of a battery and lose the benefit of the swarm.
Also, it's ok to have leaches on a P2P system if you can properly incentivize & pay the hosts. Ethereum, FileCoin, IPFS, Holochain a market for hosting data. Ethereum & Holochain will provide a market for hosting apps.
I could give a —- expletive — about what the corporations think. What about normal people?
It feels like we need a better audio mix for public policy discourse.
Let's re-emphasize: the dire predictions of removal of net neutrality didn't come to bear. They didn't pass the sniff test either. The dire predictions were fake news - however they are now the official record in the media, and won't get corrected.
It really does beg the question why it was ever needed in the first place.
Nonetheless, the very charitable "perhaps the dire predictions didn't come to pass precisely because of the amount of controversy that was generated" would be neutrally described as "fake news must be published so that bad things don't happen". That's not a valid civilizational setup: it's both ripe for abuse, and also teaches people to dabble in double-think for sake of the greater good.
Why would they start to use that power while the process is still ongoing, and there's (as can be seen in this very article) still active investigations of fraud in the underlying process?
That also brings up why do they want (and spend millions on) the removal of net neutrality if they didn't plan on violating net neutrality in the first place?
This is an undue presumption of bad faith.
There are many situations where Net neutrality legitimately harms global internet routing efficiency and user experience. For example, under net neutrality ISPs are not allowed to privilege high priority packets such as real time video streams. It’s much more nuanced than net neutrality good, isps greedy and evil. Think about it from the perspective of a network engineer working for an ISP.
Net Neutrality does not block standard QOS techniques.
Additionally, ISP level QOS of IP packets based on content rather than src/dst is mainly vaporware anyway. You don't have the time to favor certain types of packets at the routers, the streams are going too fast.
That’s not true. Nailing down what ISPs are allowed to do in terms of QoS is still very much an open question. It’s not legally clear, for instance, why Comcast would be allowed to prioritize real-time video streams yet not be allowed to deprioritize BitTorrent sharing. What you’re referring to is peering discrimination but the whole of net neutrality is a larger and more complex subject.
Additionally, I'm not just talking about peering discrimination. The residential ISPs want the ability to restrict arbitrary sources and destinations at the last mile even if capacity exists on the rest of their network and their IX.
The caveat is that throttling is allowed if it is “reasonable network management” which is defined as actions being done with a “technical network management justification.”
And this is where legislation ends, under the ambiguity of reasonable and technical. Comcast tested this when a complaint was filed against them for throttling BitTorrent. I consider throttling BitTorrent “reasonable” and justifiable under purely “technical” reasons but clearly other people disagree. Enough people probably disagree that they had to deny any such throttling practices, meaning that in practice they would likely not be allowed to throttle based on technical reasons alone.
So what you're saying is that there was a process for figuring out the inevitable ambiguities, and you just disagree with the ruling?
As an aside, I also don't see the "reasonable", "technical" justification for throttling bittorrent anymore than throttling https.
The fact that what constitutes “reasonable” and “technical” QoS is ambiguous and strongly dependent on popular opinion is a huge design smell for net neutrality legislation.
> As an aside, I also don't see the "reasonable", "technical" justification for throttling bittorrent anymore than throttling https.
If 20% of your customers are using 80% of your network capacity because of BitTorrent then yes, throttling them is both technical and reasonable.
>No, we're not setting out to throttle Netflix. (We won't need to, because use will go up, and it's already in the contract we reserve the right to traffic shape anyway, and given we won't improve infrastructure, it's a given it will happen).
The stuff in parentheses is the unsaid part that would upset the normal people. The Network Engineers in the crowd heard it loud and clear though .
If Net Neutrality (and the incentive it creates through forbidding QoS to dynamically prioritize traffic), should have lead to investment in the infrastructure to increase network throughput to meet actual demand. Instead, POTS got torn down, broadband stagnated as ADSL lines were milked for every possible cent, Unlimited data plans disappeared and were replaced with caps...
We have more IP addresses than we should ever need, yet the biggest hurdle to robust connectivity is no one wants to spend money to actually get the wires strung/buried/overhauled. Why? because screw y'all. We're near if not de facto monopolies now, wires cost money, and they'd subtract from the exec bonus that gets cut.
https://www.cnet.com/news/verizon-throttled-california-firef...
When you start resorting to rationing (data caps), you have an infrastructure problem. Fix it. Don't embrace it.
(If anyone in their respective industries is in a position to benefit over their competition by making deals with ISPs, it is likely them)
This is reflected in the Netflix CFO's comments in 2015, among others: https://www.cnet.com/news/netflix-says-it-still-supports-net...
One of the most successful was the John Olivers Last Week Tonight piece. I saw some evidence later that the ‘evidence’ provided in that piece (claims from Netflix about throttling) were either fabricated or exaggerated, although I did not look into it too much.
* Can an ISP have charge for access to Spotify but not YouTube Music (and presumably get a kickback from YT)
* Can an Mobile phone forbid you tethering on you unlimited use plan?
* Can an ISP ban you using bittorrent?
Even in the case of your Spotify but not YT music example, what if there was a 5/month offer for 15mb internet with spotify streaming. Because spotify and your provider made a deal that subsidized the plan? At the end of the day, partnerships are not bad for the consumer. YT music could make a competing for deal for less money with more Google offerings. Also, this wouldn't eliminate an unlimited option across the board. When has competition made something more expensive? Its regulations like NN which pour amber over a system and make it impossible to be cheaper or better.
Peering is for fellow ISPs. Netflix, as you may or may not be aware, is not, in fact, an ISP. They already pay for their internet access. Do they pay enough? That's between them and their provider. Does their provider need to pay more for peering with its fellow ISPs? That's between the provider and its fellow ISPs.
What we do not need is for me to need to pay for my internet access and a Netflix subscription, and then pay extra just to let Netflix's internet traffic actually reach me.
Net neutrality would say no - they are a defacto utility and need to treat the traffic objectively and fairly without prioritizing their own at the expense of others.
Reprioritizing say ALL video traffic? Perfectly fine. Peering arrangements at the network link layer that would benefit some players due to locality and not others? Perfectly fine as long as they don’t do it explicitly to penalize a competitor (and even then probably fine).
Targeting certain services or protocols because they are a threat to their own products? Or asking for upsell money to get useful speeds for certain protocols not due to network management/bandwidth and handled objectively, but for revenue extraction? Not fine.
Would you like USPS or UPS to be able to charge you extra (the package recipient) to ACTUALLY deliver Amazon’s packages at the rate Amazon paid those companies already to deliver them, since they know you’re buying expensive things a lot and obviously have money? The extra load on their trucks from these packages is awfully expensive after all. Surely once a week deliveries will be fine for now unless you want to kick in? Don’t worry, the spam mailers will still be free.
Most of the peering talk was leaving that part out, trying to pretend that Netflix was using Comcast's network to reach third parties like is the model being addressed by peering agreements between tier 1 ISPs.