Choosing the Wrong Lane in the Race to 5G | Hacker News Reader

Choosing the Wrong Lane in the Race to 5G

wired.com

51 pointsOsiris3056 comments

High band internet just isn't very practical as a replacement for wired internet in most of the country, but everyone wants ignore our broadband infrastructure issues because 5g will save us.

Using smaller cells is great for servicing more people on the same frequency. When signals don't go as far, you have fewer other people to contend with for space to send. In the real high-density places, this makes a lot of sense. If you can stick with bigger cells, that is obviously cheaper.

In the majority of the country that isn't high density, high band doesn't make sense and point of the article is that there isn't enough mid band frequencies being auctioned off.

Did anyone else think there was a race?

How about just 4G? Can we get consistent nationwide 4G? I'd be pretty happy with that.

In my tiny EU country, 4G already covers 99% of the population and unlimited data plans for homes are available from 19 EUR / month. Yet, nobody is talking about 5G or the urgent need for it.

> In my tiny EU country, 4G already covers 99% of the population and unlimited data plans for homes are available from 19 EUR / month.

Mine as well - in peak hours you're lucky to get close to 1MB/s in two bigger cities I've lived in. Also latency spikes even when the bandwidth is OK. Don't know how much 5G will help with all of this but 4G is not a good option as a primary internet connection if you work online and any upgrades would be very welcome.

That sounds like the backhaul is congested, which 5G won't do anything about at all. The good news is, it's likely much easier to upgrade the backhaul than it is to change up all the radio hardware - it's simply a decision your mobile provider/ISP has taken to not do it.

> Mine as well - in peak hours you're lucky to get close to 1MB/s in two bigger cities I've lived in. Also latency spikes even when the bandwidth is OK.

It's the same in my country.

> Don't know how much 5G will help with all of this but 4G is not a good option as a primary internet connection if you work online and any upgrades would be very welcome.

The thing is, very few people work from home, and even those who do rarely require more than 1 MB/s for the most common office tasks.

It's true, though, that 4G connection is not the best option for 8-5 software engineering. But it still makes it possible to work from anywhere in the countryside during the entire summer.

In my average EU country 4G has also great coverage, but for USA that is still a challenge: there are many low density rural areas that are too expensive to cover, so nobody does it. That is why getting 4G coverage there would be more relevant than going for 5G in a few cities.

In my G7 EU country, I regularly go out of not just 4G, but 3G coverage. E for Edge is quite common to see, which with current apps and sites effectively means no data. OK, I'm relatively rural, but doesn't require me going far out into the wilds, or even out of town.

It wasn't much better when I was last in a major city. Regular, though admittedly brief, dropouts of coverage.

No doubt in both cases the mobile company's map says great coverage. I'd settle for pervasive 3G or 4G long before caring about 5.

In a lot of the US you don't get an "E for Edge" you get a "No Service." I was just in Mansfield Ohio and my freind's T-mobile phone was more or less useless and had to be tethered to mine the whole trip, mine (which uses one of the two larger carriers with better rural coverage, mostly because of where I grew up) worked enough that we could occasionally get text messages out, we often had to find a cafe if we needed a new route or map tiles though.

Germany?

In my large East Asia country-China,4G already covers 99% of the population（1.3 billion） and unlimited data plans for homes are available from 129 rmb / month（nearly 17 EUR） for China Unicomm user.Yet, everybody is talking about 5G or the urgent need for it :-)

For the companies, academics, and countries putting time and money into developing the technologies for 5G, yes, it is a race.

I wish it were more of an archery match. How close can we get to what's right through calm carefulness and skill? Races are full of compromises and waste.

When AT&T had a monopoly, they had cell phone technology available in the 40s. It was not pushed because the thinking was that only rich people would ever need it.

Careful plodding thinking and planning ruled the day, and it took 40 years.

https://reason.com/2017/06/11/we-could-have-had-cellphones-f...

According to your article it was due to a lack of licenses, similar to the fact that there are no community run cell networks (like WiFi networks) due to a lack of licenses.

Unfortunately, we've put a giant distributed paperclip optimizer in charge of resource allocation and it really likes races.

There is a technological race. 5G will provide infrastructure where new businesses and can companies emerge.

There are tons of places in the US that are blanketed with LTE. I don't see "new businesses" emerging. Taking credit cards? Electronically managing inventory, payroll, and scheduling? You don't need much more than dial-up to do 99% of those things. In fact many large Brick and Mortar chains are still on AS/400 and do fine.

without 4G, there is no Uber/Lyft, Instagram, whatsapp, cashless etc mobile revolution. So, don't stay in your cave.

Uber/Lyft, whatsapp, cashless payment apps all work reasonably on 2G and fine on 3G. Real time video calling works great for me over 4G/LTE, in the US, China, Argentina, everywhere I’ve tried it.

I’m all for faster/better mobile Internet, but I keep seeing these claims that there’s going to be entirely new business models because of 5G and I just don’t get it. Better coverage for 4G/LTE seems like it would do a lot more good for businesses.

They can’t trick you into upgrading your phone with 5g dreams that way. 5g is what a new version of Windows used to be for computer manufacturers.

None of the things you mention require 4G. Heck, we were doing social media and messaging before 3G even existed...

Uber and Lyft exploit drivers and are only around because they can finance their losses through venture capital. Which they only get because the promise autonomous cars that will never come and because the taxi industry was broken due to lack of working regulation.

All instagram does is enable influencer and bullying.

Whatsapp is just another way for facebook to hover up all data about you and intrude on private conversation to rent out your eyeballs to advertisers.

Cashless is not a desirable state. And is much more driven by credit card companies that hope you overspent once you don't handle physical money and that hope to make a side profit by profiling the hell out of you.

So if that was the only thing that 4G provided I would be happy without it in my cave.

all concepts you mention aren’t conceptual problems. the world is better with the concepts of technology-assisted taxi hailing, and not having to carry cash (i think you’re basically alone on this one; this is the biggest and best technology change in my life in years)

regardless of your opinions of the specific problems with the current platforms, they’re not problems with the concepts... and those concepts are enabled or improved by high bandwidth mobile internet

and regardless, you’re kinda missing the point: 5g will absolutely enable new businesses (and more importantly new concepts), because it’s happened with every single bandwidth upgrade we have ever been through! you can’t conceptualise them now, because nobody has yet! if you could, you’d have started uber before uber, google before google, etc

5G solves different problems.

Panasonic(at a trade show) says they can do 'next gen GPS' with 5g. Basically better location. This could be the solution for the winter problem in self driving cars.

If (as the article claims) you have to put an access point every 700 feet, I guess location isn’t very hard to determine!

Anybody can use a differential GPS currently, you put the GSM base stations in survey mode and you make them broadcast their L2 phase.

I know there are networks doing that in some parts of the world, you have to pay to decrypt the signal they broadcast.

GNSS locations can be accurate to within a few centimeters using the right equipment as long as you have a clear view of the sky. 5G isn't going to do better, and even if it could that wouldn't be sufficient for autonomous vehicles to operate in snow.

The military owns that part of the wireless spectrum. The DoD did a study which said the military should cede control of it for that reason. This is one of the main reasons the US hasn't adopted 5G in the “sub-6” space (3 and 4 GHz bands primarily).

The strategic problem is that the US will be fundamentally different from the rest of the world, basically ceding 5G ownership to China. No one will want to go along with our broken 5g spectrum adoption and the US will find it hard to get cheap components. Along with a host of other issues like national security, integration with other nations etc.

https://media.defense.gov/2019/Apr/04/2002109654/-1/-1/0/DIB...

One thing this opinion article got wrong is the US govt is trying to make more spectrum available via a plan called 5G FAST. So it's not like the problem is being entirely ignored. There are plans to open up huge blocks of the spectrum to companies.

https://www.fcc.gov/5G

It is likely that the future of 5G in the US is a mix of spectrum, some providing the ultra fast connections for urban areas and mid range spectrum for more spread out areas. But it's looking like we will surely lose the race to become a leader in that space.

> The United States, however, has made zero mid-band spectrum available at auction for the 5G economy.

This is simply not true. Being an employee of Dish at the time, I happen to know that Dish spent over a billion dollars buying spectrum in the 600 Mhz (low-mid) range in 2018 and they were told they must monetize the spectrum by 2020 if they wish to keep it. Dish has a strong hold on rural America and is definitely keeping them in mind when it comes to 5G.

>They are also not penetrating walls or windows, making indoor coverage difficult.

Not even windows? That's just straight up bizarre. I foresee a frustrating future of a billion small blind spots ahead for our American friends. Sorry guys.

Fast internet is cool. Fast patchy internet is not.

Modern windows can greatly interfere with wireless signals. I believe many make use of a very small (barely noticeable) metal mesh which can break havoc with wireless signals.

Manufacturers are likely aware of the issue and test for it but still, I get better wifi through the wall to my house rather than through a large window with line of sight to the accesspoint.

Curious about the mesh, what purpose does it achieve? Strength, insulation, shatter resistance, shatter size control? Are there any documented examples of the technology that I could read about?

All info I've read has been in swedish and don't remember where. But unless my memory fails me the purpose is heat insolation.

A quick google suggested that it can also used for fire safety, to increase strength and resist large temperature differences. But that's not the type of mesh I'm thinking of (it seemed those meshes were quite noticeable).

The phone will presumably drop to 4G inside or switch to WiFi. You can't have it all.

You can't have it all? This quote from the article makes it sound like you can barely have anything:

> In fact, recent tests of newly launched commercial 5G networks in the United States are showing that millimeter wave signals are not traveling more than 350 feet, even when there are no major obstructions. They are also not penetrating walls or windows, making indoor coverage difficult.

350 ft with no walls or windows in the way?? So you'd need about 4 antennas just to cover the vertical height of the Empire State Building? How is this considered in any way viable, even in urban areas. Why not just build out a WiFi mesh network instead if the numbers are that poor.

Really hoping there are some important details missing here.

Short range is a feuture of 5G not a bug, to prevent congestion.

Probably it will not be usefull outside really crowded areas or a set of high throughput sensors that need to be wireless for some reason.

If it's a feature, not a bug, why are all the other countries outlined in the article choosing bands with greater range?

I just can't imagine anyone being happy with 5G if it doesn't work in your car.

I don't really know the trade offs. Maybe they believe a longer short range would be best?

There's a vehicle test track with Ericsson 5G in Sweden (Astazero) where you can stream while driving. I don't know how many radios they use. There are more installations around the globe for showcasing that.

There has been trails of multiple Gbit link to race cars, but that's more or less lab tests. With more cars you will need to share the spectrum.

> Really hoping there are some important details missing here.

The laws of physics dictate that the only way we can have large bandwidth AND large number of simultaneous users is to have tiny cells which cover only a few square meters.

5G, and future Gs will put these micro cells everywhere, on top of each street light, buried under the sidewalk...

Just like in a building every apartment has it's own private Wifi and all of them have fast Wifi, while at the same time all of them have mediocre mobile connection.

It could have some serious niches. For example sensor/camera networks on an airport site. Or as a cheaper alternative for short range microwave fixed links.

I'm actually quite happy with 4G. I can't watch 4k streaming videos on my phone... Which I don't want to do anyways. I don't think we actually need another iteration.

Going with millimeter wavelength in cell signals could result in some interesting ancillary uses. Reminded me of this: https://www.youtube.com/watch?v=uQm_7aPjBUM

It would be great for deploying thousands of cheap surveillance devices. A camera on every street lamp?

This article is strange enough that I suspect astroturfing.

The reason 5G is being deployed in 24GHz is because we don’t have wide enough blocks elsewhere. The 800MHz and 1.9 GHz spectrums have existing services cellular services, and putting 5G there would have to block off spectrum that would otherwise be used by older technologies. This would increase congestion for existing users while very few people would be on 5G.

Unlike other countries, US govt does not force phase out of services—we still have 2G service in many parts of the rural us, and cell companies aren’t going to upgrade them voluntarily.

Eventually most of the handsets would be compatible with 5G, then 800MHz-2GHz bands will be converted over to 5G. This happened with LTE and 3G before that.

Maybe this article is being pushed by telcos, who wants the current TV frequency reallocated to them. Of course the broadcasters want to hang on to those to provide 4K TV over the air service.

TV frequencies have already been reallocated. Channels 52 through 69 (700 MHz band) were reallocated in 2008 and channels 38 through 51 (600 MHz band) were reallocated in 2017.

The 600 MHz band is currently being cleared and will be finished on July 3, 2020.

https://www.fcc.gov/about-fcc/fcc-initiatives/incentive-auct...

According to this:

https://en.wikipedia.org/wiki/2G

2G is already decommissioned for AT&T, and Verizon is this year. Next year is T-Mobile.

My experience is that 2G is nowhere to be found anywhere, and 3G also seems to be being phased out. Places I would get 3G signals last year are now all 4G.

I still occasionally get 2G signal. Mostly EDGE when I am in suburban GA, and occasionally GPRS. I’m on T-Mobile.

It would be awesome if I instead got 3G or 4G signal, but I’m worried if 2G is shut off I’ll be left with no signal at all.

I strongly suspect that they won't let new coverage holes happen. They already have the stations anyway.

There is no 'G' after the 5 in the title (I only clicked the link because I had no idea what the title meant).

It’s basically a joke in Canada: our providers charge so much per gb that nobody compares or complains about speeds.

So much wasted capacity...

I’m guessing that people in the city are effectively subsidizing access for Canada’s many far flung remote areas. Unless the carriers offer separate plans that only work in the city, this is what’s going on.

Updated. Thanks!

Yes, thanks. I didn't click because the title didn't make sense, finally checked it out after the update.

Reply on news.ycombinator.com