If AT&T Puts A Meter On Your Broadband, But That Meter Is Grossly Inaccurate, Is That Meter Really There?
from the questions-to-ponder dept
With AT&T introducing its new metered broadband plans, one of the key parts of the plan was an actual digital meter to tell you how much of your quote you used up. While we can argue about whether or not this acts as a massive disincentive for use, one thing that I don't think can be debated is that if AT&T is going to offer such a meter, it really ought to work.Tragically, it doesn't.
Instead, early reports suggest that the meter is often wrong -- sometimes by tremendous amounts. Now, if AT&T is relying on the same metering technology to make its determination for whether or not people go too far, it seems likely that a lawsuit will be the eventual end result. How can a company like AT&T move to metered broadband with meters that just don't work?
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Reader Comments
Subscribe: RSS
View by: Time | Thread
New Business Model
In California the new 'smart' electric meters occasionally double or quadruple persons electric bills for no good reason--and they're forced to pay or go without.
In Georgia (I think it was) the exact same situation is occurring with the new 'smart' water meters.
So, 'smart' internet meters will undoubtedly go the same route--and the customers will be forced to pay or go without.
[ link to this | view in chronology ]
First, a bit of background. Before, what Shaw would do is to give out "warning" to people who are consistently over the "suggested" cap. There weren't any specific cap on the agreement, but internally there are (60GB/Mon for most cable modem users). For those people, they'll also turn on this internal application, which they claim to use internally, that shows Up/Down traffic for a 12 month period.
As someone who do heavy surfing, watch/dl lots of video etc online, I was concerned although personally I never was warned. I have DUMeter which tracks locally on my computer and a router that tracks usage. Out of curiosity (and caution), I asked customer service to turn on this internal application for me so I can track my own usage from their end.
It was never, ever, accurate.
We are not talking about 1-2GB difference. Those I would accept as margin of error on a 60GB theoretical limit. I am talking about double digit differences in GB, from month to month, constantly. During the 4 years I have this thing on, there was never a month that was accurate.
Now, with this kind of track record, how, if ever, can Shaw make me believe they can track my usage correctly and not charge me extra for data I've never used?
Now, margin of error is acceptable if this is only used to warn people, but what they were trying to do with the cap has money on the line. Shaw can probably get sued for charging for services that was never delivered.
[ link to this | view in chronology ]
Re:
If so, perhaps they're counting bits that you're not counting (not that they SHOULD be counting them).
I would expect from your comment that the number was some random difference from yours, but then if the number was randomly different in BOTH directions, I'd expect that 1 month out of 48 you'd randomly get something CLOSE (hence my first question).
If you have any real numbers, I'd be very interested in seeing them, even if it's just a few months' worth (I'm a data geek).
[ link to this | view in chronology ]
Need for government oversight
[ link to this | view in chronology ]
Re: Need for government oversight
/sarc
[ link to this | view in chronology ]
Re: Re: Need for government oversight
[ link to this | view in chronology ]
Re: Re: Re: Need for government oversight
Err, not if they want your repeat business they don't. The case you're describing only happens when the company has a monopoly and doesn't have to give a crap about customer satisfaction.
However, the main cause of monopoly creation in a market is . . . the government.
[ link to this | view in chronology ]
Re: Re: Re: Re: Need for government oversight
Funny thing about gov't granted monopolies. They really don't have to care about you - you still have to pay them or do without.
[ link to this | view in chronology ]
Re: Re: Re: Re: Need for government oversight
Funny thing about gov't granted monopolies. They really don't have to care about you - you still have to pay them or do without.
[ link to this | view in chronology ]
Re: Re: Re: Re: Need for government oversight
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
Government often strikes deals with these companies to allow only them in a particular area. This isn't a condemnation of companies as much as it is the government for removing the consequences for bad customer service.
[ link to this | view in chronology ]
Re: Re: Re: Re: Need for government oversight
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
Telephone and cable are usually monopolies as well, but, for a number of reasons, are not regulated as such. Then when broadband came along, they ended up pointing at each other as "competition". In reality, they usually don't behave competitively at all, which is hardly surprising for duopolies. Again, there was some government involvement in all of this. However, it would have ended up this way without any government intervention anyway. The Devil's Coachman is right, the *only* thing that could have effectively stopped it is the government. Yet, the liberals screwed up the regulations and the conservatives decided that a "hands off" approach was better. Bad government can be just as worse off as no government at all! This keeps happening because rarely does anyone impose the third option--EFFECTIVE GOVERNMENT!
And now it looks like the cell phone industry, which is merely split between two standards, is poised to fall into the same de facto duopoly.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Need for government oversight
Politicians screwed up, the whole lot not only a subset of them.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
I think, as Bastiat would say, you are not taking into account what is unseen. :)
It's easy to point at big mergers and immediately conclude that markets always consolidate into large monopolies and that the government is the only thing that can stand in their way, but that doesn't take into account the wide variety of other ways that the government impedes or removes competition beforehand so that only the large companies are left. See: Regulatory capture.
Usually, when people support regulation to "stick it to the big corporations", they only end up ensuring that only the big companies remain.
"We're tired of these big corporations selling dangerous toys! We demand that every new toy go through a million dollars worth of safety testing before it can be sold! Also, why is Mattel the only toy company? Stupid market; all it creates are monopolies!"
[ link to this | view in chronology ]
Re: Re: Re: Re: Need for government oversight
You, sir, are probably not a student of economics. Here's a short list of the MAIN causes of monopoly creation in a market:
- scale
- concentration of power
- purchasing power
- control of input supply
- exclusive deals
- natural monopoly (such as infrastructure)
- M&A
- regulatory capture
- anti-competitive behavior
- dumping
Free markets surely don't need the government to form monopolies. In fact, many markets have a natural tendency towards monopoly. Governments play an active role in both creating monopoly (ex: tariffs) and busting apart monopolies (ex: DOJ, AT&T)
Competition is wonderful, but it is not the natural consequence of free markets. Competition must be curried and managed with great care.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
...
Free markets surely don't need the government to form monopolies.
While I can think of several examples of monopolies created as a result of government interference, I'm having a hard time coming up with any to the contrary. Could you please give some good examples?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Need for government oversight
Do you know what happened since then? It has been re-consolidating steadily since.
In the US, we started out with hundreds of cellular phone companies, because our gov't wanted it so (not that this was a good thing or not, just that it was the case). Do you know what happened to those hundreds of small cellular companies?
Ever hear of Microsoft. No government created it, but they have surely tried to break it apart. It was a near monopoly.
The market rewards players that have power and scale economies. These big fish eat the smaller fish to retain and grow those advantages. Monopolies don't need a government to make them happen, the market is capable of creating them, too.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Need for government oversight
A monopoly does not create an incentive to break it, it creates dis-incentives.
If there were a monopoly on, say, garbage removal in a given country, then one trash firm would have scale economies. They would have brand awareness, retail locations, sales staff. They would buy their trucks at a discount because of volume. This monopolist currently has ALL of the customers that buy trash removal.
Now, say you are a new entrant. You could invest your capital in trash removal company to compete. You would pay more for your trucks, you would have no brand awareness and need to invest in it. When you launched, you would have a higher cost basis, and no customers. You need to win customers from the incumbent. Oh...and when you try, the incumbent lowers prices on the accounts you're trying to win. He can actually afford to lower them below HIS costs (which are lower than yours) just to drive you out of business.
...or, you could invest in something else without an incumbent monopoly and get a decent ROI.
Tell me again how there is an "incentive to break it"?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Need for government oversight
Neither, sir, are you.
Governments play an active role in both creating monopoly (ex: tariffs) and busting apart monopolies (ex: DOJ, AT&T)
Heh, good way to argue against yourself. AT&T wouldn't have had a monopoly in the first place if it hadn't been for the government.
I notice someone asked you to provide some examples to support your position and you failed to do so. No wonder.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Need for government oversight
You: "Neither, sir, are you."
Actually, I am a rather accomplished economist. You see, I'm referring to actually having studied economics -- not just thinking you understand it without having every actually done the work.
"Governments play an active role in both creating monopoly (ex: tariffs) and busting apart monopolies (ex: DOJ, AT&T)"
That, sir, is what in debate is called a "concession". I admit to a part of the opposing argument, because I'm an honest person. Of course gov't can cause a monopoly. F'n duh! But my argument is against the wrong claim above that "the main cause of monopoly is the government". In fact, the main cause is that larger scale means more market power.
"someone asked you to provide some examples to support your position and you failed to do so"
...ummm. Because I have a job.
Are you folks of the "the market is perfect" school of economics? Most economists adhere to the "the market is the best we've got" school, which illustrates an important difference.
[ link to this | view in chronology ]
Re: Need for government oversight
[ link to this | view in chronology ]
Re: Need for government oversight
so what..... only pay one 1/10 of the bill?
[ link to this | view in chronology ]
bad programmers
[ link to this | view in chronology ]
AT&T
Things that make you go Hmmm...
AT&T's entire strategy for making money.
Throw money at a problem
Find someone else to fix problems
???
Profit!
[ link to this | view in chronology ]
ISPs somehow get a pass on it. and I no more trust an ISP to police their own meters than I would the gas station.
[ link to this | view in chronology ]
Why bother metering?
[ link to this | view in chronology ]
Re: Why bother metering?
True, but is it more profitable?
[ link to this | view in chronology ]
Re: Why bother metering?
There are costs involved for higher bandwidth users.
More high bandwidth users you have on a network the fewer people you can have on that part of the network or the more you have to spend to upgrade that part of the network.
These are real costs.
I am not saying that ISPs are not complete and total fuckwads. The problem though is that when idiots like you rail against them with shitty arguments that do not hold water it makes the bad guys look good.
So do us all a favor and choose from the following.
A: STFU
B: Educate yourself.
C: Argue for the other side.
Thank you.
[ link to this | view in chronology ]
Re: Re: Why bother metering?
[ link to this | view in chronology ]
Re: Why bother metering?
They need to upgrade the bandwidth coming out of the CO. The cold hard truth is it's much cheaper for them to charge for bandwidth, than to upgrade the network.
[ link to this | view in chronology ]
Also, for the "up to" marketing tactic. I would like to see a ruling saying that if you want to use "up to" in your lingo, that means that service under that number is appropriately prorated. For example, if an ISP offers internet up to 10mbps for $50/month, then if on average the monthly service was actually 5mbps, you pay $25/month. As in: "I'll pay 'up to' $50/month for 'up to' 10mbps."
Of course, in real life, regulation would probably not work out that way, but you can't blame a guy for dreaming.
[ link to this | view in chronology ]
Re:
yes, i can..... GO AND PROTEST
[ link to this | view in chronology ]
Re: Up to...
Attenuation, line quality, distance from the CO, even things like the jacks in your house can cause problems with speeds and connection.
The company I work for doesn't throttle or cap, or meter bandwidth, we turn it up as fast as it can go. We offer up to 8 megs, I've seen people with 20megs.
The question I have is if you get pro-rated for lower speeds does that mean you pay more if your speeds are higher?
[ link to this | view in chronology ]
Re: Re: Up to...
No, it's a marketing scheme because they still charge for the higher speed even if they don't deliver it.
The question I have is if you get pro-rated for lower speeds does that mean you pay more if your speeds are higher?
No. The service provider shouldn't provide higher speeds than they are willing to supply for the agreed upon price. To try to backdoor charges in that way would be as dishonest as delivering a lower speed while still charging for the higher one (which is typical).
[ link to this | view in chronology ]
Broken meters...
One day I watched 2 hrs of Netflix (TV shows) and used 2.5 GB of data (according to their meter).
The next day I watched 4 hrs of Netflix (2 high def movies) and used 0.98 GB of data.
Pardon me, but... huh?
[ link to this | view in chronology ]
metered plans
For the last few years my Grandmother did not have a computer or at times a phone to contact us with. So for her Birthday this year we got her an Ipad and set up a data account in our name. I was concerned about the data limit already as this was going to be her primary connection to the internet. What makes me cringe is that my mother got her set up with Hulu+ and installed both it and the netflix application on the device.
I am already imagining the first months bill.
[ link to this | view in chronology ]
Re: metered plans
mmmm...cookies :)
Try bribing a neighbor, it may work and save you $100s/yr.
[ link to this | view in chronology ]
Wonderful tool for nerds.
I like the graphics they are so beautiful.
[ link to this | view in chronology ]
Wonderful tool for nerds.
LoL
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Re:
I, unfortunately, doubt I'll live to see it. Which is shame because I'd probably give my life fight fighting against corruption given the chance.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Oddly enough...
[ link to this | view in chronology ]
AT&T's meter actually exists.
-Hire 3 college students, all in need of extra cash and pay them minimum wage. Their task is to use the internet as most college students would, sans the porn sites.
-Two lawyers, because no one else can do it, watch intently at the sites they visit. If there's any streaming video, the meter is charged double, because video is not a web page. If the webpage loads within 2 seconds, it's billed at a rate 10x the data transferred, because a rounding-up is needed to justify the charge.
-Once the lawyers are done "calculating" the "average", the usage is cut by 1/7th the total amount, and this is now the true cap. The customer cap is not determined until the day of billing, to which prices will range from acceptable (because customers are on the most expensive phone plan) to actual (cheap-ass phone users), and thus the overage fee applies.
-When customer complain, and band together for a class-action lawsuit, AT&T will not admit any wrong-doing, but will "temporarily remove" the cap data plans and for punishment, will offer customers a low cost data plan so buried on its website, not even the web programmers know where it goes once they hit the "publish to web server" button.
Wait. Why does all this sound so familiar?
[ link to this | view in chronology ]
Oh wait, didn't we address this last week?
"Why bother competing if you can just get a patent instead? Nielsen, who was once sued for violating a patent ("5675510") for the invention of a "computer use meter and analyzer," has now sued competitor Comscore over the same patent, which it eventually came into possession of after settling the lawsuit by buying the patent from Jupiter Media Metrix. Once again, however, this seems to demonstrate the pointlessness of the patent system. Metering and analyzing internet traffic is something that came about because everyone needed it and lots of people tried to tackle the problem. No one went into that space because there was the availability of a patent. People got into the space because there was a need and a market. And, now, Nielsen seems to want to beat the competition not through competing with better products, but by suing over a questionable patent."
[ link to this | view in chronology ]
Re:
Oh wait, didn't we address this last week?
Um. Totally different type of metering.
And, if you think that Nielsen's patent covers *accurate* metering, you've clearly never talked to anyone who runs a website.
[ link to this | view in chronology ]
Re: Re:
And, if you think that Nielsen's patent covers *accurate* metering, you've clearly never talked to anyone who runs a website."
Ok, cool, I'll totally trust you on that.
I have another idea, why doesn't someone invent a meter that works in this context?
But let me guess, when someone does invent a meter that works, we'll all be pissed because they patented it, right? Because hey, everyone is in the field trying to make a meter that works because there is a huge market for such meters and we totally need them.
To be truthful, these "software" meters probably ought not be patentable. But hey, if they require a box at your house to meter it then I see no reason why that box and whatever parts are inside should not be patentable.
I note that such a box might already exist and may in fact already be patented.
[ link to this | view in chronology ]
Re: Re: Re:
[ link to this | view in chronology ]
The thing to remember about metered usage is...
For example, if you have an ADSL modem plugged into a router, your usage statistics counted by your router probably don't include the PPPoE encapsulation overhead.
So when you transmit a 1KB packet, and receive a 1KB packet, while your router says you've sent 1KB and received 1KB, the ISPs devices at the exchange could see a different data amount. For example, if the PPPoE encapsulation requires 100bytes, the ISP could have you metered as having sent 1.1KB and received 1.1KB. Spread over several million packets, this could have a huge discrepancy when comparing your routers stats vs the ISP provided stats.
Also, your router probably doesn't count dropped/lost packets. If you are receiving more packets/second than your router can handle, it'll start dropping packets. These packets won't be counted by your router as data received, since it's dropped them. But whether your router accepts them or not, they WERE sent by the ISP down your pipe. It's not THEIR problem your hardware was unable to receive them. They sent 10 million 1KB packets down your pipe to you, it doesn't matter if your router accepted only half of them. As far as they are concerned, that's 10MB they've sent you, not the 5MB your router says it's received.
Other situations include the fact that usually a well configured firewall drops any packet sent to you that is not part of a session initiated by you. Unsolicited ping floods, someone probing your system for open ports, even a full on DoS attack could all be dropped by your firewall. If the packet is dropped by your firewall, again in all likelihood the data isn't counted by your router as being received. However since the ISP did send it down your pipe, again they would have counted the data as sent data to you, even if you've rejected it.
I'm not defending the ISPs or excusing deliberate miscalculations, or even those caused by buggy software.
However, there usually is a discrepancy due to where in the pipe the 'probe' is placed to count the data.
I'm in Australia, and here an unlimited plan is a rare beast. I think I've been on some sort of quota for the last 8 or 9 years now. Some good, some bad, some horribly overpriced. So I feel the pain of being quota'ed.
[ link to this | view in chronology ]
Re: The thing to remember about metered usage is...
Should be 10GB and 5GB...
[ link to this | view in chronology ]
Re: The thing to remember about metered usage is...
Depends on what you are using to count those things, but your router can see droped packages and can count them, also the little overhead can't explain gigantic discrepancies alone and can be easily accounted for since your computer is capable of counting dropped and received packets so you can adjust the numbers and you will see it still is wrong.
[ link to this | view in chronology ]
Re: The thing to remember about metered usage is...
i don't pay for spam mail delivery, i sure as hell wouldn't want to pay for these too
[ link to this | view in chronology ]
Re: The thing to remember about metered usage is...
"Unsolicited ping floods, someone probing your system for open ports, even a full on DoS attack"
Now that is an interesting angle I would not have thought of. Cellular providers were banned (are they still?) from selling lists of cell phone numbers to solicitors and can only send you free text messages because you have to actually pay for that activity on your cellular phone. It is (rightfully) unfair for them to charge you for incoming phone calls and then sell your phone number to a bunch of telemarketers to start calling you.
Will metered broadband lead to internet providers being required to block unsolicited traffic from costing you money? Can they charge you for the bandwidth used to download messages they have sent you? I'm actually going to send this to my state attorney general to see if I can get an answer for CT residents.
[ link to this | view in chronology ]
Re: The thing to remember about metered usage is...
Would that be like the phone company charging you for calls you didn't receive? Hey, it's not THEIR fault you didn't answer, right? Or maybe they could just stream some unsolicited videos to your phone, even if it's turned off. Again, they WERE sent. It's not THEIR problem your hardware was unable to receive them.
I could even extend the idea to non-electronic deliveries. Just send some merchandise to people whether they ordered it or not and then demand payment. I mean, the packages WERE sent.
Well, I didn't actually come up with that scheme myself, it's an old one. And, oops, federal law these days says you don't have to pay for unordered merchandise. Maybe there's a reason they made that law?
I'm not defending the ISPs or excusing deliberate miscalculations, or even those caused by buggy software.
However, there usually is a discrepancy due to where in the pipe the 'probe' is placed to count the data.
1. Establish cap and overage plan.
2. Flood customer with unrequested packets.
3. Profit!
[ link to this | view in chronology ]
[ link to this | view in chronology ]
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Not Really "Metered"
Except, it's not really metered service. It if were, you'd only pay for what you used. Instead, it's a cap and overage scheme. Big difference.
[ link to this | view in chronology ]