In general, I think if it's the case that you can't reasonably defend a view you don't agree with then there's a very good chance you don't understand it. I was in the same boat as you until I started looking at Pai's comments [1][2][3]. Notably those are all transcripts. What he actually says, and the implications of the headlines and comments snipped by third party organizations tend to be quite different indeed!
While certainly missing plenty of nuance, his view is that rules should be written as necessary instead of creating an enormous rule book based entirely on today's tech. And to some degree, I think this has a lot of merit. Take for instance SpaceX and Google's variations of widespread air-based internet. If the throughput for these systems starts to become an issue would it be so unreasonable for them to throttle or block sites consuming disproportionate resources? And yes, this could absolutely be used in an anti-competitive fashion. What Pai is saying is deal with specific bad behaviors as they arise instead of building an encyclopedia of rapidly dated laws against many actions - which in some cases could be completely justifiable and beneficial.
Maybe the most basic argument is that it should be clear that net neutrality has not really achieved anything. Many people seem to think we always had net neutrality and stripping it away is some radical shift. In reality the government had no formal regulatory role on the internet until 2015. What was the state of the internet beforehand? Has it worsened or improved since then?
>> Take for instance SpaceX and Google's variations of widespread air-based internet. If the throughput for these systems starts to become an issue would it be so unreasonable for them to throttle or block sites consuming disproportionate resources?
No this is not appropriate. Sites do not 'consume' resources, users do. Let the market do it's normal job of supply and demand. If millions of users want to match netflix at the same time and the network cannot supply that, then raise the prize and invest in more infrastructure as any company with a demand surplus would. Markets _work_ in such situations, there's room for competitors, investments, increased profit margins and all those other good things. There is little information asymmetry: either netflix works when you want to chill or it doesn't. People _will_ vote with their money. Protecting a company from a demand surplus?? MADNESS!
Instead of nitpicking further on this flawed example instead turn your argument against itself. Why create rules of which you _know_ (as you yourself mention) that they are going to be abused, only to later 'deal' (take more 'campaign' money) with them. There is a very clear conflict of interest between ISPs that are also content creators and their content creating competitors.
I too ascribe to the notion that, whatever the legal code might say, companies are not people. They should not be able to abuse the rights that people have to achieve their business agenda. Their customers' bytes on their switches are not 'speech'. They should be dumb pipes, selling access to their dumb pipes and compete with other designers or dumb pipe networks, and that should be it.
> Let the market do it's normal job of supply and demand.
I agree with your point in general, but careful with this argument as it can favor those wanting to remove rules instead of add them. When it's phrased as "what if there is market demand for a cheaper internet package for lower-byte-producing sites or specialized-cheaper sites only?" You might say "Whoa, wait, I didn't mean let the market do its normal job of supply and demand, I meant require X regardless of demand."
Of course we can all agree that the market can't do its job anyways if a new ISP can't be easily built to satisfy consumer demand requirements.
It's a very important argument for NN though. One of the biggest problems with removing NN is that it will stifle competition by allowing pipes to interfere with normal market forces of supply/demand for the flow.
I admit that it requires actually thinking for a moment to see how regulation in this case can actually promote competition, and you'll have trouble getting many politicians to admit that or even understand it (when, as someone or other said, their paychecks depend on not understanding it).
Just because there is demand for something, doesn't mean it should be legal. Anyone claiming otherwise is deliberately trying to deceive you or themselves (or truly believes in human trafficking should be legal).
Saying that the market can effectively limit demand for something that was a limited supply (air waves) is VERY different from saying that anything that there is market demand for is good for society or the economy as a whole and should be allowed.
> No this is not appropriate. Sites do not 'consume' resources, users do.
You're drawing an arbitrary distinction between push and pull. Nexflix uses resources by sending content in the same way I use resources by uploading a file. Internet communication is a two way street. Far more resources are used when you talk to Netflix than when you talk to the NYT and someone's paying for them. Users don't think it should be them so why not the other end?
> raise the price and invest in more infrastructure as any company with a demand surplus would
Right, so they'll charge extra if you want to use Netflix and use the profits to fund that network. Err... "We charge an additional fee to access sustained video content as that puts a large strain on our network and reduces the quality of service for your neighbors." You could charge everyone more across the board but but that's unfair to people who aren't streaming.
> There is a very clear conflict of interest between ISPs that are also content creators and their content creating competitors.
Perfect, so fix the conflict of intersts and there's no need for NN. If my network oeprator is using QoS I can be reasonably sure it's to provide the best service to everyone.
> They should be dumb pipes.
Not possible. Cross-network traffic is inherantly political and even without that benevolent network operators still need to have policies to mediate allocation of a shared resource. Unless you're paying for a dedicated line between your source and target your traffic is being governed by some set of policies.
>> No this is not appropriate. Sites do not 'consume' resources, users do.
>You're drawing an arbitrary distinction between push and pull. Nexflix uses resources by sending content in the same way I use resources by uploading a file. Internet communication is a two way street.
You both seem to be laboring under a misunderstanding. Both Netflix and the customer are consuming resources and paying for them. Netflix pays for it's access and usage of the network, and so does the customer. Their networks have peering agreements to pass the data between networks. The question isn't if both should be paying, but if either one should have to pay more depending on where in the network the other one is or on the 'flavor' of the packets that are being sent. Net neutrality is supposed to protect Netflix from having to Pay Comcast for access to Comcast customers (in addition to what they pay for their network access), and it protects customers from having to pay Comcast for access to Netflix (in addition to what they pay for their network access).
> You could charge everyone more across the board but but that's unfair to people who aren't streaming.
Or you charge people for the capacity they use and charge more during peak times. This protects the problem without allowing the granting the current oligopolies massive market distorting power.
> Perfect, so fix the conflict of intersts and there's no need for NN.
Ok, fix those conflicts of interest first, then we can consider taking a look at crafting exceptions to the laws that protect us from.
The current effort isn't to make exceptions for QoS, the current effort is the repeal of the framework that allows ANY legal protection for Net Neutrality.
> Not possible. Cross-network traffic is inherantly political and even without that benevolent network operators still need to have policies to mediate allocation of a shared resource.
In what way is it not possible to create fair rules for who pays to build connections between networks and to mandate fair treatment in the creation of peering agreements between networks? Perhaps perfect fairness is not possible but you are saying that we shouldn't even be allowed to try.
> Or you charge people for the capacity they use and charge more during peak times.
It really doesn't matter. Since NN really only matters in the content delivery game this is just more directly giving ISPs what they've been asking for. It might make some people feel better but it's still a surcharge for competing video services.
> In what way...
Because it's still just as political, now the negotiation just takes place in government offices and lobbies. You might as well let companies hash out the agreements between themselves because the law will still be written to favor large incumbent players but now it's even harder to change. If things get dire the gov't can still step in.
> Netflix has peering agreements with Comcast, Verizon, AT&T and a lot more
Comcast's anticompetitive rent seeking behavior was a strategy of deliberately degrading Netflix's network access to Comcast customers to the point where Netflix was forced to pay Comcast or start losing customers.
This is the EXACT reason we we need better net neutrality protections, so that this type of oligopolic rent-seeking behavior is illegal.
>Comcast was the first large terminating access network to successfully implement a “congest transit pipes” peering strategy to extract direct payment from Netflix, but it is not the only one to do so. Since agreeing to pay Comcast, Netflix also has agreed to pay TWC, AT&T and Verlzon for interconnection. [Redacted section.] Netflix is not the only edge provider to encounter Comcast’s peering strategy. In a 2011 filing with the Commission, Voxel, a hosting company relying on Tata for interconnection with Comcast’s network, noted that “[w]here broadband ISPS typically ensure that links connecting their customers to outside networks are relatively free from congestion, Comcast appears to be taking the opposite approach: maintaining highly-congested links between its network and external ISP.” The letter concludes that Comcast, through its “interconnection relations,” had “deployed an ecosystem in which hosting companies such as Voxel are effectively forced to pay Comcast to serve its broadband subscribers.”
https://qz.com/256586/the-inside-story-of-how-netflix-came-t...
> It really doesn't matter. Since NN really only matters in the content delivery game this is just more directly giving ISPs what they've been asking for. It might make some people feel better but it's still a surcharge for competing video services.
What are you talking about? Charging customers more to a specific streaming provide is anti-competitive since they are at a disadvantage compared to other streaming providers. Charging customers more to access any streaming provider directly benefits Comcast's competing cable bundles and is also anti competitive.
Network neutrality is important for all kinds of stuff. From allowing the development of new P2P protocols, to the experience people have on different websites. Latency in browsing has a HUGE effect on how people perceive the comparative quality of sites.
>> In what way...
>Because it's still just as political,
You said it was NOT POSSIBLE, please back that up. Nothing you have said about negotiating peering agreement negotiationss and network interconnections makes dumb pipes not possible.
That's an uncharitable reading of the article you posted. Netflix pushes so much traffic that any company they buy transit from ends up using up all the good will of the people they have peering agreements with.
> Netflix attempted to address congested routes into Comcast by purchasing all available transit capacity from transit providers that did not pay access fees to Comcast.
And Netflix apparently knew they couldn't afford the traffic if they actually had to pay for it so they tried to route around it. And when they saturated those connections they had the audacity to tell Comcast it was their fault and they should foot the bill for infrastructure upgrades.
Netflix went to an all-you-can-eat restaurant, ordered two of everything on the menu, complained that the food didn't come out fast enough, tried to pay everyone else in restaurant for their seats, and then demanded the restaurant hire more chefs.
> “it is simply not possible for competing external providers to deliver gaming, or streaming video services to Comcast’s broadband subscribers” without directly or indirectly paying Comcast"
Well no shit since any company providing those services is going to push so much asymmetric traffic that either they or the companies they buy transit from are not going to be able to negotiate settlement-free peering agreements.
This isn't a "I send you some traffic, you send me some, no biggie" kind of agreement, this is a "I'm going to use 70-90% of your bandwidth at peak".
This article is trying really hard to paint Comcast as the bad guys but I actually don't think that's the case here. Netflix/Cogent had a similar spat with Verizon.
> What are you talking about?
I'm saying that allowing companies to price by volume and on/off-peak is really just a surcharge for streaming video which could be used to hurt competitors. It's not really a solution. So I think we're in agreement.
> You said it was NOT POSSIBLE...
I think we need to agree on what 'dumb' means. Connections between networks are inherently political no matter what entity presides over the negotiations. And since network access is a shared resource there will always be a need for network management policies to maintain service, respond to incidents, prevent spam, stop attacks, etc.. Both of these things make networks 'smart'.
I've read these arguments over and over, including your [1-3]. First, Pai clearly decided whose side he was on, then came up with these arguments. (And yes there are sides: the abolition of NN will transfer wealth to ISPs from everyone else in a zero-sum process.)
Second, lots of what he says are clear lies or contradictions. "I don't have predetermined views on where this will go" ... he announced many times exactly where he wants it to go! "I just want a competitive marketplace" ... he is trying to remove limitations on size of marketshare!
Third, we've seen his "arguments" over and over and they don't hold up. His main argument completely lies in revisionist history. The idea that somehow there was a "bipartisan light-touch" approach that was carefully applied and managed to grow the Internet, therefore imposing NN will destroy growth. The truth is very simple, that NN has almost nothing to do with infrastructure investment, it will just allow them to exploit their position as middlemen, and the truth is that large ISPs don't invest in infrastructure because they are unregulated near-monopolies and abolishing NN won't change that.
The "wait-and-see approach" is a classic vague generality that applies very poorly to this situation. There's a specific thing that the vast majority of consumers don't want these companies to do (discriminate traffic), and they have past histories of trying it only to widespread outrage, so we got a very simple, general regulation in place to stop them from doing it.
If anything, the wait-and-see approach he claims to support would be the one that leads to "an encyclopedia of rapidly dated laws". He says he wants to play whack-a-mole with laws, when we already have one extremely simple law that will age just fine: when a consumer pays for traffic, they get that traffic treated equally regardless of content or source.
Apologies if I seem angry at you but it really galls me when people are taken in by this.
> What Pai is saying is deal with specific bad behaviors as they arise...
You sound so sure, and you just take Pai unskeptically at his word, but is there a specific reason to think that this will actually happen? If you look at the most famous recent case of deregulation (in the financial markets), what eventually happened was that all the bad behaviours enabled by deregulation just continued without any regulatory action until the whole thing blew up. That's no model for governance.
If there isn't a specific reason (like a plan with some details, so that it's actionable), then it's also quite reasonable to think it's a bit of an excuse, intended to give cover, and a fairly transparent one at that.
>What Pai is saying is deal with specific bad behaviors as they arise instead of building an encyclopedia of rapidly dated laws against many actions
What evidence do we have that the US Government will do this? What mechanism has Pai proposed we put in place to ensure that we address them?
It's true that I don't understand his position. And that is a fundamental failure of Pai and his team. A failure to properly explain to the American internet-going public the benefit of his position beyond "it isn't our job to police the Internet." Sorry to reduce his stance to a meme, and one that might not even accurately encapsulate his positions, but I'm a reasonably well informed individual and all I can see is higher prices for Internet ahead.
> What evidence do we have that the US Government will do this?
They tend to be quite reactionary (to our detriment in many cases). I don't necessarily agree with Pai, but many of his ilk can point to preemptive restrictive regulations as bad.
If there is a side effect, it might just be that those clamoring for unrestricted internet access force their municipalities (and states) to publicly commoditize ISPs. But yes, you are more likely to see a "internet basic package" come through at a lower price at first.
>If there is a side effect, it might just be that those clamoring for unrestricted internet access force their municipalities (and states) to publicly commoditize ISPs.
Force them? Here's an example of his ilk pre-empting that:
Thanks for the links. I haven't read through all of them, but my first impression is that Pai is underestimating the impact of structural changes in the ISP industry. He keeps saying he wants to go back to how things were in the 1990s, but I don't think that's possible. The degree of industry consolidation, and of regulatory capture by the largest ISPs, have changed the landscape, and the approach that worked well until maybe 2007 won't necessarily work anymore.
I do appreciate that he's at least trying to increase infrastructure investment, and I guess I'll have to think some more before I have an opinion on whether the proposed rules are likely to do that. I do know of one small ISP that supports his approach.
As others have said here in discussions like this one, I think that net neutrality is something of a second choice. What we really need is meaningful competition among ISPs. I would be happy if we could take all this political energy being spent opposing Pai and channel it instead toward removing barriers to municipal broadband.
We had net neutrality from the creation of the internet until 2002, from 2010 until 2014, and from 2015 until now.
DSL started out regulated under Title II as a common carrier as an extension of the regulation for the phone network. Common carrier status automatically applies a limited form of network neutrality as it doesn't allow arbitrary blocking or discrimination of access. In 2002 the FCC decided the relatively new cable internet access should be regulated under Title I and then moved DSL to Title I in 2005 to match.
Also in 2005, the FCC give its definition of the requirement for an "open internet". These essentially described network neutrality. In 2010 the FCC issued their Open Internet Order which made the requirements for an open internet mandatory for ISPs.
In 2014 the DC court of appeals struck down the most important requirements from the FCC order in Verizon v. FCC. In doing so the court informed the FCC these regulations could only be applied to services covered by Title II. In response in 2015 the FCC moved all ISP services under Title II, back where they had started.
That's extremely misleading. The FCC has been trying to take regulatory control over the internet for some time now, but their attempts have been rebuffed by the courts in literally every single case. So no, we have not had any sort of net neutrality.
Your comment doesn't make sense to me given the information in your sources. I don't agree with your characterization of the FCC as "trying to take regulatory control over the internet" (implying that it doesn't have that authority) or your statement that "...their attempts have been rebuffed by the courts in literally every single case."
In your first source the case United States Telecom Association, Et al., v. Federal Communications Commission is mentioned, where the ruling was made in _favor_ of the FCC by a 2-1 margin. Additionally, the decision you cite in your third source was a loss for the FCC, but clearly their authority to regulate the internet was not in question, to whit, "The court upheld the FCC's authority to regulate broadband providers to encourage nationwide broadband deployment, agreeing with the FCC's interpretation of section 706 of the 1996 Telecommunications Act."
Check the dates. The article discusses an issue from 2005. The ruling the "subsequent actions" sections mentions is from 2015. And is why, in my original comment, I stated that we did indeed have net neutrality from 2015 onward.
We have always had net neutrality, it just hasn't been required. We need rules now because the incentive to violate it has recently become much stronger.
> In general, I think if it's the case that you can't reasonably defend a view you don't agree with then there's a very good chance you don't understand it.
While I appreciate this sentiment, it really only applies to arguments that are being made in good faith.
Pai may be saying "rules should be written as necessary instead of creating an enormous rule book based entirely on today's tech.", which is a supportable position that I can reasonably defend despite not agreeing with.
However, the action they are taking includeds removing any legal ability for the FCC to write and enforce these rules. Now congress can create these types rules without the FCC having title II authority, but relying on congress to write the rules as needed in response in innovation is going to be MUCH SLOWER to respond to innovation and bad behavior than giving the FCC the authority to adjust and enforce rules as necessary.
> What he actually says, and the implications of the headlines and comments snipped by third party organizations tend to be quite different indeed!
This is because what they say they are doing (and why) is quite different from the actual actions and motivations.
Pai keeps saying "we've got to remove title II to help small ISPs", yet the is the big ISPs that are constantly pushing for it's removal, Title II was put in place to regulate large monopolies just like them (Ma Bell), and many small ISPs actually favor maintaining Title II status. It is hard to take someone's arguments in good faith when they consistently misrepresent the issues.
> If the throughput for these systems starts to become an issue would it be so unreasonable for them to throttle or block sites consuming disproportionate resources?
Yes it is unnecessary. If they can't scale those systems to match supply to demand, they can increase the amount charged to limit that demand to match supply. Giving them
> Maybe the most basic argument is that it should be clear that net neutrality has not really achieved anything. Many people seem to think we always had net neutrality and stripping it away is some radical shift.
We have had network neutrality (mostly anyways) for a long time and proponents argue that net neutrality (and it's resulting competition) is what has caused the internet to innovate as such an amazingly rapid pace. That neutrality was not enshrined by law in the US until 2015 but that doesn't mean that the principle itself wasn't being followed by convention.
I hope your misleading arguments were made in good faith rather that being a deliberate attempt to mislead people like so much of the anti net neutrality rhetoric. (There is misleading pro net neutrality rhetoric too but I see more proponents making good faith arguments)
While certainly missing plenty of nuance, his view is that rules should be written as necessary instead of creating an enormous rule book based entirely on today's tech. And to some degree, I think this has a lot of merit. Take for instance SpaceX and Google's variations of widespread air-based internet. If the throughput for these systems starts to become an issue would it be so unreasonable for them to throttle or block sites consuming disproportionate resources? And yes, this could absolutely be used in an anti-competitive fashion. What Pai is saying is deal with specific bad behaviors as they arise instead of building an encyclopedia of rapidly dated laws against many actions - which in some cases could be completely justifiable and beneficial.
Maybe the most basic argument is that it should be clear that net neutrality has not really achieved anything. Many people seem to think we always had net neutrality and stripping it away is some radical shift. In reality the government had no formal regulatory role on the internet until 2015. What was the state of the internet beforehand? Has it worsened or improved since then?
[1] - https://www.cnet.com/news/fcc-chair-dishes-on-plan-to-rewrit...
[2] - https://www.recode.net/2017/5/5/15560150/transcript-fcc-chai...
[3] - https://www.npr.org/sections/alltechconsidered/2017/05/05/52...