I will start off by stating that I am a libertarian and that it is an infringement on the rights of the people running these companies when the government steps in to regulate how the Internet service providers bill us. The statists among us will insist that the ISPs need regulating because they have a quasi-monopoly on delivering this access… ignoring the fact that it was created for them, and is sustained for them, by the CRTC in the first place.

It is the usual Charley Foxtrot of the bureaucrats and politicians using problems they created as justification to step in and intervene even more. Like the financial mess in the US where they regulated themselves into a real estate bubble and then come to the rescue by “solving” the problem with… yet more regulations? These are perpetual job creation program for bureaucrats and a never ending source of vote purchasing for politicians. It also allows corporations to buy their way into protected and very profitable situations. This is never in the best interests of the population at large and often not in the corporations bet interests in the long run. I think this is one of those cases.

That said, I have no moral or philosophical objection to usage based billing. It seems eminently reasonable to pay for what you use. I do have a pragmatic objection to it as they are proposing to implement it here.

You see, their state protected position allows them to artificially throttle the usage and cripple access to the Internet in the name of short term profits. They set the baseline usage ridiculously low and then charge several orders of magnitude more than it costs for the inevitable overrun. This is bad for the customer but also damages the business in anything but the next few quarters.

IP is poised to take over our communication infrastructure, entertainment delivery and work via interaction with data in the cloud. Everything we exchange that can be reduced to bits will be and because of that, we can set a baseline of usage that encompasses everything we do now that falls under that umbrella.

Our protected ISP industry has held us back to a developing world level of Internet access- we’re lucky if we can get a 10Mbps feed and then they pretend that it is high speed Internet access. Uhhh no, 100Mbps is high speed Internet access; 10Mbps hasn’t been high speed since the last century and 1 Mbps being advertised as “High-Speed Lite” is actually insulting. To make matters worse, this advertised 10Mbps is small-printed as “up to”. This means they have covered their ass as long as, every once in a while, they manage to get up to 10Mbps. I have yet to meter out more than 7.5Mbps on an “up to 20Mbps” package.

If they were honest, they would advertise the minimum they hit rather than the maximum. It should be “no less than” not “up to”. The only use for listing a maximum theoretic limit is to deceive the customer. The only use.

So the first thing they need to do is drag themselves out of the dark ages. Make no mistake, this isn’t easy or cheap. Small countries with large populations have a distinct advantage in building out the infrastructure. It costs far less to do much more when you have a population density of South Korea at 487 people per square kilometres rather than our 3.4 people per square kilometre.

But it certainly isn’t a hundred times more expensive, much of our population lives in metropolitan centres located within a hundred kilometres of the Canada/US border. While rural homes may have to accept 10Mbps access for a while, the ISPs have some explaining to do about urban homes where they want to charge $150 for a hobbled “up to” 100Mbps package.

And we will need true high speed access if we want to fully utilize the Internet and stop falling further and further behind.

The big data hog is entertainment, television being the main culprit. Whatever form the broadcasters and cable networks take in the future, we will always watch a lot of television. What we won’t do is all watch the same television. The average home has three televisions and three people living in in so the Internet feed needs to handle three HD streams at once.

Good compression algorithms mean that an adequate image can be streamed for about 10Mbps… so you can see where an “Up To 10Mbps!” is some seriously weak sauce. The average peak television usage would be 30Mbps and would allow the Internet Service Providers to offer an acceptable base package starting at “No Less Than 50Mbps” that would meet demand and leave a comfortable buffer for the next decade or so.

There are a significant number of households that would require more than this. Some would simply have more people living in the household while others would be more intensive in their use of the Internet with multiple HD feeds going in the background along with several streaming radio and a self hosting website that consumes a few tens of megabits per second on its own. There are also going to be more and more people working with software and data that lives in the cloud – many of who will need tens of Mbps for that alone.

To service these higher usage households, the ISP would be remiss if it did not add one or two tiers above it- let’s just go with 75Mbps and 100Mbps. The pricing of these services should reflect what we are currently paying for Internet access as well as our cable bill and a land phone line- with a healthy discount for consolidation. It would be crass of them to take undue advantage of the situation to try and hide an increase inside that consolidation… so I fully expect them to try to do that very thing.

In a competitive market, the delivery of actual broadband services should making a healthy profit at $1 for 1Mbps. Practically every urban household in Canadian will be subscribing to Internet access and it would behove the ISP to try and deliver the best bang for buck to capture those millions of customers for decades of $50, $75 or $100 every damn month.

So now that we’ve spent several screens to establish how fast the data need to come into the house, we need to figure out how much data would be pulled down every month.

Going back to that three people per household and using the average of four hours of television watched per day- let’s round down from that and assume there is some overlap on what is watched and go with ten hours of distinct television per day. That is mostly for ease of use in the napkin math I am using here but I also think it is a reasonable metric to use.

Since we are using 10Megabits per second as the HD data rate, that works out to 1.25 Megabytes per second which is 75MB per minute for a total of 4.5 GB per hour. I know that using the better algorithms and lowering our quality demands can give us a compromised image for only 1.5GB or a little more. I really don’t want to watch an image full of compression artefacts and so I won’t use numbers that reflect the lowest common denominator when projecting forward.

So, using my quality standards and some napkin math and statistics gets us to 45GB per day on average for television alone. One of the three is streaming radio stations, another is playing a serious game online and the third is working from home complete with video conferencing and cloud computing… we can easily round that up to 50GB per day as a working basic usage rate.

So, in a household in a developed country- the Internet Service Providers should be looking at 1.5 Terabytes drawn down over a 50Mbps service as a working model for what they need to deliver monthly for not a great deal over $50. Right now, that 1.5TB would cost them between $30 and $40 to deliver and they would have to work on the economy of delivery as well as alternative revenue streams… but building a working business model based on the real world is their job. I am just pointing out the parameters to them.

Now, there are other households, with six or seven occupants who have a serious Netflix addiction and watch two or three movies in addition to the twenty hours of TV every day. So these people will be using closer to 5TB of data per month and so the ISPs will want to meter that and charge a reasonable rate. If it costs them two and a half cents to deliver a GB of data then we are looking at an additional cost to them of about $90. This household is a high usage should probably already be paying the $100 for the higher bandwidth and their usage would be partially covered by that higher rate… but the ISP would have to charge an additional $75 to $100 for the higher than average usage.

This household’s bill could easily hit $200 a month and I don’t have a problem with that- it is a reasonable usage based billing scenario.

The closest I could find to this hypothetical baseline service is the maximum service offer of SHAWS “Nitro” package. This advertises “Up to 100Mbps download speed” which realistically might be able to sustain a “No less than 50Mbps” data rate in the few places that it is offered. How does the rest of the package measure up? Well, they cap it at only 350 GB per month which is about 23% of a reasonable baseline and they want $150 which is about three times what it needs to be.

So in reality, they are not that far off what they should be delivering. Since it is actually closer to a true 50Mbps service than a true 100Mbps they need to charge a third of what they are charging and allow four times as much data transferred. These aren’t order of magnitude differences and should be reachable in a reasonable timeframe. We don’t have to live in a second rate information topology.

Now that we can see what they can and should deliver, what are they actually proposing?

The big telecom companies that built and operate the pipes for the Internet in Canada want to impose a 25GB per month cap and then charge for every GB you go over that. The CRTC was created to consolidate the bureaucracy into one-stop shopping for the telecom industry so they rubber stamped it and gave the companies $1.90 per GB in English Canada and $2.35 per GB in French Canada.

So let’s get this straight, they want to cap it at less than 2% of a reasonable base usage and then bill the customer $2,850 per month in overages to get the other 98%?

How about those poor ghettoized Quebecois who could then be charged $3,525 per month? Or that hypothetical high usage household above that would be dinged $9,500 or $11,750 respectively?

Obviously it would not be possible for Canadians join the developed world if they have to pay what the telecom/CRTC are asking for – and we will fall further and further behind the rest of the world as they build out for the future and Canada entrenches themselves in the 1990s.

So my message to the ISPs of Canada is to look out past the next few quarters of their bottom line and work toward a business plan that will make them competitive in the developed world and deliver what the customer needs rather than crippling them with last century’s data plans. If they are too blatant about the money grab and do to much harm to the customers- that leaves them vulnerable to politicians sacrificing them on the alter of votes.

I am pretty sure that I am yelling into the void on this one.