The internet is a bunch of tubes

Problem is that we're still using the same over-sold crappy connections we had ten years ago, while the amount of transferred data has exploded.

Someone has to pay for the tubes that can handle all the crap we want to flow through them. Nobody's going to want to cut profits, so how do you expect this to not end up as an increased cost for end users? It's not like ISP's and content providers are communally owned and working for the best interest of everyone - the owners want our money, and providing a service is usually just a way of obtaining them.
 
Someone has to pay for the tubes that can handle all the crap we want to flow through them.
Yes. Obviously. That would be the customers. But the customers DO pay. Some of that profit HAS to go back into investing in the infrastructure but if you think it better spent on private jets and yachts ... well, I see a problem there since the big providers are virtual monopolies so we can't leave just because the service is crappy - no competition - effective monopolies.

The telcos were regulated for this reason - since their business model pretty much meant being a virtual monopoly provider in an area to make financial sense, they were regulated to require them to invest in service while not gouging customers. AND they were required to be common carriers.

Nobody's going to want to cut profits, so how do you expect this to not end up as an increased cost for end users?
Of course the end users pay. That's not what this is about. What it's about is whether SOME users can pay to get their packets put ahead of other people's packets. I.e. can some services pay to beat other services on latency. Right now you pay by your connection bandwith generally, but should services like youtube and netflix have to pay after that to get their packets sent at higher priority in the general internet? Or is all data just data. These guys already pay for fat connections to the internet. Should they also pay to make sure their packets get to you in a timely fashion? What about a new service that has fewer customers so has less ability to pay for packet speed. Sure, they can buy a big pipe like the other guys but in terms of latency they become laggy and less usable, not because they have insufficient bandwidth or servers or software but because someone like google is paying to get their packets put ahead of the little guy.

Customers are willing to pay for higher bandwidth but what is the point of paying for higher bandwidth if the packets you want on that bandwidth are being throttled by the in-between providers? People pay for bandwith so companies find it worth putting it in - but being able to charge some companies higher rates to get available bandwidth first actually works against that dynamic.
 
I ride the train to work every morning. It takes about one hour each direction, and i pay a fixed fee each month.

If I want to be 100% sure that I get a seat each morning, or that I can actually get on the train, I would have to pay about 7€ extra for each trip.

If I don't pay for a seat, I can still get get on the train, but I'll have to stand up all the way, or wait for the next train, if there are people that have paid for all the seats or if the train is already full when it reaches my station.

I don't pay for a seat every morning, I know that I will most likely just be able to one anyway. It's a calculated risk, which I am willing to take. In return, my monthly travel costs are much lower. If I wanted to be 100% sure to get a seat, I will have to pay for it If everyone did that, the carrier would have to get longer trains, or they would have to have make the trains go more frequently. Regardless, there would be more money for expanding, and so everyone would get the service they pay for.

When you pay for an Internet connection, you know the connection will be oversold. If you want guarantees for your bandwidth, you need to be prepared to pay for more, and the fact is that that's what you do when you start expecting to be able to stream things in 1080p from Youtube/Netflix/HBO/Whatever, for 24 hours a day.

How much do you think the average daily utilization of current Internet connections have increased for the last ten years? 10%? 20% 50% How much have the prices increased in the mean time? Where I live, they have decreased by 50%.

We have similar discussions and expectations from our customers at work (I work for a web hosting company), and the fact remains that most people expects prices to go down while their hosting accounts should get more and more bandwidth, storage, etc. It just doesn't add up.
 
Last edited:
I ride the train to work every morning. It takes about one hour each direction, and i pay a fixed fee each month.
But if you don't pay the higher fare your train ride takes two hours.


If I want to be 100% sure that I get a seat each morning, or that I can actually get on the train, I would have to pay about 7€ extra for each trip.
But if they weren't allowed to make extra cash from selling their seats at a higher price for the same service you used to get for the lower price they would have to expand to meet the demand for capacity. Alternatively they can continue to jack up the prices for the same ride on the same equipment by selling "super-priority" seating for 15 euros a pop which give you the right to take a seat from a poor old regular priority passenger.

We have similar discussions and expectations from our customers at work (I work for a web hosting company), and the fact remains that most people expects prices to go down while their hosting accounts should get more and more bandwidth, storage, etc. It just doesn't add up.
So how does your web hosting company feel about paying for the bandwidth they need and then paying extra to get their packets to be put ahead of other users' packets?
It's not about getting a seat, it's about whether you are getting on the train. Charging extra to get your packets on the wire ahead of other people's packets when things get busy is a way to make boat loads of extra money WITHOUT adding capacity. If everyone's packets were treated the same telcos would HAVE to build capacity. Reselling current capacity at premium rates takes the pressure off - it's just free money.
 
Actually, the train ride won't take longer. I would just have to wait for the next train if there's no more room on the one I expected to board.

If they weren't allowed to charge extra for a seat on a specific ride, I can promise you that DSB wouldn't buy more trains to increase the number of departures. It would simply be too expensive, and that would actually result in higher costs for everyone. With the current service, I can chose to pay less and thus save money in exchange for a small risk of having to wait for the next train.

We own our own cables all the way to the IX so paying for the bandwidth simply doesn't fit into the equation. We do, of course pay for maintenance of the end-points (not to mention the ludicrous costs of actually putting it in the ground in the first place), but the cables are there and don't really cost anything.

Now, I actually don't have any idea what the average utilization is, but for the sake of the discussion, say that we are operating at 80% capacity in average. If we were to offer everyone have "Unlimited bandwidth" (which most of our competitors claim to do), the utilization would hit 100% quite fast. As bandwidth would no longer be a factor, many customers would downgrade to smaller hosting packages, and we would lose money, and be forced to dig more cables into the ground.

Looking at the loss that would incur, the board's salaries wouldn't really be that much in comparison.

We do offer customers cheaper packages for less money, but with less bandwidth usage included. Most of our competitors offer "Unlimited bandwidth" and include some obscure fair-use clause in the terms instead.

What ISP's need to do is to stop hiding the fact that they are over-selling. We both know we would most likely never be able to utilize our Internet connections if we were to use them at 100% capacity 24/7. We both know it would be completely impossible if everyone tried to do it.

Average Joe, on the other hand, doesn't have a clue. He thinks he pays for the best internet connection EVAR, and is shocked when someone tries to tell him that he would actually pay more to use his Internet connection at full capacity. Average Jane, will be even more surprised when Joe caves in and pays extra to be able to watch Netflix in 1080p, and finds that there's actually less bandwidth that remains.

In the end, it's a marketing issue more than anything else. Any ISP that tries to fool their customers into believing that they can get full utilization of their connection at any time and to any other part of the Net should be fined to Hell and back. All currently available resources are finite. It's only natural that the competition for them is getting worse once they're getting more scarce.

NOTE: I don't know what my employer's stance on this is. This is my opinion and nothing else. :)
 
Actually, the train ride won't take longer. I would just have to wait for the next train if there's no more room on the one I expected to board.

You're example doesn't really cover it, in my opinion. Let's not forget that you are on a packet based train and so, every station, everyone gets off and then they see who has the expensive tickets to get back on the trains. This isn't a seat or no seat thing, this is an on or off the train. If you don't finish your travel in a short enough time your trip dies and you have to start again from home.

Meanwhile, the job that you are trying to get to, the train company starts up their own company to do the same thing and ... guess what? All those employees seem to have good tickets for some mysterious reason and your company goes out of business because their employees can't get to work while the train company employees can.

That is the sort of situation we are looking at, and there is no good reason for it because bandwidth is cheap. Admittedly, out in the country with big gaps between houses any service is expensive (people out in the boonies don't get water or sewage for those reasons) but in the city bandwidth is cheap per home.
In the end, it's a marketing issue more than anything else.
That's it in a nutshell. How can you sell people a crap deal and make it sound so reasonable that they thank you for it.

Meanwhile AT&T's bandwidth issues go away when they are selling new connections.
 
Back
Top