Consumer data consumption has increased exponentially in a little over a decade. According to broadband insight reports from OpenVault, monthly household data usage has skyrocketed from an...
The infrastructure over which that data travels isn’t free. If you have a resource and it has any kind of scarcity, you want to tie consumption to the cost of producing more of it.
You can reduce the transaction cost – reduce hassle for users using Internet service – by not having a cap for them to worry about, but then you decouple the costs of consumption.
Soft caps, like throttling, are one way to help reduce transaction costs while still having some connection between consumption and price.
But point is, if one user is using a lot more of the infrastructure than any other is, you probably want to have that reflected in some way, else you’re dumping Heavy User’s costs on Light User.
Like, what kind of costs exist? Lines, network hardware, putting up the tunnels and poles that hold up lines, the network admins who deal with issues on them. Your ISP can’t just push a button and instantly provide 1Tbit bandwidth capacity at no cost to themselves to every subscriber.
Oh you mean like the $400 billion the industry has taken to adopt Fiber-optic high speeds, but somehow Fiber access has never materialized in most US cities? You mean like that infrastructure? That we’ve already fucking paid for through grants and other federal programs handing money to the ISPs?
Are you having a laugh or do you work for these fuckers?
I’m not disputing the costs, I’m disputing that they already have money to cover the costs (taxpayer money, I might add) and they’re bilking the consumer on top of it.
EDIT: Also, let’s not forget that they got this money during a period of media consolidation. Why was Comcast using its money to buy NBC in 2011 instead of spending that on (*audible gasp) infrastructure?
And I’m not arguing that. I’m arguing that it’s clear that pricing has been decoupled from consumption, but in the other direction, where the ISP’s are setting prices arbitrarily. That’s been a choice on their part, and a big reason why people like me are distrustful of any data they claim shows their case. They have been caught lying so many times before. I’m old enough to remember Comcast paying homeless people[1] to stuff a courtroom to make it seem like more people supported them (once again if they don’t have money to cover infrastructure costs, why are they instead spending their money on things like this?). There’s also issues like when they bundle unnecessary services, essentially consumers paying for nothing, like when the AG of Washington State sued them in 2016.[2] I could go on for pages about shit like this going all the way back to illegally shaping traffic with Sandvine targeting BitTorrent traffic.[3] I honestly don’t wish to and maybe you ought to do more research on how much money these companies ream the American consumer for before acting like there is any connection between pricing and consumption here.
EDIT: Further, if that’s not enough, here’s the Huffing Post with some raw data[4] on how Time Warner was making 97% profit on their internet offerings in 2015. Here’s the Wall Street Journal[5] in 2012 discussing how many ISPs had something like 90% profit on their internet offerings:
On the face of it, such offers appear counter-intuitive. Cable executives and analysts say that about 90% of the money cable operators charge for broadband goes straight to gross profits, since there are minimal operational costs for providing Internet service.
Finally, here’s a breakdown from the EFF[6] on why building infrastructure is ISP’s main cost, and that maintenance and upkeep is a sliver of costs, once infrastructure is rolled out:
Some estimates suggest the cost of deployment can be close to 80 percent of the entire cost portfolio of an ISP. Note that means operations and maintenance of the network (that includes all of the broadband usages of its customers) could be as little as 20% of the ISP’s costs. This is acutely true when it comes to a fiber to the home deployment where the infrastructure (fiber optic cable) is effectively future-proof and can be upgraded cheaply with advances in electronics.
It is worth remembering that our current incumbent telephone and cable companies have made back their initial investment costs because they entered the market as monopolies in the old days and likely enjoyed favorable financing as safe bets (nothing is safer to invest in than a monopoly). Our current incumbents enjoyed a litany of advantages for being the first to deploy. For example, many buildings as they were constructed prospectively required the installation of a telephone and cable line, which in essence gave them virtually a free ride to customers that new entrants will not enjoy.
It’s funny that I have symetical LAN Gb all over my house and beyond the initial cost it costs pretty well $0 per month to operate and maintain. My ISP gives me a limited use Gb line outside the house (and a tiny fraction of that speed on the uplink side for arbitrary reasons) and somehow it costs upwards of $100/month. That has no rational correlation to reality in terms of cost to provide the service.
Massive enterprise systems are on a daily basis maintained by reletivlely small teams of specialists, most of who never have reason to physically touch a piece of gear. That happening in dynamic environment always looking for the next big step. An ISP has a comparativly simple task to direct traffic flows for the most part dictated by automated protocols.
You pay them for a certain throughput, that is your limit, if they can’t provide that limit then they need to advertise and sell the actual limit they are comfortable providing.
If they’re advertising a guaranteed rate, sure – and there are contracts that exist where one does buy guaranteed rates (usually over some period of time, though). Some businesses may buy that. But if you look at a typical consumer ISP, they usually aren’t selling that. They’ll have something saying that the speed isn’t guaranteed, or “Internet speeds up to” or something along those lines.
Internet: Actual speeds vary and are not guaranteed.
The ISP I use (small, most people won’t be using it) says “Up to X speed” next to each price on their pricing page.
Like, consumer ISPs are not going to generally sell guaranteed-rate service, and most customers aren’t going to want to pay for what that would run. That’s not just a function of some users using a lot more than others, but because they’re also overselling the infrastructure. They maintain infrastructure sufficient to handle load if customers are only using a portion of that maximum – that is, if every one of their customers decided to simultaneously saturate their line, even if those customers aren’t particularly heavy users normally, they’d simply overwhelm what infrastructure is there.
Now, that being said, I do think that it might be legitimate to ask ISPs to disclose overselling ratio (or maybe there’s some kind of better metric, like how percent often their internal infrastructure to an average customer is above N% utilization). Or to explicitly disclose soft caps or something. Those might be useful numbers in helping a customer compare ISPs. But they aren’t presently selling and won’t be providing guaranteed sustained rates – that’s just the reality of what kind of Internet service that can be provided at what consumer prices are.
I’d argue that the FCC’s recent Broadband Consumer Label proposal is more important. Part of the problem with broadband as a market is that providers are able to bury the true cost and product under reams of legalese that no one ever reads. Economists refer to this as asymmetric information, where one party to a transaction has vastly more information than the other. Forcing providers to show all costs and restrictions up front would go far in preventing them from fooling customers.
I would also like it to be harder for providers to change their rates. It’s frustrating to constantly have rates jacked up when I’m not seeing much of an increase in service. I finally left Comcast over their rate increases and calls trying to upsell me on services I had no interest in.
The infrastructure over which that data travels isn’t free. If you have a resource and it has any kind of scarcity, you want to tie consumption to the cost of producing more of it.
You can reduce the transaction cost – reduce hassle for users using Internet service – by not having a cap for them to worry about, but then you decouple the costs of consumption.
Soft caps, like throttling, are one way to help reduce transaction costs while still having some connection between consumption and price.
But point is, if one user is using a lot more of the infrastructure than any other is, you probably want to have that reflected in some way, else you’re dumping Heavy User’s costs on Light User.
I want to know where the storage tanks of gigabytes are hiding
Like, what kind of costs exist? Lines, network hardware, putting up the tunnels and poles that hold up lines, the network admins who deal with issues on them. Your ISP can’t just push a button and instantly provide 1Tbit bandwidth capacity at no cost to themselves to every subscriber.
Oh you mean like the $400 billion the industry has taken to adopt Fiber-optic high speeds, but somehow Fiber access has never materialized in most US cities? You mean like that infrastructure? That we’ve already fucking paid for through grants and other federal programs handing money to the ISPs?
Are you having a laugh or do you work for these fuckers?
I’m not disputing the costs, I’m disputing that they already have money to cover the costs (taxpayer money, I might add) and they’re bilking the consumer on top of it.
EDIT: Also, let’s not forget that they got this money during a period of media consolidation. Why was Comcast using its money to buy NBC in 2011 instead of spending that on (*audible gasp) infrastructure?
Federal subsidies to telcos were not intended to provide Internet service for free, but to reduce costs.
You could argue that a subsidy should reduce prices relative to what they should have been, had no subsidy existed.
But you cannot argue that pricing should be decoupled from consumption as a result of that.
And I’m not arguing that. I’m arguing that it’s clear that pricing has been decoupled from consumption, but in the other direction, where the ISP’s are setting prices arbitrarily. That’s been a choice on their part, and a big reason why people like me are distrustful of any data they claim shows their case. They have been caught lying so many times before. I’m old enough to remember Comcast paying homeless people[1] to stuff a courtroom to make it seem like more people supported them (once again if they don’t have money to cover infrastructure costs, why are they instead spending their money on things like this?). There’s also issues like when they bundle unnecessary services, essentially consumers paying for nothing, like when the AG of Washington State sued them in 2016.[2] I could go on for pages about shit like this going all the way back to illegally shaping traffic with Sandvine targeting BitTorrent traffic.[3] I honestly don’t wish to and maybe you ought to do more research on how much money these companies ream the American consumer for before acting like there is any connection between pricing and consumption here.
EDIT: Further, if that’s not enough, here’s the Huffing Post with some raw data[4] on how Time Warner was making 97% profit on their internet offerings in 2015. Here’s the Wall Street Journal[5] in 2012 discussing how many ISPs had something like 90% profit on their internet offerings:
Finally, here’s a breakdown from the EFF[6] on why building infrastructure is ISP’s main cost, and that maintenance and upkeep is a sliver of costs, once infrastructure is rolled out:
https://www.sfgate.com/business/article/Cable-giant-Comcast-hires-seat-warmers-to-pack-3226430.php ↩︎
https://fortune.com/2017/12/24/washington-comcast-service-plan-fraud/ ↩︎
https://consumerist.com/2007/10/27/damning-proof-comcast-contracted-to-sandvine/ ↩︎
https://www.huffpost.com/entry/time-warner-cables-97-pro_b_6591916 ↩︎
https://www.wsj.com/articles/SB10001424127887324073504578109513660989132 ↩︎
https://www.eff.org/deeplinks/2018/05/big-lie-isps-are-spreading-state-legislatures-they-dont-make-enough-money ↩︎
It’s funny that I have symetical LAN Gb all over my house and beyond the initial cost it costs pretty well $0 per month to operate and maintain. My ISP gives me a limited use Gb line outside the house (and a tiny fraction of that speed on the uplink side for arbitrary reasons) and somehow it costs upwards of $100/month. That has no rational correlation to reality in terms of cost to provide the service.
Massive enterprise systems are on a daily basis maintained by reletivlely small teams of specialists, most of who never have reason to physically touch a piece of gear. That happening in dynamic environment always looking for the next big step. An ISP has a comparativly simple task to direct traffic flows for the most part dictated by automated protocols.
You pay them for a certain throughput, that is your limit, if they can’t provide that limit then they need to advertise and sell the actual limit they are comfortable providing.
If they’re advertising a guaranteed rate, sure – and there are contracts that exist where one does buy guaranteed rates (usually over some period of time, though). Some businesses may buy that. But if you look at a typical consumer ISP, they usually aren’t selling that. They’ll have something saying that the speed isn’t guaranteed, or “Internet speeds up to” or something along those lines.
Lemme grab Comcast, for an example.
googles
https://www.xfinity.com/learn/deals/internet#Pricing&otherinfo
The ISP I use (small, most people won’t be using it) says “Up to X speed” next to each price on their pricing page.
Like, consumer ISPs are not going to generally sell guaranteed-rate service, and most customers aren’t going to want to pay for what that would run. That’s not just a function of some users using a lot more than others, but because they’re also overselling the infrastructure. They maintain infrastructure sufficient to handle load if customers are only using a portion of that maximum – that is, if every one of their customers decided to simultaneously saturate their line, even if those customers aren’t particularly heavy users normally, they’d simply overwhelm what infrastructure is there.
Now, that being said, I do think that it might be legitimate to ask ISPs to disclose overselling ratio (or maybe there’s some kind of better metric, like how percent often their internal infrastructure to an average customer is above N% utilization). Or to explicitly disclose soft caps or something. Those might be useful numbers in helping a customer compare ISPs. But they aren’t presently selling and won’t be providing guaranteed sustained rates – that’s just the reality of what kind of Internet service that can be provided at what consumer prices are.
I’d argue that the FCC’s recent Broadband Consumer Label proposal is more important. Part of the problem with broadband as a market is that providers are able to bury the true cost and product under reams of legalese that no one ever reads. Economists refer to this as asymmetric information, where one party to a transaction has vastly more information than the other. Forcing providers to show all costs and restrictions up front would go far in preventing them from fooling customers.
I would also like it to be harder for providers to change their rates. It’s frustrating to constantly have rates jacked up when I’m not seeing much of an increase in service. I finally left Comcast over their rate increases and calls trying to upsell me on services I had no interest in.