Real Net Neutrality Problem: 'Edge Provider' vs 'End User' 97
An anonymous reader writes At the Washington Post, Brett Frischmann elaborates on the theory that the continuing flaw with the FCC's Net-Neutrality strategy lies in the perverse distinction between "End User" and "Edge Provider". Succinctly: "The key to an open Internet is nondiscrimination and in particular, a prohibition on discrimination or prioritization based on the identity of the user (sender/receiver) or use (application/content)," and then, "Who exactly are the end users that are not edge providers? In other words, who uses the Internet but does not provide any content, application, or service? The answer is no one. All end users provide content as they engage in communications with other end users, individually or collectively. ...
Think of all the startups and small businesses run from people's homes on home Internet connections, using WordPress tools or Amazon hosting services. Are they 'end users' when they email their friends but 'edge providers' when they switch windows to check their business metrics?"
Re: (Score:1)
Which way are the bits going? (Score:3)
It sounds like they're intending to draw a distinction between nodes that principally receive data from those that principally transmit data.
If the node has a high ratio of bits received to bits transmitted, it's an "End User." If it has a high ratio of bits transmitted to bits received, it's an "edge provider."
ISPs are neither. They presumably have similar numbers of transmitted and received bits because they are mostly actiing as conduits between data sources and data sinks.
Re: (Score:1, Interesting)
It sounds like they're intending to draw a distinction between nodes that principally receive data from those that principally transmit data.
If the node has a high ratio of bits received to bits transmitted, it's an "End User." If it has a high ratio of bits transmitted to bits received, it's an "edge provider."
ISPs are neither. They presumably have similar numbers of transmitted and received bits because they are mostly actiing as conduits between data sources and data sinks.
I provide a service where users upload files and my system identifies similarities to other files that have been previously uploaded. I have a received to transmitted ratio of around 50-1. By the logic you describe this makes me an end user rather than a service provider, which is utter bollocks.
Re:Which way are the bits going? (Score:5, Insightful)
It sounds like they're intending to draw a distinction between nodes that principally receive data from those that principally transmit data.
If the node has a high ratio of bits received to bits transmitted, it's an "End User." If it has a high ratio of bits transmitted to bits received, it's an "edge provider."
It's like that because of the artificial restrictions placed on upload speeds by the DOCSIS and ADSL protocols. Which is just the big boys trying to protect their business model by keeping us from being creators and sharing on a peer-to-peer basis.
High speed, symmetrical network links for everyone, and peer-to-peer protocols for social networking, sharing creative content and ensuring freedom of speech could be incredibly empowering and liberating technologies.
Unfortunately, those technologies which would allow users to share directly with each other, as well as strong encryption would most certainly limit the ability of the corporatocracy and the governement to spy on us for their benefit. So of course it must be stopped.
I really hate how cynical I'm getting, but our corporate and government overlords keep taking our freedoms and most people are cheering them on. Good consumers. No need to be a citizen. Just be a good little consumer.
Re: (Score:2)
It ought to be, nodes are nodes, but we're talking about the difference between telco legacy interconnect and the dawn of Internet "hotels" which were aggregations near NAP points and convenient telco interconnects. This is what was the problem: ATM, SONET, and other L1/L2 problems. This allowed the concept that some junction points were more important than others, and that an edge device could be poorly provisioned, while big junction points could have nearby CDNs, huge hosts, and so forth.
Add in isometric
Re: (Score:2)
Who is going to pay for all of that fiber - and associated changes to the network to allow it to go the last mile (so far, the only fiber we've seen to the home is in very small enclaves of people who can afford premium services anyway)?
If you believe that should reside in the corporate realm, then how do you as a corporation turn a profit while also investing in a universal fiber network?
If you believe it should be in the government realm - how do you get politicians to support fund allocations for it -
Re: (Score:2)
Most of it, to the last mile or so, is in the ground in the US.... there's tons of dark fiber waiting to be lit up.
Fuck corporations turning a profit. This is a utility, not a bunch of regional monopolies masquerading as public beneficiaries. Governments and PEOPLE get easement and right-of-way income. There are lots of models as to how this can be done.
Pioneers like Loma Linda, CA, DIgital Cities, and others show how to make it work financially, and no, not some sort of neo-socialist/commie model.
Should th
Re: (Score:2)
Step one: propose bonds that would pay for last mile fiber networks, with rollout decided primarily by neighborhood population density.
Step two: ?
Step three: Money goes to municipalities which in turn put network installation and management projects up for bid. Municipalities choose to provide service as a utillity or allow private ISPs to compete to sell broadband services.
Re: (Score:1)
they got all of the infrastructure FOR FREE ALREADY by federal subsidies when they first started offering data/internet access.
then they used that infrastructure to make billions over the last 15 years.. now they don't want to upgrade the last mile, and the only real reason is to keep artifically low connection rates.
TFB. nationalize our data highways. they can go take a flying F the same way they have treated their "customers" for years.
P2P infringement (Score:2)
peer-to-peer protocols for social networking, sharing creative content
But do most home users, especially those who aren't paid for producing "creative content", have the legal right to share most of the "creative content" stored on their devices?
Re:P2P infringement (Score:5, Insightful)
peer-to-peer protocols for social networking, sharing creative content
But do most home users, especially those who aren't paid for producing "creative content", have the legal right to share most of the "creative content" stored on their devices?
Who knows what folks would do with the opportunities provided by high-speed symmetric links? Individually made feature films distributed by the filmmakers. Bands distributing their music directly. Authors selling their books. P2P social networks that are secured (in that you maintain control over your data on your own systems) and include only those you choose. And on and on and on.
The lack of symmetric links props up the business models created in the era of mass marketing. It could be quite disruptive to the content providers who dominate the ISP market (and generally only offer asymmetric links to their customers) as well. Is that a coincidence? I don't think so.
As for using such network links to distribute the intellectual property of others, the current model doesn't really stop that anyway does it?
The only folks that asymmetric (with hobbled upload) links benefits are those who profit from controlling the distribution of creative content, those who profit by creating centralized servers for social networks, product sales and other human feedlots so the information stored on their servers can be analyzed and PI based on viewing, browsing and buying habits as well as personal interests can be sold to marketers. And since most of those don't use any sort of encryption, the government gets to spy on you too.
So. Do you want freedom to communicate, collaborate and share only with those you choose to do so with, or do you want the big corporations and the government six inches up your ass? Yes, that's a straw man. But through it, I hope you get my point.
Re: (Score:2)
Who knows what folks would do with the opportunities provided by high-speed symmetric links? Individually made feature films distributed by the filmmakers.
How would the filmmakers recoup their investment in the film? Most commonly, people have suggested the use of copyright, a measure that legislators have enacted to let the author of a work have a chance of seeing rev
Re: (Score:3)
I agree that symmetric connections to the Internet could help open up public participation in culture. But lack of a symmetric connection isn't the only obstacle to self-publishing. If we solve the other obstacles first, people can be ready to go once the symmetric Internet connections are ready
There are obstacles to just about everything. Don't make the perfect the enemy of the good. Let's move forward and deal with whatever issues may arise as they do so.
I'm not a lawyer or a talent agent. I see wealthy business interests stifling freedom, creativity and innovation. As a networking guy I see solutions in the technology.
Are there issues? Yes. Are those issues any different without high-speed, symmetric internet connections? No.
I say implement the technologies and the rest will sort itself
Re: (Score:2)
You may not be aware of it, but you have technically strayed off-topic a bit. Or at least there is need for clarification. There are two quite different kinds of symmetry being discussed here. What the article was talking about was not symmetric bandwidth links (e.g. 10Mbs upstream and 10Mbs downstream), but rather the symmetry in being allowed to use your data transfer (upstream and downstream) to act as a webserver, instead of a webclient. If the FCC version of Net Neutrality was only updated to reflect the latter, then I believe the market would sort out the former. In all likelyhood it doesn't make sense to architect the last mile presuming symmetry which we know doesn't exist at this point in the evolution of the net. The key is to at least allow the current asymmetry to be fully utilized. I.e. allowing full non-discriminatory freedom for all end users to use their share of traffic resources in an application/service/device agnostic way. Myself, 15 years ago, had dreams of getting a full 1.5Mbs/up/down T1 and being able to host something like the 1999 version of slashdot. The fact that 15 years later, GoogleFiber allegedly offers 1Gbs up and down, yet terms of service forbids any commercial server hosting is the problem. It leads to a skewed supply/demand scenario where Google and the rest of the established players tilt the rules of the game in favor of the servers they operate.
While the FCC may be focusing elsewhere -- and on the wrong issues, IMHO, that doesn't mean I can't address the real issue and the real promise of the Internet. In my initial post on this thread I said:
The "evolution" of the last mile is being hobbled (via techno
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Overkill the capture res for more flexible post (Score:2)
GoPro already has 1080p 120fps cameras.
Is the output of these necessarily intended to be viewed at 1080p 120fps, or is it so that the video's producer can digitally stabilize, crop, and slow-mo it in post-production?
Re: (Score:2)
Do you know any one who pirates media any more?
The vast majority of people who have seen the film Song of the South in the past 30 years have pirated it.
Saying we shouldn't have full speed connectivity directly to other end users because some people might share copyrighted material
Or because far more people would use it to infringe copyright than to make and distribute original works. A non-infringing use must be substantial in order to outweigh infringing uses in a defense against an accusation of contributory copyright infringement.
is like saying all phonecalls and texts should be monitored because some people can use those forms of comunication to organize crimes.
I agree. It's just that people who support these "alphabet soup organizations" currently fund the election campaigns for both major U.S. political p
Re: (Score:1)
Once you invite the FCC into regulating the Internet, you end up with a few appointed non-technical guys whose first loyalty is to a political party (In this case, Democrat, but I'm not saying GOP-controlled FCC would do a much better job of regulating) or to the special interest groups they support defining the various parts of the Internet in ways that make no technical sense, but allow them to accomplish their supporters very financially and power-based objectives.
As a result, the Internet becomes requir
Re: (Score:2)
I'm sorry. Were you replying to my post? Because your ramblings have fuck all to do with what I posted. If so, please explain what FCC regulations have to do with last-mile protocols developed by private groups and how they're designed to prop up failing business models?
If not, carry on.
Either way, have a great day!
Re: (Score:2)
I know it's unusual, so I apologize for the shock, but while I was replying to your post, I was actually agreeing with you.
Specifically, "I really hate how cynical I'm getting, but our corporate and government overlords keep taking our freedoms and most people are cheering them on. Good consumers. No need to be a citizen. Just be a good little consumer.", but just expanding on the mechanism a bit.
The FCC will inevitably kowtow to the corporate and other interests and lock in their vision of what the Interne
Re: (Score:2)
I know it's unusual, so I apologize for the shock, but while I was replying to your post, I was actually agreeing with you.
Specifically, "I really hate how cynical I'm getting, but our corporate and government overlords keep taking our freedoms and most people are cheering them on. Good consumers. No need to be a citizen. Just be a good little consumer.", but just expanding on the mechanism a bit.
The FCC will inevitably kowtow to the corporate and other interests and lock in their vision of what the Internet is and is for, discarding the reality of what it can be and what the rest of us would like it to be.
So, carry on...
Ohhh. My apologies. It's business as usual of course. Rock on. I'd only suggest that rather than focusing your ire on the symptoms (regulatory capture), perhaps focusing on the root cause (the monied interests which are the sources of capture and corruption) instead. Have a lovely Sunday evening.
Re: (Score:2)
Huh? There's nothing artificial about it in the case of DOCSIS. Cable was originally designed to multicast video using a shared medium. Putting Internet on top of that is a very clever hack, but it doesn't get around some of the fundamental assumptions and designs of the system.
To download data from the node to the user you merely need to put it on one (or more) 6MHz channels, and the user's modem
Re: (Score:2)
This has been the traditional way to measure the type of peering between two networks: are you a net supplier of data or a net sink for data. Networks with a high number of web browsers would be a net sink, because HTML requests have high ratio of data received by the customer to the data sent. Networks with a high number of web servers are just the opposite. Mos
Re: (Score:2)
The obvious solution is you pay twice: once to the content provider that pays the copyright fees and sends you the content and once to the ISP that delivers the packets to your home. Volume-based billing absolutely makes sense. It makes no sense to the ISP to charge the same amount to a customer that uses 1000GB/month as one that uses 1GB/month. (Well, they'd love to charge everybody as if they were using 1000GB/month but they recognize that many users are unwilling to pay for that class of service.)
Re: (Score:2)
Volume-based billing absolutely makes sense. It makes no sense to the ISP to charge the same amount to a customer that uses 1000GB/month as one that uses 1GB/month.
Volume-based billing doesn't make sense. The operating costs are incredibly low whether it's 1 GB or 1000 GB (if you look at figure 20, you can clearly see high speed data is $2 -- and this is for google fiber's gigabit service -- http://www.businessinsider.com... [businessinsider.com]). The true cost for bandwidth is in the infrastructure.
Re: (Score:2)
The biggest users drive the infrastructure cost. The ISP isn't going to have to upgrade infrastructure because grandma sent her kids a 30K Christmas email. It needs to be upgraded because a user wants two hundred movies, or operate a game server that serves 200 users.
Re: (Score:2)
Re: (Score:2)
You are conflating price and cost.
Re: (Score:2)
Re: (Score:2)
Flat-rate pricing is worse.
Re: (Score:2)
Re: (Score:1)
This last option is what some of the transit and ISP networks
Transit networks love Netflix, it's some of those pesky last mile ISPs that don't like them.
doesn't matter (Score:4, Insightful)
this is about artificial scarcity
how they use technical-sounding language to contextualize that fake distinction is interesting to note, but the core of the matter is that none of the distinctions are actually relevant
also "data sources and data sinks" is also equally non-technical a description of network topology
data "flows" but it doesn't pool up like water in a retaining pond
Re: (Score:2)
edge user (Score:2)
I completely agree that the "end user" and "edge provider" distinction in this case is absolutely ridiculous...it's an non-functional abstraction layer that has absolutely nothing to do with network engineering...there is an "edge router" but the distinction between "end user" and "edge provider" is not technical
however, we cannot be tempted to think that because this confusing, stupid distinction exists, **if we fix it, we fix Net Neutrality**...that's wrong
no matter what the language, it's about the data,
RF (Score:2)
Re: (Score:2)
When I pay for my Internet access, I had better get at least a good faith "Best ef
It has always been this problem (Score:2)
The provides always have created an assymetric internet, in which there are users and servers. While I welcome dynamic ip addresses for privacy reasons, static ip addresses have the advantage of being... static. And with ipv6, you could create static global ip addresses with ipv6, and still have a dynamic ip address for privacy. Does one provider support this? No, because they want to separate between "end users" and "internet services".
If the internet would be fully symmetric (upload speed == download spee
This is war over content, not neutrality. (Score:1)
Peer to Peer (Score:2)
The internet is meant to be peer-to-peer. Sure, some realities mean that we need some dedicated servers. That doesn't change the fact that when we do anything other than treating all peers by the same standards, we impede the open use of the internet.
Re: (Score:1)
Then bring it on, lets BGP to everyone!
If we tried to use BGP clearly the internet asplode. But perhaps if we used modern mesh networking protocols...
Re: (Score:1)
Not going to happen. You will need some AAA from RIPE to distinguish legit/malicious IP subblock announces for both Alice and Bob, even while they're still not connected to some backbone where RIPE is available.
Sure, there will need to be some kind of centralization. But there can be multiple authorities with just a little cooperation. Probably with centralization of its own :)
Re: (Score:2)
Nonsense, the Internet is mostly client-server for 25 years now, not peer to peer. The Internet would not be possible without massive servers and business grade infrastructure. End users are indeed a seperate thing from providers, making emails and facebook uploads is merely end user activity.
real simple... no priority bit increase (Score:2)
no matter how much you pay, no priority bit jump to get your traffic ahead of anybody else's. you can buy a gig if you want to, and it's offered in your area. but you're the same priority 0, or 2, or 3, or whatever the provider grants to non-critical traffic like VoIP or management commands, that all internet traffic users have. that's net neutrality in a nutshell, and it applies to Google as well as Billy Joe Whistlebritches and his 600 kB line ten miles out in the sticks.
Congestion (Score:3)
um no (Score:2, Interesting)
Are they 'end users' when they email their friends but 'edge providers' when they switch windows to check their business metrics?
and that right there sums up why people who don't know how networking works, shouldn't write news articles about it.
In both cases in this example the user is consuming data. A better example would be once the user is hosting data locally, like a webpage... And then, in fact, they'd be required by nearly every ISP in the county to have a business account.
The author acts like he's come up with some novel argument that questions the basic foundations of the ISPs business model, but in fact, it's a ques
Re: (Score:2)
Exactly! Well said.
Just like banking (Score:1)
Just as banking and investment should be considered distinct and often conflicting interests, so should network connectivity and content distribution. These two types of business should not be allowed to vest in the same company. Connectivity is as essential to the modern economy as transportation, and should be provided on the same fair nondiscriminatory terms to everyone.
Traffic, not content (Score:2)
I always defined it in terms of traffic:
An end-user is a network (possibly consisting of only one machine) or group of networks owned/controlled by a single entity that doesn't carry traffic bound for networks owned/controlled by any other entity. So a residential Internet subscriber or their home LAN would be an end-user, as would a typical corporate network that doesn't carry traffic for anyone but the company. An ISP wouldn't be, since it hands traffic off to the home networks controlled by it's subscrib
End asymmetrical billing (Score:2)
If end users are all providers then they should be paying the business rate. Right now everyone is under the mistaken belief that a fair systems as an asymmetrical rate where some users who consider themselves providers are being charged far more for the same bandwidth and covering almost 100% of the cost of delivering internet while others are getting tons of bandwidth at the cost of offsetting some-all the cost for the last few miles. So if you want to claim that everyone is a provider then everyone sho
Re: (Score:2)
Yes. They don't come anywhere near what it costs to move that much bandwidth through the country. Not remotely.
No, just someone who doesn't live in fantasy land about stuff costs. You can't have a car for $20 and you can't have your own personal 1Gbps fiber running over thousands of miles for $7
Re: (Score:2)
I agree. The ISPs are the ones who IMHO created this confusion by mislabeling the services they are providing.
Again. Consumers should mostly be consuming bandwidth from people paying full fare. Business accounts can generate bits. Other than that basic asymmetry they
Re: (Score:2)
Re: (Score:2)
Bengie take a look at what they charge for real business lines: http://www.sonic.net/solutions... [sonic.net]
They also want to distribute via. colo not at endpoints: http://www.sonic.net/colocatio... [sonic.net]
Now for small business at endpoints they offer cheap packages but all the carriers do that. The small business products are mostly offering bandwidth down plus phone (i.e. they look like residential from a network standpoint). Maybe they won't freak about small business servers but they certainly would if the business wer
Re: (Score:2)
Re: (Score:2)
Cost is defined [wikipedia.org] as the value of the next-best alternative that was given up.
In major data centers, download is used very little, to the point of being free.
Upload to customers is what is costly, and since different channels are used for upload and download, the law of supply and demand dictates higher prices for upload in datacenters.
The inverse tends to be true (if not as much) for residential connections.
What you're proposing is eliminating these cost signals that help allocate capital, and keep usage of
Re: (Score:2)
Where are you getting that from? More importantly how is that relevant? Major carrier based data centers are about every 100km along all the major fiber routes. Even if traffic were one direction, which it isn't, from the standpoint of the national map it would still be very expensive to distribute between them.
Different c
Re: (Score:2)
Have you deployed to a data center recently? Like, any data center anywhere? Unless you're in with people doing really bizarre, high-download stuff.. How about how about Amazon's AWS [amazon.com]:
You can saturate your download speed 24/7 and they won't charge you a penny, whereas "Data Transfer OUT" starts at $0.12/GB, past the first GB.
In any event, there's this thing called supply and demand, which says (among other things) that prices of products are set com
Re: (Score:2)
Amazon is allowing you to upload free because when you upload you are generally going to be buying storage. That doesn't reflect Amazon's underlying network costs. They don't have the same cost structures as carriers because AWS offers far more services. So, no it does not cost Amazon 0 for you to upload data. They offset the cost. When they aren't offsetting the cost they charge between $30-120 / tb to move data out. Which is being sold at a reasonable
Re: (Score:2)
Amazon is allowing you to upload free because when you upload you are generally going to be buying storage. That doesn't reflect Amazon's underlying network costs.
Upload (out) is what's expensive. Download (in) is what's nearly free.
Amazon likely pays for the entire pipe which is rated at a certain capacity; they charge those particular prices because of supply and demand. Same thing with your commercial ISPs. The fixed costs and cost of the pipe is irrelevant; the price they charge is chosen because that is what that market can bear.
I think you should go back read about supply and demand. The supply curve is a curve which indicates how supply increases as cost increases. The whole point of the intersection point being on that curve is that price is not independent of cost of providing the service. In particular, carriers are not going to give you services for far less than they cost to deliver.
You're diving into the finer (but equally important) points of a change in supply vs. changing the quantity supplied. Yes, a change in
Re: (Score:2)
You have upload and download reversed it is to and from them not to and from your local systems. But ignoring the terminology reversal, they don't charge you for the bandwidth when you are putting stuff on their server because they are going to collect storage fees.
No they charge those particular prices because in one direction they make lots of money on storage and
Re: (Score:2)
"Upload" and "download" here is from the viewpoint of that server.
Sending data from my computer here on my desk to my AWS instance, AWS bills me nothing. Why? Because it's so under-utilized that it's practically free. People just don't use servers for consuming stuff.
Upload (from server to my desktop) is what it utilized, and pushes prices upward.
Residential connections tend to be the opposite, though there's no hard rule that this must be true, as TFA points out. The pricing phenomenon is decided by how pe
Re: (Score:2)
No that's not why. I've already explained to you why. You can continue to live in fantasy world or you can look at how everyone, including Amazon, is charged.
Re: (Score:2)
Citation needed.
When you OWN the pipe you don't have marginal costs until the pipe becomes saturated. Get out of your little fantasy world where prices are crapped out by little price pixies.
Re: (Score:2)
A small data center is relatively easy to make efficient, but after a certain point, it starts to go backwards. Sometimes you st
Re: (Score:2)
Re: (Score:2)