Wireless ISPs like Verizon and AT&T are up in arms against the FCC’s new Net Neutrality rules. The FCC plans to formally adopt six principles – the first four generally enforced since 2005, plus two additional rules:
1. Consumers are entitled to access the lawful Internet content of their choice.
2. Consumers are entitled to run applications and use services of their choice, subject to the needs of law enforcement.
3. Consumers are entitled to connect their choice of legal devices that do not harm the network.
4. Consumers are entitled to competition among network providers, application and service providers, and content providers.
5. Broadband providers cannot block or degrade lawful traffic over their networks, favor certain content or applications over others and cannot “disfavor an Internet service just because it competes with a similar service offered by that broadband provider.”
6.Broadband providers must be transparent about the service they are providing and how they are running their networks.
Apparently the second two rules stifle innovation. How? Well, for one it provides an equal footing for all internet services, regardless of the bandwidth it wants to use in relation to competing services. It lets a service develop without any hindrance from the ISP. It also means ISPs can’t put off upgrading infrastructure by bottlenecking services at their discretion, instead having to fill the demand for faster bandwidth lest their competitors do it first. Oh wait… these things aren’t stifling innovation. They spur innovation.
Their argument is fairly simple. There might not be enough bandwidth to go around. Users can move from node to node on the network very quickly. An area that was serving very few can suddenly be serving very many. If there’s an event, concert, emergency, etc., there would be an unforeseen increase in network load on a node which might have usually served few customers. To handle this you would do one of two things – cap every person’s network speed equally (which causes a slowdown for everybody) or bottleneck the guy who’s downloading a game on Steam, so the user who’s trying to access school email doesn’t notice a slowdown. That makes sense, right? After all, it would be terrible if you couldn’t access a mission-critical service because someone who paid the exact same amount of money for the exact same service in the exact same area can use said service to for something recreational.
When you reach a certain number of users and it’s too expensive to profitably upgrade your infrastructure, and there’s not enough bandwidth to handle all of those users, what would be the logical thing to do? The above? I don’t think so. Either cap the bandwidth evenly for all users and all services, so each user can use the system equally, or increase the price of all users’ plans evenly, so you can afford to upgrade your network and handle higher speeds.
Is this not basic supply vs. demand? Am I missing something? If there’s not enough bandwidth to go around, I pay more. Sure.
However, what is not okay is for the dumb pipe to tell me I can only use (100-N)% of the bandwidth I paid M dollars for because I’m using a service they deem less important, while my neighbor can use 100% of the bandwidth he paid M dollars for because he’s using a service they deem more important.
What does it mean to cap each user equally? If a wireless tower serves out X Mb/s, and Y people connect, then each person is capped to X/Y. However generally someone would only use a fraction of X/Y, and each time they do, there’s some unused bandwidth, (1-P)(X/Y) per person where P is the percentage of bandwidth they’re using that’s not being allocated to a service.
There is a certain amount of bandwidth, S, wasted:
. Y
. S = Σ (X-XPi)/Y
. i=1
The amount of wasted bandwidth is the sum of the unused bandwidth of each user, so long as the bandwidth is capped evenly among all users.
To keep this bandwidth from being wasted, you would allocate it among the users who are currently capped at the physical maximum, X/Y. If there are Q users capped, then their total cap would now be X/Y+S/Q. Eventually one user exists who is capped at a very high speed, or more than one user are using the same amount of network, capped at the same speed.
Isn’t this giving the heavy users priority over the lighter users? No. Why? Because, theoretically, all users now have a theoretical cap of X/Y+S/Q. This is because as a lighter user increases his network utilization, P, the sum of wasted bandwidth, S, goes down, and so does therefore the additional bandwidth granted to all capped users. As less network is used (fewer or lighter services are being run) then the network will be faster for all users, if all users are using more (let’s say close to 100%) of their network speed, each user gets closer to that X/Y. However it is important to note that no user would experience less than that even distribution of X/Y. No service is arbitrarily deemed less important than another.
If an ISP wanted to promise, say, 2 Mb/s as the absolute physical minimum, they would have to estimate the maximum number of users, Y, and implement a system that can handle 2Y Mb/s, then create an infrastructure that could handle that. Obviously it’s more complex than that, since you can figure the average amount of wasted bandwidth and try to get the theoretical minimum as close to the promised bandwidth as possible. If they mis-estimate, then they’ll have to upgrade their infrastructure or promise lower speeds. This is still oversimplified. The carrier doesn’t want a lot of wasted bandwidth at night, for instance, but at the same time they don’t want a slow useless network during peak times.
I’ve seen a post proposing bandwidth caps. When you reach a certain amount of bandwidth in a month, your overall speed gets capped. You could have weekly or daily allowances. You can lessen restrictions at night, for instance to encourage users to download large files or use heavy services (the basis of “free night a weekend calling” which has existed since the dawn of time). So long as they are transparent, there could be many reasonable and viable methods of bandwidth management implement.
What I’m trying to get at is this: It’s not outside the realm of reason to say every user can have an equal amount of bandwidth. That’s not the hard part. It’s possible to do this without wasting bandwidth. If during peak times, bandwidth is too low, then advertise it as that low, or find new technologies and implement a faster system and charge more. I don’t mind good clean economics.
AT&T says this:
“On a wireline broadband network, you know where your customer is,” Jim Cicconi, AT&T’s senior vice president of external and legislative affairs, said in a statement, “So you can build capacity to handle the peak demands. But on a wireless network, you have a crowd converge on a site that suddenly has 10 times or 100 times the users competing for the same resources.”
If I move into a crowded area, then there will be more users there. Of course I’m going to have a slowdown. Verizon/AT&T/Whoever can take averages, gauge usership, etc. in many areas, and upgrade the infrastructure there.
If there’s a sudden influx of people anywhere (for example, there’s a large concert that attracts a lot of users of your network in an area where you had a weak infrastructure), there will probably be a slowdown.
However, as I described above, we should all have equal slowdowns. Cap everyone equally. Guess what? The user who is downloading a movie or streaming video will hit a slowdown. The user who is accessing web pages or tweeting or chatting might still be under that X/Y cap, and might not notice a slowdown at all, even when there is one. My own network utilization, for instance, is about 75 KB/s while Thunderbird checks three mail servers, I’m running a DC++ client, an Pidgin IM with 5 services installed and communicating, and Sunbird syncing with Google Calendar. These services don’t use much bandwidth and send/receive data in bursts, like when checking a server or sending or receiving a message.
Let’s say the tower can handle 100 Mb/s and there are 1000 users. Each user’s “cap” will be .1 Mb/s. Seems slow. I’d expect most of those 1000 users to be idle (or close to it) at any given point in time. The unused network bandwidth is applied to whomever’s streaming Pandora, YouTube, or what have you. In the “large concert” example (or “sudden crowd convergence” as AT&T puts it) most users would at most likely be using communication protocols. Very rarely will a “crowd converge” and all try and install a Steam game or stream YouTube HD at once.
What wireless providers, are arguing, I suppose, is this:
Rather than let bandwidth be spread evenly across all users, which would cause slowdowns evenly, since one would have to cap everyone evenly, instead we’ll cap users who we think don’t deserve to use the same percentage of bandwidth as everyone else.
Net neutrality is all about keeping providers from discriminating against various content or services. This provides an equal and open ground for any service to, without an arbitrary bandwidth restriction, connect to and communicate with its users, and keeps data just that – data – without putting a higher or lower value on those 1s and 0s.
I’d rather have a clear cut Max Bandwidth vs. Users vs. Network Utilization equation spelled right out to my face by my ISP than for them to decide which services, at their discretion, deserve priority over others. We all pay the same price for the same bandwidth. Nothing makes my bits better than anyone else’s bits. They should deliver the amount of bandwidth I paid for. If they can’t, then they should advertise it at a value they’re able to deliver. From there I can, as a member of a free society, make an informed decision on whether or not such a service fits my needs, purchase, and use the service.
Generally AT&T is on top of things, though. Wireless carriers do occasionally listen to their customers, and they do make upgrades. Why, then, must they whine about these net neutrality rules? There’s no reason for it.
