Comment Re:Download caps are not as bad as they are made o (Score 1) 656
I would have to agree that reasonable caps are actually a good thing.
From 1995 to 2001, I worked for a dial-up ISP. We serviced about 15,000 customer across 30 cities (most of them small towns) in Arkansas.
Back in the those days, the number of hours you spent on-line was the "capped" resource since the telecommunications economics of the time meant that the server end of each dial-up connection cost the ISP well over $100 per month, IIRC. We were charging $20 a month for an initial block of hours and then some amount for each hour over the limit.
I think the initial hours allotment was around 120 hours per month, which was fairly generous for the time. Only a few of our users went over this. Those that did were offered a more expensive account that had a larger allotment.
However, when all the other ISPs went to "unlimited" accounts, in order to compete, so did we. With no incentive for people to disconnect, many of our customers would start their connection, start something (like a ping) that would keep the connection active, and leave it connected 24/7. As this began to tie up more and more lines on the server side, more people started doing this so that they wouldn't get a busy signal when they actually wanted to use the service. Then there were those that gave their login information to all their friends. With unlimited usage, why not?
Obviously, this killed our quality-of-service. Just "adding more lines" was not an option unless we doubled or tripled the cost of an account. The economics just didn't work.
We considered implementing some complicated (for us at the time) monitoring so that we could detect those who were abusing the service. We could try to detect sessions that were not *really* doing anything and limit simultaneous connections on the same account. However, when we realized how much work this was going to be for us, we decided to try something else first.
We changed our unlimited accounts to have an initial allotment of 360 hours, which was triple the amount we had before we went unlimited. In order to use this up, people would have to be connected for 12 hours every single day of the month. If people actually did this, the economics still wouldn't work, but this was not a decision based on economics, but on psychology.
At the time, most people's actual usage didn't come close to 360 hours. The psychological effect of knowing that the cap was there had the desired effect. Our customers only connected when they actually needed to do something. Our quality of service returned to its previous higher-than-average level.
Of course, there were a few customers that had been using more than 360 hours per month. A few of those were using more than double that amount by using multiple modems. Most of these left our service for someone else with an unlimited offering. Good riddance.
I know that comparing hours and bandwidth is comparing apples and oranges. My point is that once you get people actually thinking about how much of the resource they are using, they will monitor their usage themselves, usually not coming close to the cap. Also, the abusers of the service will either cough up the funds to pay for what they are actually using, or go somewhere else. The point is that the service provider can actually afford to provide a quality service.
It is interesting that this dial-up usage issue seems to have been re-incarnated in the age of broadband.
From 1995 to 2001, I worked for a dial-up ISP. We serviced about 15,000 customer across 30 cities (most of them small towns) in Arkansas.
Back in the those days, the number of hours you spent on-line was the "capped" resource since the telecommunications economics of the time meant that the server end of each dial-up connection cost the ISP well over $100 per month, IIRC. We were charging $20 a month for an initial block of hours and then some amount for each hour over the limit.
I think the initial hours allotment was around 120 hours per month, which was fairly generous for the time. Only a few of our users went over this. Those that did were offered a more expensive account that had a larger allotment.
However, when all the other ISPs went to "unlimited" accounts, in order to compete, so did we. With no incentive for people to disconnect, many of our customers would start their connection, start something (like a ping) that would keep the connection active, and leave it connected 24/7. As this began to tie up more and more lines on the server side, more people started doing this so that they wouldn't get a busy signal when they actually wanted to use the service. Then there were those that gave their login information to all their friends. With unlimited usage, why not?
Obviously, this killed our quality-of-service. Just "adding more lines" was not an option unless we doubled or tripled the cost of an account. The economics just didn't work.
We considered implementing some complicated (for us at the time) monitoring so that we could detect those who were abusing the service. We could try to detect sessions that were not *really* doing anything and limit simultaneous connections on the same account. However, when we realized how much work this was going to be for us, we decided to try something else first.
We changed our unlimited accounts to have an initial allotment of 360 hours, which was triple the amount we had before we went unlimited. In order to use this up, people would have to be connected for 12 hours every single day of the month. If people actually did this, the economics still wouldn't work, but this was not a decision based on economics, but on psychology.
At the time, most people's actual usage didn't come close to 360 hours. The psychological effect of knowing that the cap was there had the desired effect. Our customers only connected when they actually needed to do something. Our quality of service returned to its previous higher-than-average level.
Of course, there were a few customers that had been using more than 360 hours per month. A few of those were using more than double that amount by using multiple modems. Most of these left our service for someone else with an unlimited offering. Good riddance.
I know that comparing hours and bandwidth is comparing apples and oranges. My point is that once you get people actually thinking about how much of the resource they are using, they will monitor their usage themselves, usually not coming close to the cap. Also, the abusers of the service will either cough up the funds to pay for what they are actually using, or go somewhere else. The point is that the service provider can actually afford to provide a quality service.
It is interesting that this dial-up usage issue seems to have been re-incarnated in the age of broadband.
