One Argument Why Data Caps Are Not a Problem (fierce-network.com)
- Reference: 0175328825
- News link: https://tech.slashdot.org/story/24/10/26/2118227/one-argument-why-data-caps-are-not-a-problem
- Source link: https://www.fierce-network.com/broadband/are-data-caps-big-deal-not-really-says-openvault
> OpenVault believes that data caps on broadband [2]are not a problem because most people do not exceed their existing data caps . OpenVault contends that people that do exceed their broadband data caps are simply being forgetful — leaving a streaming device on 24x7, or deploying unsecure WiFi access points, or reselling their service within an apartment building.
>
> Yes, there may be some ISPs that have older networks that they have not upgraded. Or maybe they are unable to increase network capacity in "the middle mile" of their networks, but the Covid pandemic certainly encouraged many ISPs to upgrade their networks and capacity while many ISPs that had broadband data caps ended that feature.
>
> Perhaps the biggest problem, according to OpenVault, is that most broadband users do not really have any idea how much bandwidth they "consume" every month. If Internet access is a service that people want to treat as a "utility", then you have to ask, Would they keep the water running after finishing their shower?
In the article Ookla's VP of Smart Communities adds that "Scrolling through social media feeds for hours can 'push' hundreds of videos to the user, many of which may be of no interest — they just start running." So the main driver for usage-based billing wasn't to increase revenue, OpenVault CEO Mark Trudeau tells the site, but to "balance the network a little more..." (Though he then also adds that sometimes a subscriber could also be reselling broadband service in their apartment building, "And that's not even legal.")
"If one or two customers on a given node is causing issues for 300 others, where those 300 are not getting the service that they paid for, then that's a problem right?" he said.
Having said that, the article also points out that "Many major fiber providers, like AT&T, Frontier, Google Fiber and Verizon Fios, don't have data caps at all."
[1] https://slashdot.org/~NoWayNoShapeNoForm
[2] https://www.fierce-network.com/broadband/are-data-caps-big-deal-not-really-says-openvault
What a nice "argument by hallucination"... (Score:5, Insightful)
Also, if, supposedly, most people do not exceed their data-cap, why does it even make economic sense to implement them at considerable effort and cost?
Re: (Score:3)
Data caps aren't exactly hard to implement. My router will let me do it for anything on my LAN with a fairly easy config setting. I think the only reason they exist is to upsell consumers to an unlimited plan.
They are rather pointless though. Even if someone were pirating as much content as their connection would allow, there's no way that they could consume it anywhere near as fast as they could acquire it. Once upon a time there may have been concerns about that, but the pipes got fatter, the compressi
Re: (Score:2)
How about somebody hosting several BitTorrents? Depending on how popular they are, they might well reach their data cap in one or two days.
Re: What a nice "argument by hallucination"... (Score:2)
Yes. The Comcast data cap applies to the sum of downloaded and uploaded data.
Re: (Score:2)
Like a lot of stuff in a typical EULA, it's an extra stick to beat the consumer with if they need a beating. Most people do not exceed the data cap and in most cases the network is able to cope well with traffic, but if there's a case of a subscriber using massive amounts of data on a somewhat oversold link, they can choose to enforce the data cap rather than make a potentially expensive investment just to accommodate traffic from that one user.
Re:What a nice "argument by hallucination"... (Score:4, Interesting)
It seems unlikely that data caps are primarily about protecting overloaded links at peak times: they do keep you from just running at advertised speeds at all times; but, were they the only mechanism at play, a heavy user could still be overloading the line for some days after the cap resets before they hit it again. Any congestion management you want to do(and aren't restricted from doing, it's not like there are SLAs here) you'd want to do significantly more quickly than that.
It's possible that the psychological effect of a looming fine helps keep utilization down, especially among people who don't really know much about how much various things consume; but as an actual congestion management tool a meter of "X GB/month" is pretty poor: it counts traffic the same regardless of how heavily loaded the line is at the time, so does nothing to encourage off hours usage or backing off in response to signs of congestion; and, in the case of all but the most draconian caps or the most questionably supported nominal peak speeds, it generally won't stop one or more heavy users from hammering the line for days after whatever the reset date is.
Re: (Score:1)
Suppose 100 people use a service. 99 of them use 1-5 units per day. The 100th uses 200 units per day. If you offered that service and units cost money to produce, you might consider implementing a cap of 10 units per day that wouldn't affect 99% of your users, and removes the one who costs you the most.
I'm not in favour of caps generally, but I can see a circumstance in which they would make economic sense to a comany.
Re: (Score:1)
Bandwidth is not oil or fresh water. No one is going to run out of it. It also has zero marginal cost so there's no cost to "produce it".
As long as power is supplied a router will deliver bits. At ISP scales there's very little power difference between full utilization and partial utilization.
Re: (Score:2)
Indeed. Here (Europe), I pay for _bandwitdh_, not usage (both wired and mobile). Incidentally, bandwidth is also what the provider pays for. It makes no sense anymore to have data-caps. The only use-case is to squeeze money out of customers, i.e. market failure.
Burstable bandwidth (Score:2)
Do you pay for burstable bandwidth or sustained bandwidth? As I understand it, caps are meant to deter customers from misusing burstable bandwidth as if it were sustained bandwidth.
Re:What a nice "argument by hallucination"... (Score:4, Interesting)
1TB is nothing when 4K TV is involved. I'm on FIOS and I just checked my devices. Our main TV 4K streamer is at 1TB just for that single device. Overall the house is at 2TB and the month isn't finished yet. 95% of the 2TB is streamed video. If the ISPs truly want to lower bandwidth demands they'd allow the streaming companies to set up more caching servers without charging them outrageous fees. To put this in perspective, if I ran my gigabit connection at capacity it could deliver about 300TB of data so I am only using 0.6% of my available bandwidth.
Re: What a nice "argument by hallucination"... (Score:2)
No, fuck you, ISP man.
Re: (Score:2)
Except that's not how their cost structure works. Their line provides and costs based on peak units per second. If their line is provisioned for 150 units per second, it costs them fuck all extra and implies no requirement to increase if one guy is using 2000 units or 2m units all during times when total utilization isn't exceeding 150 units per second. If the line becomes saturated, then it makes sense to throttle people disproportionately contributing to it to keep everyone with a fair share. But not days
Re: (Score:3)
The 100th uses 200 units per day. If you offered that service and units cost money to produce, you might consider implementing a cap of 10 units per day
Except computer networks do not work like that. With computer networks your cost is Peak units being used when the network is ccongested , and the infrastructure you had to build to support that peak. Units taken outside the congestion period do Not cost anything.
This is why when cities have roadway congestion - they Don't create a "Miles driven" overa
Re: (Score:2)
> if, supposedly, most people do not exceed their data-cap, why does it even make economic sense to implement them
Their purpose is to deter people from exceeding them (and it seems to work).
Re: (Score:2)
The purpose of the limit is to be a limit? Seriously?
ISPs these days pay for bandwidth in their upstream. They have always paid for bandwidth in their own networks. Bandwidth restrictions make sense. Data-caps do not.
Re: (Score:2)
> The purpose of the limit is to be a limit? Seriously?
I was wondering why you asked. It's pretty simple really.
> Bandwidth restrictions make sense. Data-caps do not.
The data cap sets the moment where you get a bandwidth restriction. Such that you know in advance, and can plan for your usage.
I have a fibre landline with "unlimited" and the ISP told me before signing that there is actually a reasonable limit, but they can't tell me what it is. I found it confusing. I totally prefer the function of my mobile phone where I have a data cap, I know how much it is from the start and I can find where I am within the limi
Approximates burst billing (Score:2)
> Bandwidth restrictions make sense. Data-caps do not.
I think ISPs impose data caps as a proxy for the difference between burstable and sustained bandwidth. They must figure that nontechnical residential customers are less likely to understand burst billing than a cap.
Re: What a nice "argument by hallucination"... (Score:2)
No, their purpose is to extract more money from laying customers. Cap fees are like overdraft fees. They're made up to extract more money from paying customers, and oftentimes in a way that you won't know until the bill comes due.
Most is not everyone, there are spoilers too (Score:2)
> Also, if, supposedly, most people do not exceed their data-cap, why does it even make economic sense to implement them at considerable effort and cost?
Because most people is not everyone. There are enough spoilers outside of everyone that will over consume and ruin it for everyone.
Quid pro quo. (Score:4, Funny)
Sure. Let's treat broadband as an utility as long as those who send data over my connection do not send me ads, TV commercials, trackers, session-recording addons and similar crap. Quid pro quo. Deal?
I'd prefer the telcos follow the cell co's lead (Score:2)
Why not just slow people way down when they get above whatever arbitrary threshold the company picks? Seems like that would discourage the people who actually intend to abuse the system while not punishing the rest - including the "forgetful".
Re: I'd prefer the telcos follow the cell co's lea (Score:1)
Telco towers are hooked up to cable net lines..
Re: (Score:2)
As long as they're honest with what they are selling, and not just in the fine print, it's ok.
Re: (Score:1)
> Why not just slow people way down when they get above whatever arbitrary threshold the company picks? Seems like that would discourage the people who actually intend to abuse the system while not punishing the rest - including the "forgetful".
Dafuq? That IS how it's done in most case.. It's called "throttling"....
That's a problem right? (Score:3)
> "If one or two customers on a given node is causing issues for 300 others, where those 300 are not getting the service that they paid for, then that's a problem right?" he said.
Ya, the node isn't capable enough. But the company would have to invest money to fix it...
Or be honest about what they sell. (Score:2)
Kind of like overbooking a flight by the airlines, then the potential passengers must deal with the repercussions.
Why aren't 'sales/purchases' considered a contract?
Re:That's a problem right? YUP! (Score:2)
There are a gazillion ISPs out there running ancient TCP software, and we get to suffer from it. I wrote about it at [1]https://cacm.acm.org/practice/... [acm.org] For a really short explanation, there's a 5-minute video there.
And yes, Hanlon's razor applies: never attribute to malice that which is adequately explained by stupidity...
[1] https://cacm.acm.org/practice/you-dont-know-jack-about-bandwidth/
Re: (Score:2)
>> "If one or two customers on a given node is causing issues for 300 others, where those 300 are not getting the service that they paid for, then that's a problem right?" he said.
> Ya, the node isn't capable enough. But the company would have to invest money to fix it...
Which means raising the rates on the 298 to service the 2 over consuming. Or just have a two tiers. One the 298 fit in and one the 2 over consumers fit in, charge according.
Bullshit (Score:1)
Data caps are a way for ISPs to lie about bandwidth, and prevent them from running into issues when the cut corners and drastically under-provision.
Re: (Score:1)
Absolutely this. That this guy is spewing nonsense is proven by this: "Having said that, the article also points out that "Many major fiber providers, like AT&T, Frontier, Google Fiber and Verizon Fios, don't have data caps at all." I guess those four are just operating as charities and eating the tremendous expense associated with customers using too much bandwidth.
Re: Bullshit (Score:2)
This, this, and this.
just wait for cable co's to see TV drop and then t (Score:2)
just wait for cable co's to see TV drop and then the caps will come down to make up for the loss of $
Wait a moment (Score:2)
US-ians have data cap on residential broadband? I thought they were a thing only on mobile...
Re: (Score:2)
Some do, some don't
Re: (Score:2)
I was prompted to check and I do not.
Re: (Score:2)
I've had three different ISPs since 1998, and have never been subject to a data cap.
This isn't an "argument" (Score:5, Informative)
It's shilling by a lying corporate weasel who knows nothing about network engineering and is attempting to cover for greedy behavior by incumbent ISP monopolies/duopolies who under-provision their networks and over-charge for them.
As someone who does know about network engineering (after 40+ years I ought to), I can tell you that it costs more to implement data caps than not. Why? Because they have to be put in place, mechanisms built to implement them, accounting and billing set up to handle them, customer support set up to deal with the fallout, etc. It's not just a technical measure that exists in vacuum, it incurs a lot of cascading costs including significant human time. It's easier and cheaper to just add capacity -- and it gets easier and cheaper every year.
So why do ISPs do this? Artificial scarcity and plausible deniability. It's an excuse for exorbitant pricing and network congestion. And this dirtbag is playing right along with them.
Re: (Score:2)
> It's shilling by a lying corporate weasel who knows nothing about network engineering and is attempting to cover for greedy behavior by incumbent ISP monopolies/duopolies
Well yeah, that is literally his job, although he would describe it in a more positive light.
But they've spent the effort to support it now. (Score:2)
I agree they're a shill, and there is added work/effort to support data caps...
But big companies already support the ability now. Incremental work is likely small compared to the added profits.
I'd look at this like tax preparation though. It's extra work that someone who won't be doing it decided everybody else must perform. It has no or little real value to the people doing the work. And somehow society decided it's normal/OK.
Re: (Score:1)
> not just a technical measure that exists in vacuum, it incurs a lot of cascading costs including significant human time. It's easier and cheaper to just add capacity -- and it gets easier and cheaper every year.
Data caps, create revenue streams. They also do not incur capital expenses that would otherwise dig into the executive bonus coffers.
Increasing capacity, will incur cost. And might result in increased revenue.
Not really a dilemma for the executives making that decision.
Re: (Score:2)
> As someone who does know about network engineering (after 40+ years I ought to), I can tell you that it costs more to implement data caps than not.
Ok but this is really a question for accounting, not engineering. Engineers aren't very good at accounting, that's why you almost never hear them talking about things like technical debt.
Natural levels of usage, tiers (Score:2)
> It's shilling by a lying corporate weasel who knows nothing about network engineering and is attempting to cover for greedy behavior by incumbent ISP monopolies/duopolies who under-provision their networks and over-charge for them.
Is it? Do you find it hard to believe that people can be segmented into tiers representing their natural usage and pay accordingly?
The real argument is whether or not the tiers offered represent these natural levels.
I'm shocked, shocked, I say! (Score:3)
per wikipedia:
> OpenVault LLC provides network management, policy control, data integration, and business analytics software as a service that is designed to help communication service providers (CSPs) achieve revenue and operational goals.
So the guys that sell software to nickle-and-dime customers don't think data caps are a problem? Astounding! /s
Re: (Score:3)
And there it is, the hidden agenda. Their sales go down if ISPs simplify their network management by committing to no caps.
A gigabit connection can download 400GB an hour (Score:2)
Meaning that the 1.2 TB cap common on certain isps can be reached in 3 hours. That means only 0.4% of your connection's maximum potential can be used before you are capped. This is why gigabit connections should make data caps obsolete, as most people only use the max potential for short term bursts like downloading a game and then then only using a small trickle for general web browsing.
And for others they won't hit 1.2TB in month (Score:2)
> Meaning that the 1.2 TB cap common on certain isps can be reached in 3 hours.
And for many that 1.2 TB of data is more than they consume in a month. Not all users are the same. Hence tiers make sense. Pay for what you need without.
And you can set a car on fire too ... (Score:2)
> Meaning that the 1.2 TB cap common on certain isps can be reached in 3 hours.
Yes, I can fill my car's gas tank, punch a whole in it, and start a fire. Consuming all that fuel in 3 hours. Or I can not set the car on fire and have the fuel last a week.
Data caps ARE a problem, because "cap"! (Score:3)
Look -- I'd find it acceptable for an ISP to throttle a really high speed connection after you exceeded a large enough monthly quota. But by that, I mean such things as "1Gbit download speeds cut in half to 512MB/sec" until the next billing period.
That would help alleviate the claimed excess traffic generated by forgetful people who leave some streaming video going non-stop while they're out of town for 2 weeks or ?
But these greedy bastards always have proposals that start charging you excessive "per GB" type fees once you go over their arbitrary limits, or they just effectively cut off your broadband until you pay for another month at full price (regardless of if this happened only a few days into the last billing period). Ridiculous levels of throttling also amount to cutting off your service, since it becomes unusable if your broadband speeds drop to the point you can't have more than one device doing things at a time or your remote work via VPN or remote terminal server connections gets impacted.)
On top of that? It's like someone else already posted -- where the limits they set are absurdly low. A 1 or even 2TB cap is not going to do, especially when people want to start streaming 2 hour long movies in 4K resolution, increasingly have a lot of data backed up in the cloud with services like Microsoft OneDrive or Apple iCloud, and when you can't even use many devices you buy without a broadband connection. (EG. My Bambu Labs 3D printers default to uploading each print to their cloud server to be sent to the printer from there. It's nice in the sense I get remote control of my print job from anywhere via my smartphone. But a detailed print can easily be hundreds of megabytes in size, and I own 5 printers right now that I keep fairly busy with projects since selling these prints is a side gig for me on weekends.) I rely on a fast and reliable broadband connection for my day job in I.T. too, remoting in to machines all day long to provide remote support or sysadmin tasks. My kid and I like to play online games too.) It's not gonna fly that because we binge watched some streaming shows and then she listened to a lot of streaming music in the form of YouTube videos, a data cap wound up exceeded before the month was up, preventing all these other things from working right.
Re: (Score:2)
I would not disagree that there are greedy corporates out there who want to maximise revenue while minimising investment.
However, if you turned it around, if there are 2 out of 300 customers per node who are causing issues, then why wouldn't the company just want to get rid of those two customers and not have to invest more and actually have 298 happy customers and however many happy executive and shareholders. I am sure they would rather not have the other 2 as customers.
That being said, I would expect a c
Great, we agree on tiers ... (Score:2)
> I would expect a company providing, say, a gigabit internet connection to not have a data cap that you could hit in 3 hours.
Many people do not exceed 1TB a month. Gigabit would still make sense for them, less latency, no buffering pauses, etc.
> So a gigabit internet connection with a 1TB cap would be silly. But a 10TB cap might be justifiable.
Great, we agree that tiers make sense. Buy what you need. The only real argument is whether the current tiers offered represent natural levels of usage by some segment of users or not.
Bad argument (Score:2)
Fuck off, ISP man
No problem? Great! Then don't have data caps (Score:2)
Somehow the thing that incurs fees in this article is not "the problem". Instead "the problem" is just the same 15+ year old BS about bandwidth scarcity that is every day even more untrue than it was the day previous.
Lack of tools (like OpenVault supposedly sell)? (Score:2)
"We cannot manage the network to prevent that person hurting 300 other people, so we're just going to charge them extra." Sounds closer to reality.
And I'm confused, was that statement by OpenVault too (that Google says create broadband management tools)? This summary mentioned Ookla as the source.
I can't remember what Ookla does beyond my landline across internet device that they recently stopped supporting [bought forever ago... was likely to happen eventually].
What a stupid argument (Score:1)
How retarded do you need to be to come up with this nonsense?!
Simple solution (Score:2)
> Scrolling through social media feeds for hours can 'push' hundreds of videos to the user,
I just turn off DRM. And now the advertisers are afraid that I might seal their precious ad copy. So they pop up a warning "This video cannot be viewed without DRM".
Great! Job done.
Fuck off (Score:5, Insightful)
Get this corporate mouthpiece bullshit off /.
Re: (Score:3, Insightful)
Seriously that "just asking questions" vibe of that non-headline is infuriating.
It's bad if someone is getting paid for this, and worse if someone is so incompetent as to do it for free.
Re: (Score:2)
Yes.. Why are we asking the data cap software company their opinion on data caps? Their product has no purpose if Broadband providers simply invest the money in the network instead of spending an arm and a lag instead of on software and services designed to mitigate the lack of capacity.
Regulate those who over-consume (Score:2)
>> companies have unlimited bandwidth they refuse to share
> right, because data caps are a "feature"
Yes, they regulate those who over-consume to make sure those with a reasonable appetite can be served.
Re: (Score:2)
drnb ( 2434720 ) writes:
> Yes, they regulate those who over-consume to make sure those with a reasonable appetite can be served.
I pay for bandwidth. They supply the infrastructure to provide what I paid for. Now they want to be paid more for giving me less. They also want to be paid again for that same bandwidth from those I am connecting to. So they get paid three times.
That is essentially what you are advocating for.
People naturally fall into tiers of usage ... (Score:2)
> Get this corporate mouthpiece bullshit off /.
Really? Is it so hard to believe that people can be segmented into tiers representing their natural usage and pay accordingly?
The real argument is whether or not the tiers offered represent these natural levels.
Re: (Score:1)
> Get this corporate mouthpiece bullshit off /.
Lets hear all about your internet co-op.....