Slashdot Mirror
AT&T/Comcast Consider Aussie-Style Bandwidth Caps
LazySiow writes "Having looked at Australia's pioneering efforts in cappedband services,
AT&T Broadband and Comcast are considering applying download caps of their own. Since the two approved a merger proposal last week, they will be the largest broadband provider in the States, and will not only affect a large percentage of of users, it will set a large and potentially unstoppable precedent for caps all around the country."
Just don't call it "unlimited internet", or it's false advertising.
-John
That's actually completely reasonable. If your only browsing and checking e-mail, maybe a game of quake here and there, then you might actually start paying less than you are now, with the multi-iso/night and movie freaks paying more of the cost. It's completely fair.
personally, i hope they cap speeds, not download limits. my cable company (time warner, who privides road runner) already has an option for "business lines", cable lines that download twice as fast and upload several times faster, for about double the cost per month. there are even more choices beyond that. while i dont need the extra bandwidth, id gladly pay an extra 10 bucks a month for my service now.
Good point. It's unrealistic to think an ISP is not oversubscribed. So there will alwais be some kind of cap. P2P makes this worse, much more than the average user is willing to understand.
The conspiracy has one simple, ultimate goal: to transfer as much money from your pocket into theirs. They have the will and organized money to make it happen and there is very little you or anyone else will be able to do about it.
You can make false claims that you are all powerful and can take your business elsewhere, but then you will all realize all businesses operate in this manner. They will all charge bullshit fees, they will invent reasons to charge you more bullshit fees, and they will all utilize contracts that lock you into them. They will all, in short, steal as much of your money that can get away with.
Welcome to the "free" market---free not as in fair, but free as in free to steal.
<a href="http://www.joblessjimmy.com">Work is dumb and so is Jobless Jimmy.</a>
No, thats not the idea. The corps want you to spend more money on your broadband, not download less.
So, you can download both redhat and pr0n, but you will be priced a bit higher. Now how much is that RedHat iso worth to you ?
Probable impossibilities are to be preferred to improbable possibilities.
Aristotele
There are no doubt many other providers that would love to have you.
The funny thing is, no there aren't.
Hear me out.
If you're the kind of user who downloads a gig a day, runs a web server, a MUD, a webcast radio station, and several sessions of KaZaA, the providers don't want you. They'd much rather do without your 30 or 40 dollars or whatever you spend a month than have to spend more providing you with bandwidth and technical support. To them you're more trouble than you're worth, and if by instituting a cap they lose you, well that's the price they're willing to pay.
Users who saturate their connections should not pay the same as users who occassionally browse the web, but like to do so at high speed.
That's just wrong. Primetime traffic sets the lower limit of the bandwidth a network provider has to install. When everybody and their mother want to surf at high speed at the same time, then that too puts a limit to overselling. The rest of the time, the network is more or less underutilized. A reasonable solution would be to cap the bandwidth (not the volume) during primetime, using a bursty traffic model: This would not cause any slowdowns for webtraffic but would restrict full performance for continuous high bandwidth applications to off peak hours.
Paying for actual usage is not as easy as you think. Usage is neither "line bps" nor "total bytes". Usage is the influence on network performance for others, which is hard to measure and includes factors like time of day, short term temporal traffic characteristics, and even user base.
All I can say is that the ones hurt will be the internet marketers...People will rise to arms to keep popups, flash anims, and cascading pr0n from taking the precious 5 gig. Forget proxy systems... I forsee people using software to selectivly avoid servers that are bandwidth saps. Killing the online marketplace.
Mozilla offers a form of this, right clicking on a pic and telling it to never download a pic from this server again. dark days are coming and I see alot of Surfing dollars being spent where marketers could be footing the bill.
as a AT&T victim^H^H^H^H^H customer I think I'll be looking into this as soon as I'm done posting.
THE WORLD IS GOING TO END!!!! eventually.
I've had a fair amount of experience in rolling out DSL services, and one thing that I always found strange is that lots of people in the industry disagree in terms of the way that the bandwidth should be carved up.
/. users would prefer :
Generally the expensive part of the bandwidth is the fibre leg between the DSLAM and the equipment on the ISP's bandwidth.
The discussions always revolved around the most fair/effecient way to carve this bandwidth up, and there were basically two original ideas : firstly put a limit on the users modem and have a free for all on the fibre. The problem with this is that p2p users can degrade the perceived QOS for all. Secondly you could basically channelize this fibre and make sure everyone has their allowed bandwidth. This obviously doesn't allow for oversubscription on the fibre which is a bad thing for consumer networks (we were designing a network predominently for businesses, so it was OK).
The important thing that we came upon is that there is a third option, which I think most
Give all users a low guarantee : say as low as 64k. OK, I know that you're thinking that's low, and it is, but thats guaranteed. The point is then you can divide the bandwidth on the fiber by 64 and work out the maximum no. users that you can have, which will be a lot, definately economically viable. Even when you've got this maximum no. of users, you'd be surprised how many users don't use their bandwidth (even with the proliferation of p2p services). This unused bandwidth goes into a pool that all users can take from. It's a bit like a burstable service. It means that you'll always get 64k of low latency service (for me that's just as important as the bandwidth....) and with the tests we made, you still managed to get some pretty decent download speeds. And with this way, you don't have to start putting caps on to increase the perceived QOS
tom-george.comBecause geeks rate higher t
Dude, you have to sleep, eat, and shit sometime. In other words, you're not going to be using your connection at home more than, say, sixteen hours a day. Odds are you're going to be at home and online no more than six hours a day for your average work/school day. So in other words, instead of dividing 166 MB per day by 24, dividing by something in the 6-16 range would yield a better estimate of your max-to-avoid-breaking-cap rate. (about 60kbps constant at 6hours usage per day)
Note: this still sucks. Just not quite as bad as you're calculating. Then again for the "average" net user who just wants to get their family's e-photos and surf ebay quicker, well... They're not going to feel it. OTOH people like me that have a 128kbps music stream open for hours at a time would definitely be getting close to the cap.
Now, if you wanted to run a public server at home, yeah, you need to calculate based on 24 hour potential usage. News flash, these companies aren't in the business of being IP pipes for home servers.
I am firmly against bandwidth caps, and here's why.
The moral is: don't punish people who like your service. I don't get punished by DirecTV and TiVo because I watch 20 hours of TV in a week instead of 2. True, Internet access requires more infrastructure per user than satellite does, but DirecTV has a per-user infrastructure cahrge as well (more satellites; installation; tech support). I expect that additional infrastructure charge to be covered in my monthly bill.
Even traditionally per-use models, such as long distance, are moving to flat-rate fees for those who use them a lot. You can now get unlimited long distance for $30 a month thanks to VoIP, which was spawned by the same technologies that made the Internet possible.
Don't cripple the growth of the Internet by advocating bandwidth limits. The only thing you will end up crippling is the continuing introduction of new, interesting websites with full-motion video and audio. The last thing we want is people defecting back to 56K, or worse, moving away from the Internet completely because "it's just not worth it."
Broadband has made the Internet thrive. Don't hold that progress back.
Simpli - Your source for San Jose dedicated servers and colocation!
A large part of the problem is the misuse of the Internet big companies are trying to force. Rather than treating the network as peers they want to have a few centralized services under corporate control and lots of little users that just sit there and suck up products and canned media. Essentially trying to turn the Internet into television/newsprint. It just doesn't work well.
If ISP's would embrace people that want to run their own web servers, P2P, etc they could reduce a lot of their upstream bandwidth usage. How many people look for local news on a server half way across the country? How many check their email on servers sitting somewhere at Yahoo? How many download the newest game, movie, or music from a distant P2P peer? That is a lot of bandwidth they don't need to waste.
Smart ISP's would provide community sites within their own network (and encourage power users to make their own sites) and provide nice web-based mail. A local IM server would be nice. Offering good proxy servers for web-surfing and a local P2P server that users can connect through rather than using servers elsewhere on the Internet. All are good ways to reduce the ISP's bandwidth usage while keeping happy customers.
I've seen community ran wireless networks that offer all these things and do a very good job at it. If ISP's aren't careful with their limits eventually enough users will join such community network projects that a good deal of the ISP's business may suffer. Wireless networks now are pathworking their way into covering most major cities and even rural areas. At the same time advances are being made in long haul signals for wireless. Eventually this will be a threat to the ISP/telco business and they just accelerate the shift by driving away power users.
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
i think this was meant as a joke, but it does raise a (semi)valid point.
We've always been paying for spam via higher access charges, cost of ISP resources, equipment, etc. Nobody really cares because it's an abstract cost. But in a system where, after a certain transfer threshold, you are charged per byte(or megabyte, gigabyte, whatever) i can tell you EXACTLY how much any given spam is costing me. Sure, it's likely a trivial sum, but multiply it by the huge volume sent to everyone, and we'll have a rough idea of how much money spam costs the recipiants. That number might not be so trivial.
Then again, maybe it will be and we would need to rethink the "it costs us money" line from the antispam arguement. (not to try to vindicate spam in anyway, it mearly helps to have the facts straight) Anyway, i'm curious about it. Anyone have numbers for this?
btw, similar arguements can be made for popups and banner ads for stuff i'd never buy anyway.
Well, only temporarily. Very quickly, geeks across the country begin buying T-1's and starting their own, small, unlimited, ISPs.
What's more, when you become the ISP, you can tell the RIAA/MPAA to fuck off when they send you a cease & desist letter about one of your customers. You might end up in court (for sure, if you are so blatant about it) but that's quite rare.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Welcome to the 80/20 rule. 80% of the bandwidth is used by 20% of the users, and 50% is being used by the top 10% of users. (Or it could be the top 5%. I did the sums back when I worked at an ISP, but my memory of these things is hazy now) Now, a little mathematics. You rewrite your user contracts to target the top 10%, and they leave.
Suddenly you have effectively twice as much bandwidth for your remaining users as before. With decreased expansion costs and increased service-levels for your remaining customers, you could quite easily profit from your customers "voting with their feet".
I bet the cable companies are just shaking in their boots over your threat to leave.
Flat-rate pricing is a myth. It does far more damage to the Internet than it heals, since the need to artificially prevent people from fully utilising their connections without charging them more is is the cause of stupid rules like "You can't run a server and we'll cycle your IP occasionally" that really do impact on user freedom.
Charles Miller
The more I learn about the Internet, the more amazed I am that it works at all.
There is a perception that companies never like to use customers, but this is wrong. If you clog their pipes, they really don't want your business, and they may be saving money by seeing the back of you.
Assume Joe Mpeg "votes with his dollar" and leaves Foo Company, no longer downloading 50 GB a month. Guess what? Foo Company can replace him with 50 "normal" customers who only download 1 GB a month each. In fact, Foo Company is likely to implement download caps designed expressly to get rid of unprofitable customers like Joe Mpeg, whilst still keeping the casual user who will never hit the cap during normal use.
There are two kinds of people: 1) those who start arrays with one and 1) those who start them with zero.
Wow, how many falicies or errors in one message.
"Like cable TV" - OK, you can suck anything you want off the internet - only one catch, no uploading! You can't ask for anything in particular. You only get what somebody else decided to send and exactly the same thing your neighbor gets.
"number of miles" About 1/3 or more of the price I pay for gasoline is taxes. More miles equals more gasoline equals more taxes. We all do pay by the mile for road use.
Pipe vs. water. I'll hook up a real nice, fat data pipe to your house for a small, one-time fee. However, if you happen to want data to flow through that pipe, its going to cost you extra.
The dotcom crash happened because nobody actually had a way to make money. The ISP's are just on the tail end of that. They have some revenue stream but nothing to tie income to costs. You'd better be ready to start paying for what you get.
Profanity - The sign of a small mind trying to express itself.
Not for nothing, but if Comcrap, now that they are a complete monopoly, start capping anything (They already throttle news to 1 gig/month in my area) then they better provide software to stop FRIGGEN BANNER ADS, flash ads and ANY piece of spam from hitting my computer. Period. They want to cap stuff? Fine. I don't want to pay for things I didn't order.
refusing all non TCP packets addressed to customer's machines and all incoming TCP connections
Even doing something like blocking SYNs is not enough. You have to actually statefully track all connections. Otherwise, I can just send you tons of large TCP packets that appear to be part of an existing connection, but actually aren't. Even then, malicious hosts could "accidentally lose" your ACKs and keep retransmitting the same packet for days. I guess some will say that you could sue the host sending you packets, but what do you do if it's some 0wned PC in Korea?
Also, what if the ISP loses my ACK, forcing the other end to retransmit? Why should I be charged for that?
The core of the problem here, and the reason there is so much potential for abuse, is that the ISPs are considering charging the recipient of a packet, when the recipient has no control over how many packets may be sent to him.
And what other provider should I choose? That other Cable company that is not in my area? That DSL provider that refuses to service my area? These providers are government sanctioned monopolies. That is way they can change what they want when they want, for the most part you DO NOT have a choice. They also conspire between each other to offer similar services and pricing structures. I do have the option of dialup but this is not in the same playing field.
Bad boys rape our young girls but Violet gives willingly.
Nope. In central Maryland (Howard County) there is one cable provider (Comcast) and no DSL (too far from the telco). Satellite is laggy and generally not Mac compatible. So what do you recommend?
One addition to this... violate my privacy a bit and take note of what I'm requesting to download. If I want an ISO image from Red Hat, and my 19 of my neighbors wants the exact same thing... cut your external bandwidth by 95% by making only one download from the Red Hat server, and the multicast that out to all 20 of us. Much more efficient usage of the network for all involved.
Unless you consider "move" a choice, which believe it or not is exactly what I was once told by my cable company (before they were bought by TCI, later bought by AT&T). They had the nerve to tell me to my face that they don't have a monopoly on cable TV because I am free to move! With this attitude, is it any surprise they will cap downloads? It's simple math: Those who use the most have the least option to switch, so they're the most likely to pay whatever you charge. Those who use the least could always go back to dialup Juno for email, so you have to treat them nice.
If all this should have a reason, we would be the last to know.
If they do it with a "download agent", then you'll only be able to use their network with Windows. Think this one through again.
I think we've pushed this "anyone can grow up to be president" thing too far.