Goodbye to a flat rate for broadband?

[ic14]

Well maybe hes an xbox live user who wants to host?

Just like me, i want more than 8 in rtcw!

[Steve H]

Originally Posted by ianathuth

I am not a gamer but cannot see the logic in that statement. I would have thought that there would be no difference in gaming between a 600k and a 1Mb connection unless you was hosting a game.

oops Sorry

Meant to say Hosting games.. and as IC said, especially XBOX Live games where a 1mb/256k connection can only hold 6-(maybe)8 players.

[erol]

This article is nothing more than PR and propaganda for a company that hopes to sell lot's of sw to ISP's. Shame on you BBC.

The only basis on which pay for usage could be justified at all would be by the MB/GB. The problem with this approach is the incremental cost of moving the data is so marginal, once the network is in place, and falling at such a rate that any 'system' put in place to monitor such usage and produce variable billing will be more expensive than the 'bandwidth' that is being charged for.

The vast bulk of the monthly payment a user makes goes to 'pay down' fixed costs and not incremental ones (ones based on usage). These are the same for "Mrs Gadekar" as they are for "Mr p2p nutter". If you were to offer a price difference to these two users, that truely related to 'cost of provision', both fixed and incremental, then it would be a minscule difference and not worth the trouble and cost to implement. Similarly if you were to bill users on usage in such a way you might have a user who does not use his conection at all being charged £25 pm and one saturating it paying £26pm, where 70p of the extra £1 is actualy the cost of of 'tracking and implementing' the variable billing and 30p for the extra incremental cost actualy created by the 'heavy user'. This is just madness.

The agenda here is not one of 'fairness' at all. It is a telco agenda of 'regaining control' and a sw producer agenda of selling product. The true agenda can be seen in the statments like

"The last four or five years has been about building this infrastructure of a high-speed network, providing a dumb access," Now, he says, service providers need to make their networks "intelligent" so they can identify users and the applications used.

So the real agenda starts to appear. The whole reason the internet is so powerful, so useful and has spread accross the world like wildfire, is that it is a 'dumb' network. This does not mean their is no intelligence in it, what it means is the intelligence is at the 'ends' of the network, ie in the hands of the 'users' and not the owners of the network. This is the very basis of its power. It is also an anethma to the network companies, reducing their role from one of controling and defining and billing for every application to one of 'commondity transport movers'. This is an attempt by telco's to 'regain' the 'control' they lost when TCP/IP moved that 'control' from their domain to the users domain. Any such efforts will ultimately prove futile but they are sure gonna try in the mean time.

further reading on this subject can be found via these links. The 'rise of the stupid network' is nothing short of a 'seminal' piece on this issue and should be read and understood by every internet user and more inportantly every CEO of a network company.

http://www.rageboy.com/stupidnet.html

and generaly at

http://www.isen.com

To finish off I will just highlight a couple more examples of 'insidious propaganda' being brought to play here.

"Between 60 and 80% of bandwidth is being eaten up by a fraction of customers - who are mainly engaged in peer-to-peer activity - and, according to the industry, the rest are penalised because of the heavy users sharing the network."

ahh an old chestnut here Where do these figures come from ? If they are saying during peak hours (when congestion occurs) that a tiny 'fraction' of users are using 60-80% of the bandwidth, they are quite simply lying. You can only use as much bandwidth as your connection allows and when congestion occurs all users slow down equally. This simply means it is impossible for one user, during peak hours, to use 60-80% of the 'bandwidth'. If they are saying '60-80%' during all hours, peak and off peak, then the statment is meaningless in terms of congestion and one users impact on other users. However this 'story' is highly 'emotive' and a great 'propaganda' tool for the telcos, trying to convince a public that all the gains of a 'dumb' network should be reversed and control should be restored to telco hands, who will then decide what 'services' are offered and bill for each one, regardless of the cost of provision.

next

"Ever been out to dinner with a big group of people, only to feel cheated when everyone splits the bill equally?"

Let's imagine a senario where the actual cost of food to the restuarant is neglegable and the vast bulk of the cost on your bill is for the 'fixed' costs of building the restuarant or renting it etc, so that if you have just a starter your bill is say £20 and if you have a 3 course full on meal it is £20.33, then no one would ever suggest at such a meal that the bill is not split equally. This is exactly the case with internet access. It is even the case with restaurants to some degree when they impose a 'minimum charge' regardless of what you eat or do not eat. When the ratio of these fixed costs to incremental costs is such that the incremental are marginal to the fixed, then flat rate pricing is the only sensible approach.

and finaly

"What could be difficult is to persuade new customers to plan what they want to do on the web. At least in the short-term, perhaps many would prefer to split the bill evenly while they sample what is on offer."

You are damm right is going to be difficult to 'persuade' customers that what they really want is for all control to be taken away from them and put back into the hands of the telcos. The biggest weapon 'users' have in this fight is knowledge and self education and ironicaly the greatest means to this is the internet itself. Of course under a 'telco' model this would be packaged and billed as a specfic service by them and under their control, and priced out of the hands of the mere 'average user'.

[ian@huth]

Originally Posted by erol

"Between 60 and 80% of bandwidth is being eaten up by a fraction of customers - who are mainly engaged in peer-to-peer activity - and, according to the industry, the rest are penalised because of the heavy users sharing the network."

ahh an old chestnut here Where do these figures come from ? If they are saying during peak hours (when congestion occurs) that a tiny 'fraction' of users are using 60-80% of the bandwidth, they are quite simply lying. You can only use as much bandwidth as your connection allows and when congestion occurs all users slow down equally. This simply means it is impossible for one user, during peak hours, to use 60-80% of the 'bandwidth'. If they are saying '60-80%' during all hours, peak and off peak, then the statment is meaningless in terms of congestion and one users impact on other users. However this 'story' is highly 'emotive' and a great 'propaganda' tool for the telcos, trying to convince a public that all the gains of a 'dumb' network should be reversed and control should be restored to telco hands, who will then decide what 'services' are offered and bill for each one, regardless of the cost of provision.

I think that you have to take account of both upstream and downstream bandwidth when you look at this. If you have an upstream channel on a UBR with 200 users on it and eight or ten 1Mb users are leaving their computers running on p2p with their upstreams being saturated by people downloading from them then those few can be using up 60-80% or even more of the available upstream bandwidth and slowing down the service for all users on that upstream card.

[erol]

Originally Posted by ianathuth

I think that you have to take account of both upstream and downstream bandwidth when you look at this. If you have an upstream channel on a UBR with 200 users on it and eight or ten 1Mb users are leaving their computers running on p2p with their upstreams being saturated by people downloading from them then those few can be using up 60-80% or even more of the available upstream bandwidth and slowing down the service for all users on that upstream card.

Upstream is 256k per user 1 Mb user.

NTL claim a maximum (guaranteed !!??) contention ratio of 20:1.

So if these figure are true, contention starts with the 11th user (200 potential users / 20 = 10 users with no effect from contention)

So at this point let's imagine 10 (heavy) users online. That is 5% (of potential users) using 100% of the bandwidth, or 100% of 'active users' unsing 100% of the bandwidth - but at this point no one else is using it so it don't matter.

Lets add another 10 'light users'

So now we have 20 users all getting 128K upstream instead of thier maximum 256k. So now we have 'heavy users' representing 5% of potential users using 50% of the bandwidth, or representing 50% of active users using 50% of the bandwidth. we have the same for 'light users'

Lets add another 10 'light users' (so now we have 10 heavy users and 20 light users)

So now we have 'heavy users' representing 5% of all potential users using 33% of the bandwidth, or representing 33% of 'active' users using 33% of bandwidth. We also have the light users representing 10% of all users using 66% of the bandwidth, or representing 66% of the 'active users' using 66% of the bandwidth.

and so and so (just keep adding another 10 'light users' untill all 200 users are online at the same time)

So where oh where do these _constant_ claims from the telcos (and not just from NTL I might add) of 5% of users using 80% of the bandwidth come from?
Either I have barked way way up the wrong tree here, or someone is telling porkies. If it is the former then please please explain to me where I am going wrong ? In the absense of someone pointing out where my gross stupidity has crept in, in the above, I will stick to the view that there is a strong porcine whiff comming from the telcos.

(hey this is good stuuf. Anticap guys you should use this and you are welcome to if you like)

For more 'interesting reading' on the general topic have a look here
http://www.ntlhell.co.uk/index.php?a...=1&t=3083&hl=&

[ian@huth]

It's nothing to do with contention really. If you have the eight or ten 1Mb users maxing out their upstreams it satutares the upstream channel which is less than 2000kb after you have taken DOCSIS overheads into account. Once that point is reached it starts to affect the downstream as everything being downloaded has to have data passing on the upstream and that is already saturated. You could have a situation where the uploaders who are maxing out the upstream channel are not downloading (leaving their p2p software running for others to download from them) and only a few other users are on that card and trying to download but find their downloads seriously restricted. Robin Walkers post is worth reading at http://www.nthellworld.com/forum/sho...2&postcount=15

[erol]

Originally Posted by ianathuth

It's nothing to do with contention really. If you have the eight or ten 1Mb users maxing out their upstreams it satutares the upstream channel which is less than 2000kb after you have taken DOCSIS overheads into account. Once that point is reached it starts to affect the downstream as everything being downloaded has to have data passing on the upstream and that is already saturated. You could have a situation where the uploaders who are maxing out the upstream channel are not downloading (leaving their p2p software running for others to download from them) and only a few other users are on that card and trying to download but find their downloads seriously restricted. Robin Walkers post is worth reading at http://www.nthellworld.com/forum/sho...2&postcount=15

Thanks for the link. Heavy stuff from the 'wise man' himself but interesting none the less.

However even having read this posts (and understood a tiny fraction of it) I would _still_ maintain that the claim that 5% of 'heavy users' take uo 80% of the bandwidth or cause 80% of the congestion is at best grossly misleading and imo intentionaly so.

from Robin's post

"When we buy a train ticket, we don't get a train to ourselves. Sometimes, according to the behaviour of other customers, we don't even get a seat."

In this senario do we claim that 5% of the people on the train (that use the train when it is not packed almost continusly) do not get blamed for causing 80% of the congestion on the train when it is packed (or that they use 80% of the seats) . When it is packed they cause the same amount of congestion as every other passenger. No more, no less.

It is this constant claim that the telcos use that 5% are responsible for 80% etc that I am trying to debunk, nothin more and nothing less.

If there is a flaw in my use of Robin's analogy then I would genuinely like to know where it is because I can not see it and if it's there I would dearly love to be able to see it.

[ian@huth]

The train analogy is rather flawed as a comparison. It assumes that one passenger takes up one seat. You could have the situation where several large people take up more than one seat each, somebody has vandalised a carriage and is virtually unusable or the ultimate that a gunman has forced all the passengers into the end carriage and his accomplice is pilfering from the luggage and bags left behind. In the latter case two people could occupy 80% of the train whilst 98% of the passengers are crowded ino the other 20%.

The other point to note is what is going on during non-peak times, particularly during the night. Most users are tucked up in bed or may be posting on a forum or in achat room. None of these consume much bandwidth. The file sharers though are likewise tucked up in bed but their computers are hard at work both downloading and uploading. They could be consuming well over 90% of the bandwidth at this time.

If you look at the total data moved by file sharers and compare it to the data moved by others you can see that a small percentage of customers can be responsible for a very large percentage of data moved.

[Ignition]

The 60 - 80% of total usage being attributable to 5% of users claim *is* true, at least 60% of an average ISPs network *is* peer to peer traffic, there is no disputing this these are facts, it's that simple.

Consider that your average user probably only really uses their connection at peak times for surfing, occasional downloading, maybe some messenger / IRC chat, and compare that to someone who maxes 24/7. Approximately 1% of broadband users in the UK use their connection at as close to 100% in BOTH DIRECTIONS as they can get it.

Compare somone who uses 13GB a day to someone who averages less than 100MB a day, then tell this person who averages less than 100MB a day that their connection is more expensive so that Johnny 'Warez Hub' Smith can trade the latest moviez, gamez, appz + mp3z with his mates, and see if they don't mind.

When someone can tell me they don't mind this and prove I'm talking nonsense THEN I'll shut up. Personally like so much else in life you should get what you pay for, and pay for what you get. P2P traffic is the most expensive, providing bandwidth at local level isn't that cheap either, upgrading networks to cope with Captain Warez Hub costs.

[erol]

Originally Posted by ianathuth

The train analogy is rather flawed as a comparison. It assumes that one passenger takes up one seat. You could have the situation where several large people take up more than one seat each, somebody has vandalised a carriage and is virtually unusable or the ultimate that a gunman has forced all the passengers into the end carriage and his accomplice is pilfering from the luggage and bags left behind. In the latter case two people could occupy 80% of the train whilst 98% of the passengers are crowded ino the other 20%.

The analogy is good when Robin uses it and flawed when I use it?

The analogy is not flawed as far as trying to make the point that I am. Yes you might get very small passengers (children) or very large passengers that take up two seats. These in alanlogy temrs would relate to 150kbs connection or the 1mbs connection. The point is there is a limit. No one person is so big that they take up 20 seats, or 50 seats. The limit is the size. As for gunmen / vandalised that is irrelevant (to the point I am trying to make) and interms of the analogy would relate to the network being broken or a gunman taking over NTL and closing the network down.

Where the anaolgy is flawed however is when you consider what happens when the train reaches congestion point (when there are more passengers trying to travel than there are seats). When this happens on a train, then yes you get an unfair situation because some get seats and some do not. When a network reaches this point however things are still fair, because everyone slows down EQUALLY. There is no equivalent of one user 'getting the seat' and another not.

Originally Posted by ianathuth

The other point to note is what is going on during non-peak times, particularly during the night. Most users are tucked up in bed or may be posting on a forum or in achat room. None of these consume much bandwidth. The file sharers though are likewise tucked up in bed but their computers are hard at work both downloading and uploading. They could be consuming well over 90% of the bandwidth at this time.

If they are consuming 90% of 'the bandwidth' at these times that means that there is no problem at all. All they are doing is getting value from a resource that has already been paid for and which if they were not using would still have to be paid for. In train analogy terms again - a train that is 90% full still costs the same to run as one that is 10% full. The problems occur when the train reaches congestion point. Before that there is no problem.

Originally Posted by ianathuth

If you look at the total data moved by file sharers and compare it to the data moved by others you can see that a small percentage of customers can be responsible for a very large percentage of data moved.

The problem is one of 'congestion' not volumes moved. The 'argument' is that one person usage is affecting another person usage (unfairly) and it ratios of 5% to 80%. This is not true and not physicaly possible.

[erol]

Originally Posted by JustAnotherN00b

The 60 - 80% of total usage being attributable to 5% of users claim *is* true, at least 60% of an average ISPs network *is* peer to peer traffic, there is no disputing this these are facts, it's that simple.

Well if you say it is fact I guess it must be. Even accepting this fact what makes you think that the 60% of traffic that relates to p2p traffic is not being caused by 60% of the users of the network? Why, just because one application is popular does it have to mean that only 5% of users are using this application? You are also getting data volumes moved confused with congestion imo. How much data is moved on a network is not the same as one persons use affecting anothers. Any time a network is running at less than 100% capacity, it is actualy 'wasteful'.

Originally Posted by JustAnotherN00b

Consider that your average user probably only really uses their connection at peak times for surfing, occasional downloading, maybe some messenger / IRC chat, and compare that to someone who maxes 24/7. Approximately 1% of broadband users in the UK use their connection at as close to 100% in BOTH DIRECTIONS as they can get it.

Let's say that approximately 1% of road users use their cars almost 24/7, where as majority use them occasionaly. Do you then say that these 1% are responsible for 60-80% of all congestion on the roads ?? Then why do you support this argument in network use ?

Originally Posted by JustAnotherN00b

Compare somone who uses 13GB a day to someone who averages less than 100MB a day, then tell this person who averages less than 100MB a day that their connection is more expensive so that Johnny 'Warez Hub' Smith can trade the latest moviez, gamez, appz + mp3z with his mates, and see if they don't mind.

When someone can tell me they don't mind this and prove I'm talking nonsense THEN I'll shut up. Personally like so much else in life you should get what you pay for, and pay for what you get. P2P traffic is the most expensive, providing bandwidth at local level isn't that cheap either, upgrading networks to cope with Captain Warez Hub costs.

The cost of providing the network to mr '13GB a day' and to mr '<100MB a day' are almost identical. They both have to have the roads dug up and cable laid. They both have to have UBR's and POP's etc etc etc. As the cost of providing the network to them is almost identical then it is perfectly fair that they pay the same price. There is some incremental cost (ie cost that relates to the _usage_ of a connection) but these are marginal comapred to the cost of putting the network in and running, which is EQUAL for both users.

Costs that are incremental should be paid for on a 'pay for what you use' basis. Costs that are fixed whould be paid for on a 'fixed fee' basis. The reality is with BB almost all of the costs are fixed - hence flat rate pricing. It's simple really.

[erol]

To be honest I simply give up

JustAnotherN00b and ianathuth WANT to believe that _their_ connections are being slowed down by a mysterious '5%' and nothing, it would seem, will convince them otherwise, not logic or facts. That is their perogative I guess.

Of course _their_ usage never impinges on anyone else's. It is only the usage of this mysterious 5% that ever effects anyone else. They want a pay as you use model because they don't use it very much, but they also want to ingnore that the bulk of cost of _their_ conection is the same regardless of how much they use it. This reality does not fit their desired belief, so they ignore it.

This is not intended to be personal, I merely use these two as examples because they have been vocal on the issue. It could be any 'they' that holds this same 'mythical' belief.

[etccarmageddon]

[QUOTE=erol]They want a pay as you use model because they don't use it very much[QUOTE]

it's already available...

http://www.metronet.co.uk/adsl/paygo

[erol]

Originally Posted by etccarmageddon

it's already available...

http://www.metronet.co.uk/adsl/paygo

So why don't those that want such a model just use such a service, I wonder, rather than try and insist that _everyone_ use a pay as you go model on the grounds that anything else is 'unfair' ?

[ian@huth]

Originally Posted by erol

To be honest I simply give up

JustAnotherN00b and ianathuth WANT to believe that _their_ connections are being slowed down by a mysterious '5%' and nothing, it would seem, will convince them otherwise, not logic or facts. That is their perogative I guess.

Of course _their_ usage never impinges on anyone else's. It is only the usage of this mysterious 5% that ever effects anyone else. They want a pay as you use model because they don't use it very much, but they also want to ingnore that the bulk of cost of _their_ conection is the same regardless of how much they use it. This reality does not fit their desired belief, so they ignore it.

This is not intended to be personal, I merely use these two as examples because they have been vocal on the issue. It could be any 'they' that holds this same 'mythical' belief.

I would say that you have got an idea into your head and nothing will persuade you to change your mind.

The train and the gunman analogy is valid because a few users can affect the service of many. The two gunmen are grabbing the majority of the seating capacity because of their actions. If ten 1Mb users are online using p2p and maxing out their upstreams (there may be very little us of their downstreams) then they will have saturated that upstream card. The upstream card has a capacity of probably less than 2000 kbps after DOCSIS overheads are taken into account and the 1Mb users at max are using 256kbps. 10 x 256 = 2560 in my book and that is more than the card can accomodate so there is already a slowdown in performance for those ten. Along comes anothr 1Mb user who just wants to download a large legal ISO from a website. He can not achieve the 1048 kbps download that his connection is capable of because his upstream ack packets are slowed down because of the upstream being saturated. Another 100 users come on line at any speed and wish to browse, download or watch streaming video. What will their experience be like? Everyone on that upstream card will have a performance which is slower than dial-up. If the ten p2p users then go offline the other users will more than likely enjoy a normal experience.

Bandwidth can be said to be the amount of data that can be passed along a communications channel in a given period of time (look it up on dictionary.com). If the given time is 24 hours then it should be easy to see that a relatively small number of customers on the fastest connections that are using them to the max 24/7 can pass a hell of a lot more data than a very large number of customers that are only on for short periods and may be on much slower connections and just doing a bit of browsing and e-mail checking.