Net neutrality is a real hot-button issue for a large swath of Internet users and for good reason. The happy accident of interconnected networks and free flow of information has lead to a lot of useful and creative uses of the world’s largest network. That said, it seems that far too many proponents have added a whole host of other issues under the banner of net neutrality, diluting its value and entangling what should rightfully be separate issues. In the words of Inigo Montoya, “you keep using that word. I do not think it means what you think it means.”
Net neutrality is, simply put, not treating a particular application or destination differently than any other. A clear violation would be if Comcast degraded access to Hulu in order to promote Fancast or preserve its own video product. Another clear violation would be charging to not degrade access to particular applications or destinations. It wouldn’t be a violation, though, for a company to pay for priority access. Unfortunately, a bunch of zealots with their particular version of how things should be have started to attack any network management scheme or transfer cap as violating net neutrality. I’m not saying that their anger isn’t justified, but I am saying that what they’re talking about is often unrelated to network neutrality at all.
Network mangement schemes can be considered a net neutrality problem if they affect traffic in an adverse way with no particular rhyme or reason. Unfortunately, the people who have jumped all over Cox for their network management plans either don’t know what they’re talking about or are intentially obfuscating the issue. Since the plan only kicks in when the network is saturated, you’re still able to use your connection for whatever you want without delay during all 20 other non-peak hours. Even when you’re during peak hours, those types of traffic that get bumped to a lower priority are simply delayed, not blocked. And the delays? They are frequently under 100ms, far less than you’re likely to notice.
Even if you totally abandon network management, you’ll still get reamed for it. Comcast decided to just delay all of the traffic on your port if the network is congested and you happen to be the reason why. It seemed all good and fair to me, then the FCC started raising Cain that VoIP traffic from competing carriers would be held up. What’s Comcast to do? If they ignore protocols, they get reamed. If they degrade only certain protocols, they get reamed. You can’t have it both ways, guys. If you’re using a disproportionate amount of bandwidth to the detriment of dozens of your neighbors, it’s only fair to give your traffic a lower priority. Now pick your poison as to how that’s going to happen or, for goodness sakes, lay off the BitTorrent until off-peak hours.
I really get steamed whenever someone tries to confuse transfer caps as network neutrality. Here’s a news flash: the old pricing models where casual users subsidize the heavy used is over. Dead. Gone. Finished. Stick a fork in it; it’s done. It has nothing to do with degrading protocols and everything to do with who should pay for network usage. My water bill has a base charge that includes a certain amount of monthly water usage. If I go over, I pay for the overage. I don’t expect my neighbor to pay for my long showers under my new rain shower head or my green lawn. Why on earth should broadband be different somehow?
The problem is that unless you’re providing an uncapped, unmanaged 100Mbps connection at DSL prices, the vast majority of the allegedly l33t get their panties in a twist and start crying. They don’t seem to get that someone has to end up paying for the bandwidth and the old model of asking light users to subsidize heavy users isn’t terribly fair or even sustainable anymore. Net neutrality is just what most of them duck behind so they don’t look like selfish dipsticks.
Truly oppressive network management schemes are worth fight about. So are pitifully low transfer caps. (I’m looking at you, Time Warner, Frontier and AT&T.) But for the good of REAL net neutrality, stop abusing the term and co-mingling with other issues. It really isn’t helping.
Well said Jesse. What really rubs me wrong is when non-technical advocates for net neutrality equate it to the First Amendment and free speech. I don’t see a lot of technical nuance in the latter, while managing a complex Internet system is loaded with it.
Any bias of network traffic based on political motivation should be prohibited, but it is much easier to go after the offenders than trying to prevent it in the first place with overreaching legislation.
Right on Jesse. The thought that any service provider must allow unfetered transport for anything including his competitor is ridiculous. To think that you must transport services like Vonage for nothing is just plain crazy. This kind of thinking is the mental offspring of the socialist, something for nothing, copyright infringing, welfare state loving slacker.
I think you misunderstand me, Ben. I expect any service provider to make a best effort to transport whatever it is I’m sending that doesn’t disrupt the network for other users. If the network gets congested and I’m the jerk who did it I would expect them, in fairness to other users, to put me at the back of the line. I also expect SPs to give me a clear idea of when, why and how they do it so I can react accordingly. That’s very different from shaping traffic when the network isn’t congested.
Ben, do you think filtering spam is blocking free speech?
No Pete. I think filtering spam is a necessary and responsible thing to do. I do also believe that bandwidth usage monitoring and control is justified and important. There are those who will take every bit of bandwidth that they can just because they can.
Jesse. Of course the last mile is usually the most expensive portion of a network. Providers like Vonage get to ride this expensive path for free. Shouldn’t they be compelled to contribute to a portion of the expense. Who will build these networks if all services ride for free. It sounds like Utopia will compell the consumer to pay for their last mile. I guess that’s one way if it works. I have my doubts.
Saying that Vonage rides for free is false because their customer already paid for the data connection and the transfer that comes with it. Why should ISPs double-bill for a packet of data? It would be like the caller and recipient being billed for a phone call’s toll rate. It makes no sense.
I’m sorry I took your first post for sarcasm. Now I see you were being serious.
Vonage doesn’t pay for the last mile to their servers? I’d imagine they pay quite a bit for that size of a connection.
Metering service types on the Internet at the client end doesn’t work. Data takes the path of least resistance. If you’re going to bill my clients to receive my data, I’ll make my data look like something you aren’t metering for. If you pick apart your subscribers data, they’re going to find an Internet provider who does it like everyone else instead.
Who will build the streets if the cars get to drive on them for “free”? Who paid for the entire Internet before 1994?
network neutrality ends up being a while smattering of different idea’s and concepts bundled into one. some of these idea’s are very reasonable and some less so.
I do believe that their infact does need to be some common carrier style rules imposed onto the internet. I also like the idea of open access for devices(aloud to use any device on a network that is both compatible and legal). I don’t buy that all traffic is created equal, I have no problem with an isp QoS’ing voip up, filtering out clearly ill intentioned traffic etc.
Pete, The server end of the connection is the “first mile”. Of course it’s expensive but not when spread across several million subscribers. On the other hand the last mile is incredibally expensive when multiplied by several million installations.
If we all agree that the subscriber will pay all costs associated with the last mile and pay enough for the last mile builder to have economic incentive then I guess the point is valid. That has not been the case in Provo and it may prove out over time that it is not the case in Utopia. We will likely see many years of floundering before the free market mechanism makes enough adjustments to allow the fiber to the home business case to succeed. But hey that’s my opinion and what do I know.
By the way in 1994 the last mile was a dial-up phone line and a wonderful 28K best case. Anything beyond that cost thousands of dollars per month. It sucked but the business model worked. We live in the land of the artificial these days.
Ben, the business model of charging thousands for anything beyond 28k did not work to do anything but line the pockets of the telcos and obstruct their desire to increase capacity to the end subscriber.
If I’m paying you $5 for a gallon of milk, why would you want to sell me the cow for $10? You wouldn’t, and you would fight any attempt to force you into it.
This is precisely the reason the US started on the cutting edge of Internet access in the 90s, but now isn’t even in the top 10.
So, when you say “the business model worked”, well, if you mean it allowed big business to be profitable while providing inferior service, then you are correct. But if you mean it helped improve the data infrastructure in the United States of America, I would argue you are quite off base.
This is why it is a mistake to leave infrastructure in the hands of a regional capitalistic duopoly/monopoly. It makes more business sense for them to obstruct innovation and growth so that they can continue to squeeze profit out of outdated facilities.
“So, when you say “the business model worked”, well, if you mean it allowed big business to be profitable while providing inferior service,”
That’s exactly what I meant. I didn’t say it was good, but there was economic incentive for operation. At the end of the day, how do you finance what it takes to improve America’s infrastructure. Either the government subsidizes it like iProvo or the customers have to pay a lot more or even pay for their own installation / last mile. This of course assumes that no one like Vonage will ever agree to paying for services transport. Of course this blows a big hole in the Utopia open network model. Why should any phone provider pay them for transport when Vonage does not and it’s a slippery slope. Next it will be Internet bandwidth and video and no one paying for transport anywhere. And hey Warren, I don’t claim to have anymore than an opinion here. You guys at Xmission are way beyond me in knowledge on the subject so I should just get out of the way let you explain it.
I do appreciate the dialogue.
While the market is making a push towards end-to-end IP everything, it’s not there yet. There’s still a difference between voice, video and data from regulation and technical standpoints.
There’s also a difference to customers. Vonage piggybacks onto your data stream and doesn’t include battery backup, packet priority from your router, or whole-home wiring. Sure, you could probably hack most of that yourself, but the average customer wants to plug their phone into an outlet and be done with it. Vonage works pretty good in that regard, but their technical support stinks and you can still get choppy calls when there’s heavy traffic out of your home. There’s still an advantage to using one of the providers on the network.
Same thing with video. Web video sucks. The “HD” streaming that most providers offer doesn’t look as good as over-the-air or paid services. The interfaces to get video content to your TV are very rough and there’s nobody that I’m aware of that does a Pandora-like service for video. It’s also highly fragmented enough that you can’t find all of the content you want in one place. People will still pay for traditional video service so they don’t have to deal with it all.
I don’t think in any way that these competing services undercut UTOPIA providers. They still offer value-added services including support, convenience and quality of product. If someone pays for a video package or spends an extra $50 a month in bandwidth overages, does it really matter all that much?
Bandwidth rates have dropped across the board since 1994 and will continue to drop. The first DS3 (45Mbit) XMission bought was on the order of $40,000 a month. Now, over a decade later, I can get multigigabit connections for a fraction of the price. It isn’t artificial, it is cost-declining technology. That computer on your desk would have been only affordable to government and research institutions a decade ago. Remember racing to the phone when there was a long-distance call? Soon, paying for voice traffic is going to be like sticking a postage stamp on email. Nobody will buy it.
Korea is planning to deploy 1Gbit connections to all metropolitan households by 2012. The US needs to be competitive, or we’ll be a digital third-world soon. You may call this socialism, but I call it necessary infrastructure for the market.
The problem with these data networks is that you have the barriers to entry of a power utility with the technical advances of a computer. As Warren pointed out, any existing provider wants to maintain the status quo so they can maximize the profit of their existing infrastructure. When there are only two providers it’s easy to hold the status quo. The barriers to entry help a lot – and a little lobbying raises those barriers just a little higher. This makes it very hard for a country like ours (capitalist, relatively hands-off government) to keep up with technical advances.
Switching topics – as an ISP subscriber when I pay for my data connection I should get unfettered access to any other node on the network. Any exceptions should be spelled out to me very clearly – the same as with the telephone system and telephone numbers. There is no “first mile” and “last mile”. Every node on the network is just a node. Some are more busy than others, but each node pays to be on the network under terms with their ISP.
From the ISP’s point of view, there must be a reasonable solution to the problem of too much traffic. But that solution needs to be made clear to the customer.