> "Sometimes providers boost speeds for households during the actual FCC speed-testing period," the Journal wrote. "Comcast a few years ago upgraded speeds in some regions without notifying the FCC, making test results look stellar, people close to the FCC program said. The FCC discovered the changes after spotting anomalous data and adjusted the numbers."
This whole thing is frustrating. It is made to sound like some kids pulling some harmless pranks or something. What the heck. These companies are deliberately lying and confusing government agencies.
They're too busy rolling out IPv6 support-in-hardware and gigabit fiber upgrades everywhere nationally and obsoleting data caps, instead of taking software IPv6 shortcuts and enjoying regulatory capture and turning peering agreements into weapons and maximizing "shareholder value" (which of course we all agree isn't the real value at hand).
Seriously though, ATT was a government entity that got its huge advantage from being a monopoly. After they were forced to "split up" (wink wink nudge nudge) they then shed all their unprofitable business and recombined again as new "ATT".
At no point have they ever really become a customer-focused company and all their decisions are steeped in monopolistic style thinking.
I have had AT&T Gigabit Ethernet for 3 years. It’s $70 a month, no extra fees, no caps, and I consistently get 900+ u/d.
Compared to cable companies it’s a dream. Comcast has a maximum upload of 35Mbps on their highest plan, it’s more expensive, less reliable, heavily over-provisioned and the bills are full of junk fees.
Even at the new price of $60/month, AT&T Now was still a better deal than Comcast’s cable offerings - again with junk fees, hard to cancel service and cable box rentals.
Their hilarious pranks are linked to them receiving many billions of dollars from the government. Perhaps one of the largest dollar amount frauds out there today.
The company i recently started managing it for previously went through some business and personnel changes and so doesnt have a good hold on their IT budget. Ive been discovering and recording bills from only at&t for days now. Highlights:
1300$/month for IT support
800% difference in per month rate vs 1 year contract rate
Different departments cant talk to each other and have separate accounts. I have att.com accounts and businesscenter.att.com accounts. Is AT&T BusinessCenter different than AT&T BusinessDirect? You fuckin tell me
Before i started, we were paying over $10k a month to only AT&T, right now we're <$7k and I'm not done. We have an account manager that i havent talked to. Clearly hes either incompetent or malicious, no desire to loop him in. Never before have i wanted to killdozer a company so badly. I have better things to do than review bills (although clearly not).
I've faced similar issues.. it's almost impossible to do it directly on your own, unless you have enough business to get a real account representative.
This is why there are tons of third-party companies that are run by ex-employees that will manage all of your AT&T billing and accounts for you. There was a one-time fee, but it was more than made up for in what we saved by consolidating and normalizing all of our accounts through them.
It's annoying that I have to go through this extra hoop just to get reasonable service, but it's the only path I've discovered that doesn't make me want to tear my own eyeballs out at the end of the day.
I actually worked on the Business Center application. They are different systems, Business Center is supposed to be the future of Business Direct, which is our legacy product. The idea is that we on-board people onto Business Center, but since not all products and services are ported over, there is still a need for BD accounts.
lol. You are from SoCal or Europe, right? Sorry, not trying to be an asshole. Those are (almost) the only two places where people on hackernews have more than 1.5 internet choices.
New Zealander here. The hopeless state owned telecom was sold off then forced to break up (into a line provider and a separate phone/ISP company). It has taken a very long time to come right, but NZ is now a great place to get internet and many good options are available from a range of companies.
Gigabit fibre is here for a huge portion of residential customers. Connections of 2 and 4 gigabit are scheduled for early next year.
I just tested a wall plate I mounted and am getting 920Mbps, for about US$80.
SF is finally starting to get competitive between gigabit options from Sonic, MonkeyBrains and WebPass. There's also of course Comcast, AT&T and Verizon.
Midwest US..., with locations in several states (and internationally)
While I only have 2 choices for my home, getting Business Fiber runs has been no problems, normally at least 3 companies if not more will provide them. Of the 5 companies that will provide us fiber at main office, only 2 of them also offer Residential Services (ATT and Comcast), the others are B2B only companies, and offer services at a faction of what ATT provides business services for, but considerably more than a Residential line, which is why there are only 2 choices for a home, there is not much money there
Austin, TX. Known for being tech hub. I have two choices. Spectrum or AT&T. Both overpriced with mediocre but acceptable service. Known that we have it better than most of the USA. Your situation is an outlier and not the norm.
My wife and I mostly work at home and our kids are avid gamers so loss of internet access causes havoc, especially for the IT guy (me). I solved most of my problems by just getting the fastest service from Spectrum and a medium speed backup from AT&T.
Comcast told me they'd be able to add my house to their service - no big deal, we promise - until I actually bought the place and called their bluff. New story is that they want $500 for a site survey. And no word on whether it'd be possible to run service. If it is, I get to pay for the wire to buy the privilege of paying them for service. And I wouldn't own the wire. Others a few blocks away were quoted $13k per home. High enough, I thought, to make it obviously-stupid to pay for a survey. Three out of five of my neighbors have Comcast gigabit service.
I can get AT&T DSL. It's not cheap; sticker is $50/mo, and it has more asterisks than a mail-in-rebate form. It's not fast; 20 mbps download max. Who-knows-what upload. So why shouldn't that be included as a data point? It's my only wired option, despite decades of subsidies towards universal service.
Instead I opted for a 4G modem, but now I don't have a public IP address. Didn't expect that, and it's honestly pretty annoying to be unable to ssh home. I just moved in, and this weekend will be getting set up on some hosted VM to work around this shortcoming.
I work for Alphabet in Mountain View. I live less than five miles from my office. You'd think I'd be able to do better.
Beats me! But I suspect even if the answer were, "super awesome!" that it would not be helpful here. I think their mission is to try to go from absolutely-nothing to basic-cell-service. I've got decent cell service, so things aren't awful. They're just lame and disappointing.
Every encounter with AT&T has felt like a scam to me. I tried a mobile hotspot two years ago on a 14 day free trial. One bar of signal, so I returned it the same day I received it. It took them 14 days to "process" the return, and I ended up with a $285 bill (activation fee, two months service). Last year, I was trying to find a carrier that would work with my Sierra Wireless WLAN card, and ended up repeating the same horrible experience - this round cost me $180. Next, I tried Cricket Wireless, which uses the AT&T network. No activation fee, $35/month, works flawlessly. Its not the network, its the company.
I am not a fan of AT&T, but it appears what they didn't want included was their ADSL1 speeds, which are probably utter rubbish for obvious reasons. The problem here is that there are people still on those plans.
AT&T: "We don't want these plans included because they're obsolete and they're not the ones we are actively marketing".
Also AT&T: "But we are absolutely selling them, obsolete or not, and signing up new customers in areas where that's their option, because we've decided it's not viable to upgrade them. We just don't want them counted."
Is there a good solution for areas where it will never be profitable to run lines?
I mean my company is getting a big local subsidy for a short distance point-to-point connection and it’s still costing us about a million dollars for construction and we still have to pay a monthly maintenance fee.
I feel like this is some form of Stockholm Syndrome. You shouldn't pay more. You should be getting better speeds for the same or lesser price. Technology is advancing and the carriers are purposely holding us back, for exactly the reason you gave.
If ADSL is cheap I actually would prefer to have it for redundancy. Phone line and cable is generally the only lines that are routed to a home that can be used for Internet.
You sure? We tax payers have given it $Billions and AT&T hasn’t met its obligations for that money. I bet if one were to trace it, it was funneled to rich shareholders.
Why? We already paid them $400 billion dollars to get you decent Internet access (see: https://www.huffpost.com/entry/the-book-of-broken-promis_b_5...). I think that gives us the right to expect you to have something better than 14Mbps DSL at a decent price.
Absolutely! Even ignoring all the stimulus money and tax breaks probably given, municipalities all over the US are building out their own fiber that is blowing the others away in performance and even price usually.
Who said anything about you having to pay more? At least here, in the market that I can choose between FTTH and ADSL1 from the same provider, there's a 990 Mbps difference for the _same price_.
Though - my personal opinion is that old copper that no one really wants to maintain is wasted space for cabling that could support significantly higher bandwidth at lower latency. Telcos are already refusing to replace destroyed landlines and instead substituing LTE service or otherwise instead for a decade now.
> ISPs don’t make 75% profit margins, so that’s really not possible.
You're assuming the only place price reductions can come from is lower profits and there is no room to e.g. reduce waste or increase efficiency, which monopoly providers have a long tradition of not even attempting to do. Which usually means there is a lot of low-hanging fruit there.
They have been chastised on this repeatedly. For instance, "The FCC Had To Remind ISPs Not To Spend Taxpayer Subsidies On Booze, Trips To Disney World" (https://www.techdirt.com/articles/20151022/09232532594/fcc-h...) here, the $1,300,000 spent by an executive to buy a house for their children's use as college housing wasn't part of the profit. That's simply the cost of running business, as essential as keeping the lights on ... nothing wrong with the twice a week $96,000/year massages either, I'm sure that's got a lot to do with FTTH residential deployment. Oh wait, the taxpayers built that.
This has happened a few times, such as with verizon (https://www.huffpost.com/entry/are-the-salaries-of-veriz_b_9...) where the 4 top executive pay (without benefits) was $41 million ... which would have been fine if it was in their operating expense and they could afford it.
But they did this because it wasn't and they couldn't.
If the telcos make something called "Local Service" unprofitable when compared with "Corporate Operations Expense" (which includes jets, multiple homes, multi-million executive pay, etc) and technically run "at a loss" they can legally raise prices and reduce service.
Because they can make things look unprofitable, federal regulation permits them to raise prices to make up for it. It's an perverse incentive, an executive spends $100,000 a year on pet care to make sure local service becomes unprofitable and then they can use a legal procedure to raise prices. rinse, wash, and repeat.
So yes, it could be 1/4th the price but no, the free market is the most efficient allocator of goods, we're told. They do things the cheapest we're told, there is no alternative ... yeah, bullshit.
High speed internet is $7.67/month in Russia, $5.41 in Ukraine, $6.12 in Venezuela and $63.07 in the USA. Ah, the magic of trusting the market. Can you feel it!?
You're not in the US, so why did you bring up Google Fiber? Also if you're not in the US why are you talking about AT&T being forced to take away your ADSL?
Also the poster you replied to specifically said ADSL1 which has been superceded more than a decade ago by ADSL2, ADSL2+ and VDSL. These aren't fibre technologies and in most countries are sold at the same price.
Then you're likely in a very different situation. A condensed version of the US situation is that telcos here were incentivised through subsidies to upgrade and update the networking infrastructure to deliver more speeds and bandwidth to existing customers as well as connection to more rural customers. By and large, the telcos did not do that in spite of receiving the money.
ADSL1 from at&t tops out at 6Mbps, and 8Mbps generally, so you're most likely on ADSL2 (probably not VDSL, unless you're just plan limited, and not distance limited)
This is an important piece of context, thank you for pointing this out. Exactly - you can't classify things one way in certain contexts, but then want things viewed differently (in this case - ADSL not viewed as broadband so not included in broadband aggregate speeds) when it no longer suits you.
The FCC has defined broadband as 25/3 Mbps since 2015. Prior to that it was 4/1 Mbps. The first FCC definition of broadband was established in 1996 as 200/200 Kbps. You can't really play loosey goosey with the definition. The service is either fast enough to be considered broadband or it isn't. The delivery method is irrelevant.
I wouldn't be mad if the FCC standard was still 10Mb.
But I prefer it increasing the way it has been. The downlink, at least. Minimum uplink should be at least a quarter of minimum downlink, if not higher.
And I disagree that 4K only needs 10Mb. I think youtube uses about that much, and it very much does not reach the full potential of 4K. Netflix uses about 16 with HEVC, with massive amounts of server time spent optimizing every second. And they recommend your connection be 25. And if you wanted to watch a live stream of the same quality you'd need far more bandwidth to make it work.
> The 2017 report includes two categories for AT&T, one for its oldest DSL technology and another for its DSL-based IP broadband with speeds of up to 45Mbps. [...] The 2018 report only includes AT&T's IP broadband category, leaving out the company's worst results.
Good to know, but only if you have a choice of carriers. While I hate to drag the free market into this, having choice (read:competition) would expose this issue and be more likely to solve it.
Reminder: initiatives like Hyperboria (formerly Project Meshnet) exist. Using the cjdns protocol and some communications medium (e.g. wires, line-of-sight microwave, WLAN, IPoCP), you can become the ISP for your own local area, assigning IPv6 addresses and providing a bridge (or many) to the 'net at large.
The rebirth of Community Access TV(CATV). Some enterprising community member would set up an expensive and complicated antenna and split it out to all his neighbors. Those eventually become cable companies.
I kind of hope the next 'net is based on cjdns, because it's a much better protocol and it's fundamentally less centralised. If the new ISPs evolved from Hyperboria, it'd be a dream come true.
There are still some bugs, though (like the horizon issue limiting the usability of convoluted or large networks: https://github.com/cjdelisle/cjdns/blob/master/doc/bugs/hori...) but it's enough that people using the network don't really need to fix them, since everything still works.
If you want to help out with the code, it's probably best to look at the bugs repo (https://github.com/hyperboria/bugs/issues); the issues in the "bugs" directory of cjd's repo are mostly large, architectural issues and non-trivial flaws that are probably due to the implementation.
I'm one of the Yggdrasil developers, and FYI, the cjdns routing scheme is flawed and has severe scaling issues. It will, in all likeliness, be replaced with a variation of the Yggdrasil[1] routing scheme in future updates to the cjdns codebase and route servers.
The existing Hyperboria network already sees issues due to scale today and it's not even 1000 nodes.
Hopefully newer services such as SpaceX's StarLink will break the geographic monopolies or duopolies that have allowed these companies to profit so greatly off of such poor quality service.
Right but the end result of competition is that one company wins and eats the other, or they divide the market geographically, either way you end up back in the same place.
The alternative being what? Gov regulations? Where the companies manipulate the regs and have flat out abusive monopolies? As we have now? I don't see how more of the same is any sort of solution.
Can you truly have free markets in a decades-long timespan? Lobbying to rewrite laws seems like a given for large companies. Free market telecom doesn't seem like an equilibrium.
This whole thing is frustrating. It is made to sound like some kids pulling some harmless pranks or something. What the heck. These companies are deliberately lying and confusing government agencies.