FCC forced by court to ask the public again for feedback on net neutrality
theregister.co.uk
theregister.co.uk
The deadline for comments is March 30, and the deadline for reply comments is April 29, (though it's not clear what the difference between comments and reply comments is).
The FCC document (in all its legalistic glory) can be seen here: [1]
And you can comment here: [2]
[1] - https://docs.fcc.gov/public/attachments/DA-20-168A1.pdf
Nah, they wouldn't do anything like that...
Just as the commission has always held it was the responsibility and duty of all users of a particular part of the spectrum from intentionally or unintentionally interfering with any other legitimate users, the commission should make rulings that require that providers of Internet service (ISPs) shall not intentionally or unintentionally interfere with any legitimate use of the network bandwidth between the end user and the service provider communicating with that end user. Such policies of equal access for all legitimate uses has been a bedrock of FCC policy since the commission was established.
For people trying to fill out the form, use 17-108 as the proceeding you are commenting on.
It is completely inappropriate for Internet Service Providers to apply different traffic policies to legitimate use cases. All legitimate use cases of the service by the end user shall be given equal access rights. Restore the Title II designation for ISPs, or face the risk of a cunning and dedicated demographic working toward the destruction of the institutions that give you your power.
Did you elaborate on this at all, or do you intend to make the FCC (and us) guess?
Keep in mind, what you post to FCC is public with your contact info. Posting it here verbatim creates a linkage between your persona here and IRL. You may not mind, but it's something to consider.
You forgot the most important part of the comment that the FCC really wants to hear from the public : why?
I see governance as having two primary missions, preventing harms and preserving rights. That said, rulemaking is a policy action, it creates rules much like a programmer does, that are provided input and act based on that input.
An example of the latter that many readers here are familiar with would be sanitizing user input prior to displaying it on a web page. The programmer creates a rule like "always change the '<' and '>' characters to < and >." That is a rule and the 'why' is implicit. Without making that change, a user might type in their input an arbitrary piece of javascript which when displayed becomes part of the page. So the 'rule' prevents the 'harm' while preserving the 'right' for users to put < and > in their input, like we do here in comments all the time.
The other aspect of policies (or rules) is that making them consistent allows users or clients to not be surprised when something they think should be allowed, isn't. This is sometimes shortened to the principle of least surprise.
In this case the FCC is rule making in what is, relative to there original mission around RF spectrum, a new communication medium, the Internet.
To someone at the FCC, within the context of their past policy/rule making and their understanding of the Internet as a communication mechanism, they have the tools to understand that ruling against interference to other legitimate users of the spectrum is a core principle of their rule making in the RF world. This is particularly true in the portions of the spectrum that are unlicensed, like the 2.4GHz ISM bands.
My comment was written for a commissioner to read and to tie the 'old' with the 'new' and to perhaps dent a little bit the argument that ISPs use of "my wires, my house." Such a commissioner will already have heard hundreds, if not thousands, or arguments which stress the harm that is caused when one entity using spectrum and interfere with another using the same spectrum.
Using our WiFi examples it would be trivial for Cisco access points to recognize Ubiquity access points nearby and jam their channels. That would give Cisco users a "good" experience, and Ubiquity users a "bad" experience. And no doubt there have been examples of access points that had this sort of behavior, intentionally or not, and have been brought before the commission and adjudicated as being interferers and causing harm.
My hope is that one or more commissioners will have that context and read my comment and think, "Hmm, he has a point, it is just like spectrum and we treat people who interfere there as the 'bad guys', why should it be ok here?" And with that thought will come many, many, "Why equal access is the correct policy choice" examples from their work in the management of RF spectrum.
Shell companies were created, those companies would buy "leads" from ad-tech companies. Those "leads" were names/emails of people who checked a box saying "I'm okay if you use my name and email to comment on the net neutrality issue." There is no real way to tell where these leads come from. Theoretically the ad-tech companies will place ads on publisher websites which contain a checkbox allowing a user to "opt-in". Once the leads make it back to the shell companies, they can add those leads to the comments section because the person willingly (in theory) gave them their info for this exact purpose. In reality the ad-tech industry is rife with grey-hat SEO techniques at best, and outright fraud at worst.
They are some of the biggest proponents of it because they don't want ISPs trying to charge them for transit to their users.
I'd like to see one of these checkboxes though.
The fact that Comcast owns that cable means there's no real competition.
- Last mile sections of the city will always have winner takes most (customers) situation, making the lines built by the other providers redundant and wasteful
- There are only so many communication lines you can put on poles, and underground, before the people start complaining about ugliness, and never ending construction.
I think a good approach for local communication infrastructure is ownership by community/local-goverment and private contract bidding on construction, maintanance and support
- Companies bid on building new infrastructure,
- Companies bid on running & maintaining existing infrastructure (usually a 2-3 year bid contract that sets fixed rate for all customers for 2-3 years). At the end of each contract a new bid goes out, but the same company can "win" the bid
- People in the community vote on the bids. The POLITICIANS don't get to pick the winning bid
However, I think another overall approach would be just as effective at achieving pro-consumer ISP behavior. The approach is simply this: via regulation, force local loop unbundling, and take no other action. Suddenly, anyone would be allowed to start an ISP and rent the existing infrastructure. This would reintroduce the free market and allow it to solve the underlying behavior problems. I think that when it came time to vote with their wallet, most people would select a pro-neutrality ISP, regardless of their political alignment. Competition generally works fine in places it's actually present.
Incumbent ISPs would complain and scream, possibly louder than they have against Title II. I don't know what the legal approach would be for forcing unbundling, but I seem to recall it wasn't completely infeasible.
Local infrastructure will always be a winner takes all economic game. So its pointless to play it. (side-note: Elon Musk's Starlink might change the economic game)
The only way for people to "vote with their wallets" on local communication infrastructure, is BEFORE the winner "settles in", not AFTER.
A bid system, allows 100's of companies to compete to set the price for 2-3 years. Rather than 2-3 companies competing to set the price for perpetuity. It also creates an incentive to produce quality service, because they will be competing for another contract in 2 to 3 years (i.e. they have 2-3 years to demonstrate they are a competent provider).
A single pipe can handle only so many customers. I don't see how it limits the number of providers?
There actually are countriess with such rules, and they seem to work well. The difficult part is the need to set some uniform price providers must pay for that "last mile" connection to their customer. I seem to remember something like $8/month in Germany. That's actually low enough, it would allow healthy competition even if you set wholesale price 50% higher than neccessary.
The same mechanism is used for competition among power and natural gas companies.
The problem is that they can't compete on price (in the downwards direction), and they can't compete on building new infra with latest technology. The minimum price is established by the price the owner of the pipes charges. Which is usually set by the gov't. So the can only compete on value added services (i.e. hey we are not going to sell your data hooray!)
The core competition we want in local infrastructure is faster internet and cheaper prices. I think the dual bid system can accomplish that (i.e. separate bids on building/upgrading infra, and bids on maintenance/support)
As for gas and electricity, I think a similar bid system would work as well.
It could also make the situation worse, if everyone changes to starlink and Starlink becomes a national or global monopoly ISP.
If this were true... why would the local telecom companies fight so hard to make sure it's illegal to build competing infrastructure?
I think the flaw in your reasoning is the "genetic" logical fallacy https://yourlogicalfallacyis.com/genetic
Just because someone "bad" doesn't want something, doesn't make that thing automatically "good"
Maybe not all wireless technologies ??? I am not an expert on wireless.
Still wired internet has advantages over the other technologies (stability and latency), so I am merely suggesting an effective way to compete on wired internet can help significantly (its path of least resistance from a technology standpoint, but not from a political standpoint)
The typical scenario is Comcast asking "startup X" for money to allow them to reach you.
You wouldn't have to pay more, nor would you necessarily notice: established sites would be exempt because customers expect them to work. But nobody will be missing a new company they have never heard of.
Since you will not (immediately) feel the impact, there will be no incentive for you to chose a different ISP.
Conversely, that startup doesn't get a choice: your ISP is the only route to get to you, and they have to pay or forfeit the chance to do business with you.
ISPs would be in a position to claim the vast majority of any internet-based business' profits. As a data-intensive startup, you would also be faced with the prospect of negotiating contracts with every single significant ISP. You'd get to make decisions such as "should we fire ten people, or go dark in Florida for the rest of the month?"
Have any of these things happened? Or did that turn out to mostly be fear-mongering?
If you're hosting something, might be worth upgrading to a business connection, which doesn't have data caps and comes with static ip's and a SLA.
As an aside - I'm personally OK with data caps so long as they don't impact most users (80/20, 90/10 or something like that).
ISP's pay for ingress traffic, so by setting a data cap that impacts only the egregious data hogs, it means they can charge normal users a smaller rate, rather than subsidize a few users by charging everyone more.
Also, 3TB is not huge for a family, especially in the age of 4K streams.
Either way, I don't want to subsidize the guy torrenting Blue-Rays and pushing 10TB a month when I'm only using 1TB.
It wouldn't though - at least for most major residential ISP's. They oversubscribe the connections, knowing full well not everybody will use the connection at the same time, and even when they are - they won't be using the maximum at once.
A minor example would be bringing a single 100Mbps connection into a small building and selling 12 x 10Mbps connections. It's one of they ways they can, yes, make a profit and also offer the bandwidth at reasonable costs.
Dedicated bandwidth is expensive. Think of your old-school T1's and your typical business class connection. You pay a lot for having that bandwidth reserved for your use alone. And majority of the time you aren't using it... and when you do use it, you use very little of it at once. Debating on HN doesn't take 1Gbps... so it goes unused. Residential users prefer lower costs vs. dedicated bandwidth.
> The data caps exist solely so they can squeeze people for money beyond what they actually need to maintain the network
Yes and no. You seem to be tip-toeing around the municipality-run ISP argument - which is a fair argument to have (although I'm concerned a city can maintain this network long-term and keep it as good as a for-profit company). There's no problem with an ISP charging customers more for the connection than the ISP pays to provide it. That's the incentive to run the network in the first place - let alone spend millions building it out.
I don't think data caps are bad so long as they don't impact "normal" use. Normal as being defined perhaps by the 95th percentile of use or something. The remaining users should be obligated to pay for their usage so that the ISP isn't forced to spread that costs onto other customers - that's simply not fair.
Given the recent emergence of high-definition video streamed to every home, it seems like three times that probably wouldn't be enough.
EDIT: Capitalization error, mbps > MBps.
Ya, but you didn't do that. 1TB pushed at a constant 1Mbps, would take 88 days... 4TB taking nearly an entire year.
4TB at a constant 100Mbps takes 3.5 days. It is a HUGE amount of data.
> recent emergence of high-definition video streamed to every home
Netflix claims about 3GB per hour of HD content streamed[1]. So if you were to stream 24 hours a day for an entire 30 day month, you'd barely peak over 2TB. That's 24 hours a day for 30 days straight - something nobody is doing.
A data cap of 3TB is simply not a problem for almost everybody. If you're running torrents, or hosting files, then ya - you might have issues. But then why should I pay comcast more every month to subsidize your non-normal usage?
Ah, I see. mbps v. MBps. Capitalization error!
Also, you're arguing against the past point which is fine because I messed up the capitalization (though the point still holds, because I wasn't doing anything abnormal; it was a two person household with a game console), but not against present use-cases:
4k 24fps video for Netflix is 7GB, and 1080p 24fps video is over 3GB/hr (HD is 720p). How many TVs are you aware of that are under 1080p? A house that has two people regularly watching Netflix in it on a modern data connection seems like it'd blow through that cap pretty quickly.
Why should Comcast be able to sell you a data connection you can't actually use? Caps are only a thing because Comcast oversells every connection.
It's not impossible, but it certainly isn't blowing through the cap quickly.
The caps are a thing because Comcast pays for ingress traffic into their network (they pay other providers from which the data comes from). At a point, you cost them more than you pay for. Which is where they charge you an overage.
Seems more than fair to me - I pay currently $49.99 a month for 150Mbps connection with a 3TB cap (I'm aware they offer different rates, speeds, and caps in different areas).
I do not want to pay more than $49.99 just because someone else in my area constantly uses 10TB each month - that person should pay more, not me.
Actually, in some cases, Comcast gets paid FOR it.
https://qz.com/256586/the-inside-story-of-how-netflix-came-t...
They started charging Netflix (or threatened to deprioritize or drop Netflix traffic) because Netflix traffic to Comcast customers starting making Comcast customers unprofitable.
Comcast does pay for ingress traffic. They are not a Tier 1 provider and do not peer with other Tier 1 providers - bits flowing in cost money. They either had to charge Comcast customers more per month, or get Netflix to subsidize the ingress traffic costs from their service.
This ultimately led to Netflix making the Open Connect boxes and offering them for free to ISP's as a way to mitigate the ingress costs. (this is why ISP provided video services don't have this problem - their traffic originates in-network and doesn't cost them anything additional to provide, fair or not)
Unless we're going to change how the 3 Tier provider system works and who gets to charge who for ingress traffic - I don't see this changing anytime soon. Unless you want to regulate our way into doubling your monthly Comcast bill... I sure don't.
what is should be forbidden would be to advertise an unlimited plan and then post-facto soft ban users that actually test the unlimited description, or deprioritize connection to expensive services (like YouTube, Netflix or torrents)
I was using steadily 600-700GB of data a month on my gigabit plan. It suddenly spiked to 2TB - from what I could tell there was no change in behavior, my ubiquiti gear wasn't showing this amount of usage. This went on and used up my two 'courtesy' months of overages forcing me to pay the extra money to remove the cap.
A month after paying this my usage amounts dropped back down to where they were before.
The gigabit services works well, but I find the cap hard to accept without any ability to question it.
Fair enough - but also in fairness, your electric utility provides about as much (which is also frustrating when you're trying to figure out why your bill went up 25% one month).
I personally run a mikrotik router which tracks my usage too, so I have a reference to double check against and could wage a good argument with evidence that I didn't use the data they claim.
Not everyone can do that though... but perhaps standard routers should come with these features now-days.
It’s not worth my time to argue with them and there are no other gigabit options.
Let us know how that works out for you. My guess?
"We can waive this month as a courtesy, but we can't use your router as evidence".
That would be the goal.
So do my water and gas companies.
My ISP provides the total I used in a month, at the end of the month. That's it.
- Netflix in 4K (7 GB/hour), 2 hours per weekday on 2 screens. That's 560 GB in a 4-week month.
- Google Stadia in 4K (35 GB/hour), 2 hours per week on 2 screens. That's 560 GB in a 4-week month.
Bam, 1,120 GB/month.
What makes you think that? Most everyone will peer with residential ISPs for free. There have even been issues with residential ISPs trying to get paid for ingress traffic, because they control the only network path to their customers which means they can charge monopoly prices to video competitors like Netflix who want to send traffic to their users on that ISP.
> so by setting a data cap that impacts only the egregious data hogs, it means they can charge normal users a smaller rate, rather than subsidize a few users by charging everyone more.
An ISP's costs are dominated by things like staff and trucks and electricity that depend little if at all on how much data someone transfers. None of the other customers are going to get a discount from heavy users using less because there is no meaningful savings there.
The reason for data caps is some combination of anticompetitive practices and price discrimination.
If you tell people they have a data cap, they'll try to minimize usage. Which is to say minimize Netflix, which counts toward the data cap -- but hey look, the ISP's own video service doesn't.
They also assume that if you use a lot of data, you derive a lot of value from internet service, so even if the data hardly costs them anything, the fact that it's valuable to you makes them want you to have to pay more for a business connection.
Both of these are stupid reasons which they only get away with because of the lack of competition.
Residential ISP's are typically Tier 3 or maybe Tier 2 providers, and have to get their bandwidth from either a Tier 2 or Tier 1 provider. These providers definitely charge for traffic to pass into the ISP's network (ingress traffic into an ISP network, egress out of the Tier 1 provider). Without paying that bill, ISP's couldn't exist... and without charging that bill, Tier 1 providers couldn't exist.
Tier 1's peer with other Tier 1's for free because it benefits them and allows them to have connections to other networks that Tier 2 and Tier 3 providers want access to. They make their money charging Tier 2's and 3's for that access.
> because they control the only network path to their customers
Tier 1 providers don't care about residential users. This argument doesn't make any sense.
Comcast got away with charging Netflix for ingress traffic because comcast was going to drop or throttle Netflix traffic. Why? Because suddenly the $50 a month comcast was charging wasn't enough to cover the ingress traffic generated by Netflix content into the ISP's network. The decision was to either charge Comcast customers more, or ask Netflix to subsidize that added expense. This also led Netflix to create the Open Connect boxes they offer for free to ISP's.
> None of the other customers are going to get a discount from heavy users using less
No, but they won't have their bill raised if the ISP can just charge the heavy users more instead.
> but hey look, the ISP's own video service doesn't
This is because the ISP's own video service's traffic originates in-network, therefore costs nothing additional to provide other than the existing infrastructure costs (there's no ingress traffic to pay for). Netflix (and other out-of-network services) costs more, as previously covered. Open Connect can help mitigate this, but then it's on the ISP to rack the server and pay all the other related costs for keeping it running in their data center... something they might not want to do.
Comcast's argument was Netflix was making a ton of profit from streaming content they weren't helping pay to deliver - effectively shifting the delivery cost to the ISP, which was getting stuck with suddenly a bunch of unprofitable residential customers. Asking Netflix to help pay for that ingress traffic burden doesn't seem so outlandish from a network-ops perspective. Either that, or Comcast could have raised everyone's rates... which people probably wouldn't have liked.
A parallel would be making an ecommerce business and expecting packages to be delivered to customers for free. The customer pays for their mailbox already (in taxes or directly with a PO Box), and paid for the package, so they expect it to get delivered. At some point, the delivery company is going to need to charge for the service, since there's a cost associated in making that delivery. Either the customer can pay for the delivery cost too, or the ecommerce company can pay for the delivery cost. Either way... somebody has to pay that cost.
> only get away with because of the lack of competition.
I don't disagree that we need more residential ISP competition in the US.
You're thinking of the small dial-up ISPs from 20 years ago. The likes of Comcast and Verizon are themselves Tier 1 providers. The other Tier 1 providers' customers are predominantly data centers that want transit to the rest of the internet and aren't themselves (as e.g. Google is) Tier 1 networks.
> Tier 1 providers don't care about residential users. This argument doesn't make any sense.
Tier 1 providers sell transit to the whole internet, so they have to care about having a path to all users or their other customers will be displeased about not being able to reach those networks. The only path to users on Comcast is through Comcast, and there are enough of them that people will actually notice if they're missing.
> Comcast got away with charging Netflix for ingress traffic because comcast was going to drop or throttle Netflix traffic.
This is my point. They have the ability to drop or throttle Netflix traffic and there is nothing Netflix can do about it other than to pay them the danegeld, because they control the only network path to those users.
> Why? Because suddenly the $50 a month comcast was charging wasn't enough to cover the ingress traffic generated by Netflix content into the ISP's network. The decision was to either charge Comcast customers more, or ask Netflix to subsidize that added expense.
They didn't have to pay anything for ingress traffic from Netflix because their transit provider at the time was willing to peer with Comcast for free.
> This also led Netflix to create the Open Connect boxes they offer for free to ISP's.
Netflix was willing to offer them for free but Comcast wanted to be paid to use them. How does that make any sense if Comcast's problem was legitimately the amount they had to pay to other networks for ingress traffic?
> No, but they won't have their bill raised if the ISP can just charge the heavy users more instead.
There isn't anything there to raise the bill for since the additional usage has insignificant cost, and there aren't enough heavy users for that amortized amount to be real money on everybody else's bill even if it didn't.
> Comcast's argument was Netflix was making a ton of profit from streaming content they weren't helping pay to deliver - effectively shifting the delivery cost to the ISP, which was getting stuck with suddenly a bunch of unprofitable residential customers.
Each endpoint's ISP is responsible for carrying traffic to a mutually agreed upon peering exchange where it's handed off to the other endpoint's ISP. Netflix will basically peer with anybody at any peering exchange for free. Carrying the traffic from there to the other endpoint is what the customer is paying Comcast for. Comcast was double dipping, trying to get paid twice for carrying the same traffic. Or get an unfair advantage over a competitor for video services.
> A parallel would be making an ecommerce business and expecting packages to be delivered to customers for free.
A parallel would be Amazon paying FedEx to deliver packages and then discovering that Walmart had bought 20 million apartment units with doormen and was refusing to accept Amazon packages to the residents of those apartment buildings unless they paid Walmart for the privilege.
I am a single person but I host my personal non-business website from home and participate in as many peer to peer distributed and federated protocols as I can. I also do watch a bit of video media. My usage isn't only bursty, it's a steady 100-200 KB/s 24/7 up with bursts down on top.
I contacted Comcast Business two times over the last 2 years, the most recent just last week. If you want a business 100/10 account then it is $156/mo with a 2 year contract (do not believe their promotional lies). If you move away to an area without service you have to pay 75% of the remaining contract. If you have your own modem they charge you a fee more than twice the fee for renting their modem.
Frankly, it's insulting.
They also put your house on a difference circuit from everyone else in your neighborhood, which has a cost - which explains the ETF if you move away.
In your case, perhaps it's cheaper to get a $2.50/month VPS from Vultr instead of paying all these fees? You get a static IP to, and don't have to play DynamicDNS games.
The rep I talked to and the price I gave $156/mo does not include a static IP. That was something like $10/mo more.
Sure, the firmware they tftp to my modem will be different but it's not giving me dedicated RF bandwidth on the exact same RF hardware path that existed before I got the firmware. It's the same uplink channels. This can be confirmed by using a splitter, attenuator, and looking at the RF signals directly in the frequency domain over time with an with a software defined radio (see: http://superkuh.com/pictures/docsis3_cable_ziggo_2017-10-08-...).
You can do a traceroute and see the reverse-dns names of the hops you go through too.
Also remember, TCP has 8Mb of overhead per 100Mb, so you'll get maximum theoretical speed of 92Mbps on a 100Mbps line.
So to reach that number, I guess you're talking about 60 bytes of headers for each 1500 byte data packet, then again for an ACK packet?
That's partly or entirely wrong in multiple ways.
First, the combined IP+TCP header is only that big for IPv6 connections.
Second, under general circumstances an ACK is only sent for every other packet, halving that bandwidth.
Third, that 100Mb is going to be for one direction, so the ACK traffic going the other way doesn't affect it at all.
Typical overhead is generally 3%, not 8%.
Are you claiming that T-mobile slowing specific sites or people? Do you have a source for that?
I agree. But that wasn't my take on the claim above, so I asked for the source that this was happening selectively to specific sites, apps, or users. I guess I'll keep waiting.
;@)
Which highlights the problem with what Net Neutrality morphed into - it became something that meant everything and anything - not just "dumb pipes" like it originated as.
Ask 3 different people what Net Neutrality means and you'll get 3 different answers. How do you legislate that and make everyone happy?
This highlights the foolishness of "net neutrality", which is really just code for "my team should regulate the internet".
[0] I guess I am wrong in believing this should is respected
This is conditional on resources or on the possibility of a class action. In practice I agree that this condition is almost never satisfied.
I mean a VPN could legally do that, and in the case of a ISP my personal idea of net-neutrality would be that a neutral service needs to be available and that non-neutral offerings needs to be clearly marketed as such.
I honestly see this as a possible drive to competition, as ISP would be forced to disclose how shitty or not the service is.
0. Even assuming lossless/infinite backhaul capacity, it gets messy when guessing about what's deployed in terms of packet prioritization of tower -> handset.
1. Once deployed and maintained, the marginal cost of data is much closer to zero than what customers are being overcharged. Even if it's going from a cell tower to cheap dark fiber or trenched in, it's still nearly the same amortized cost. Also, there is plenty of spectrum channel capacity.. the "bottleneck" is the backend of the towers or their backend upstream in the region.
2. Xfinity (optical, fixed-premises "cable") has a 1 TB limit for certain US states but not others, where they have unlimited data allowances. So it's not about infrastructure or data transit costs, it's about making money from relatively richer people.
It would be a good requirement for internet-providing carriers to be required to publish their effective channel capacity oversubscription ratio (for upload & download) for each internet-connecting plan that is offered in each ZIP code.
When everyone thinks of an Internet plan as what TV shows and streaming media they get, with FAANG and remote work access on the side, why should ISPs do anything but allow access to whitelisted services and companies that pay them for access?
It's not going to be a quick process, but since telecom providers own the lines, and people have no choice but to give them money, they aren't going anywhere and can wait it out.
When you think to yourself, why do I put up with youtube, why isn't there an alternative... that's the 'thing happening'.
(Side note: one more good reason for Netflix to be in the content creation game)
Netflix gets to pay whatever rate to get priority infrastructure at the telco, you get to pay more to Netflix, and you blame Netflix (not the telco) if the content delivery is botched / not as good as what your "cable subscription over the Internet" delivery is.
That doesn't line up perfectly with the narrative about fast lanes that we were sold, but it does show some of the power being wielded by the telco today as more than just a dumb pipe, and how they leverage their position against competitors.
I'm not sure that's true. Netflix must pay a premium to the ISP to get prioritized access within the ISP in order to get their edge cache in position to compete against the ISP. That's three ways in which the ISP is not acting as a dumb pipe.
The Wikipedia understanding of network neutrality is: the principle that Internet service providers (ISPs) must treat all Internet communications equally, and not discriminate or charge differently based on user, content, website, platform, application, type of equipment, source address, destination address, or method of communication.
No, that’s not a fast lane nor would be stopped by NN. You just explained that Netflix made an investment to get CDNs where they would be effective.
So, I’m still curious to know where these priority traffic fast lanes are.
That costs the Provider significant routing and intermediate-distance (40~100km, i.e. CWDM long-range transceivers) bandwidth. This is much more expensive then feeding the google box with power.
An "internet fast lane" would be like the city government saying to Amazon, "and also, all traffic lights will immediately switch to green for amazon delivery vehicles."
No, they're still using the same roads as everyone else, they're just starting from a lot closer to the destination.
For example Verizon has the lesser known FreeBee: For pay-per-click campaigns, the FreeBee Data icon (a “bee”) appears next to sponsored content to let Verizon Wireless subscribers know that, when they click on that content, they will not incur data charges. With FreeBee Data 360, businesses can sponsor some or all of their app or website.
That’s just one example, but it does demonstrate net neutrality is dead in the US.
Net neutrality and the move to "Internet everything" is an apocalyptic combination for ISPs, who are accustomed to selling overpriced cable bundles. If all your entertainment is on Netflix, they can't charge you $100/month for a TV package. With net neutrality gone, they can move that tax to Netflix instead.
The game is older than that, and iterated.
The telcos-vs-internet fight over common carriage, network neutrality, and then Net Neutrality, has been going on since the 1990's. Telcos don't want to be in a commodity business of dumb pipes - far more profitable to coerce a piece of everyone's action. Educating congress about the architectural importance of common carriage burned a lot of IETF leadership time during the transition to Internet.
One ongoing characteristic of this fight has been telcos moderating their behavior for fear of backlash strengthening the neutrality position. Another recurrent pattern has been telco overreach, followed by regretting it. Gouge too hard now, and reestablishing neutrality regulation in some next administration becomes a no brainer. Over the decades, neutrality regulation has lost more than won, but has remained an immanent threat. Slowing the re-imagining of the internet as a centralized broadcast medium.
The internet is a critical part of both military and civilian infrastructure and should not be monitored, limited, or in any way restricted by any entity. Under the definition of arms understood at the time of the Constitution (as written by James Madison in Federalist 46), the internet and any digital tools which may be used in the defense of the nation are "arms" and therefore protected by the second amendment.
Further, because the internet is ubiquitous in modern society as a means of communication, much more so than letters were at the time of the constitution, the government has no right whatsoever to restrict access or use of the internet. Even if there were a disagreement over this interpretation, there can be no doubt that the general public 'assembles' and 'petitions' using the internet. This petition itself is being written on the internet. All this ignores the simple truth that the internet is the most common means of distribution for the press worldwide.
Finally, because the federal government has singularly supported, with enormous sums of taxpayer dollars, the internet infrastructure which is now (against the interest of the public, for which the federal government is a trustee) owned by private corporations, all internet infrastructure should be treated as public property, and ISP's should act only in a capacity of providing a utility service.
In summary - there should be no privatization of internet infrastructure. There should be no government or corporation monitoring anyone's use of the internet (without a proper warrant, signed by a judge, on a case-by-case basis). A person's digital information should be treated as 'papers' and 'effects' as they are considered to be by the general public and by most experts in the subject. And lastly, there should be absolutely zero limits, restrictions, rationing, throttling, or other mitigation of any person's access to a completely free and open internet.
I'm not sure about this. What about shovels, nuclear bombs, or meth [0]? Do we have a right to bear nukes?
[0] https://en.m.wikipedia.org/wiki/History_and_culture_of_subst...
Amphetamines or other performance-enhancing drugs are absolutely useful for military applications. As much as I loathe drugs and addiction I don't believe I have the right to restrict what another human being can or cannot put into their body.
> My personal stance on the 'right to bear nukes' question is that a weapon of mass destruction has no defensive utility and is either self-destructive or offensive by nature.
Serious question: How is that different from munition and/or firearms?
As far as I know, if you exclude weapons whose utility is only self-destructive or offensive, you'll only have (active/passive) shielding left, which precise munitions and firearms to the best of my knowledge are not. Yes, they can be used as a threat, but the same can be said for nukes.
If someone breaks into your home, you can reliably combat the intruder with a firearm - without harming a bystander.
If someone is attacking your town or village, you can at least situationally combat them with precision-guided munitions without killing any civilians (not that they are used ethically in all cases, just speaking to their potential use).
Neither of those things is true of WMDs. You have to stretch the bounds of realism pretty far before you can find a way to deploy a nuke effectively without killing innocent people, or yourself.
http://freakonomics.com/podcast/ajit-pai/
As for the article, I don't think it matters all that much except maybe possible slight embarassment?
I don't really want to listen to that guy babel pro-ISP nonsense.
* Creating competition is the only surefire path for improvement
* State/Local governments should not subsidize players to avoid entrenching ISPs and creating local monopolies.
* In areas with entrenched ISPs, the only way to break the cycle is to bring in new players (Google Fiber successfully proved this)
The FTC should regulate ISPs, not the FCC:
* An "internet service" provides internet access unfiltered subject to a well defined classes. Anything that is not an ISP is an "information service" provides a narrow walled garden and must be advertised as such.
* ISPs should be regularly investigated for false claims on advertisement.
[0] https://arstechnica.com/tech-policy/2017/07/saving-net-neutr...
Why do they need my address at all? And worse, why do they need to make the email and physical address public?
I can prioritize the packets I send to my ISP, but I have no control over the order I get packets from them.
You can either do traffic pattern analysis or just base it on what IPs they are connecting to.
previous Net Neutrality text:
> § 8.5 No blocking. A person engaged in the provision of broadband Internet access service, insofar as such person is so engaged, shall not block lawful content, applications, services, or non-harmful devices, subject to reasonable network management.
> § 8.7 No throttling. A person engaged in the provision of broadband Internet access service, insofar as such person is so engaged, shall not impair or degrade lawful Internet traffic on the basis of Internet content, application, or service, or use of a non-harmful device, subject to reasonable network management.
-
this is putting too much faith in the government (+corporations) and enables the lawful establishment of the Great American Firewall to manage what's "lawful"
What is the expected gain from Net Neutrality at this point?
NN is when you can't get to "Social Media Startup X" because they didn't pay your ISP.
You are unlikely to even notice this, because only newcomers would be required to pay ransom. Blocking existing sites would lead to customer complaints and actual competition.
That sounds bad! Any examples of that ever happening before or after NN?
https://www.eff.org/deeplinks/2008/08/fcc-rules-against-comc....
So, in no way supportive of the NN argument.