Metered Internet?

An article in the New York Times today discussed what could quite possibly be the next wave in internet access following the propagation of broadband access to the masses: metering the internet and charging by bandwidth used.

Now, as a moderate internet user, I can certainly see the validity of the argument for this style of accounting for internet use. I certainly don't mind that the p2p user next door would have to pay a surcharge for all of his 'illegal' file sharing that slows down my internet speed. And, as the article relates, at least in terms of Time Warner, they will, for the first time, have different internet plans based on the expected bandwidth use of the client. Sounds good right? If you're a light internet user, using the 'net only for mail and google searches, then you can get off paying much less than the high-end WoW gamer.

There's a catch. Anyone who has run a NETSTAT on their computer knows that there are many, many more hidden programs running in the background that use bandwidth, such as iTunesHelper or the notorious Windows Media Player Network Sharing Service. Given that the majority of computer users most likely have no idea that there are processes running that use bandwidth even if they are not actively surfing the net, this could pose problems for those users. Furthermore, from personal experience in the IT field, many, many internet users have no idea about the concept of bandwidth being limited, and some even go so far as to complain about slow buffering times when watching youtube videos, for example. Who knows, maybe if this procedure takes hold, the average user will become more conscious about their bandwidth use, and about the structure of the internet in general, instead of simply viewing it as a collection of tubes.

The other major concern I have about this is something the article briefly touches on, the procedure for actually metering internet use. For starters, the internet has vastly outgrown what anyone expected it to become in the late '80's and early '90's. There was never any thought for a mechanism to monitor bandwidth use; and while it wouldn't be too hard to implement, there are many problems that could arise. First is bandwidth piracy. If the ISP doesn't monitor bandwidth at the site of the client's router, then it would be feasible for others to tap into that client's data stream somewhere along the line and run up the legitimate user's bandwidth at no cost to themselves. The second would be wireless networks. The bandwidth of wireless networks fluctuates greatly with the number of users at any given time. This could prove tricky for a household or company who sees usage patterns fluctuate, as there is the desire not to pay surcharges for excessive bandwidth use, while at the same time to not be paying for unnecessary bandwidth. It's a tricky situation that I think needs to be pilot tested under a variety of circumstances before being foisted upon the unsuspecting public.

So all in all, I'm up in the air about this one. I would have to see some more definitive figures from the ISP's to compare them to current prices before I could judge this effectively. It is just a slippery slope to walk down, for there are plenty of methods that could prove disastrous to the entire industry if mismanaged. And lord knows what an economic disaster that would be.