Hacker Timesnew | past | comments | ask | show | jobs | submitlogin

> Mobile connectivity plays an important role but it should be complementary, an add-on to the backbone not a substitute.

Why?



Wireless spectrum is limited and thus should be used for stuff that can't be wired such as mobile phones. Using it for static connections that can use wires is a waste.


It’s expensive to maintain. It’s slow. It doesn’t scale. It depends heavily on weather. It can be interfered with by a bad party pretty easily.


Expensive to maintain: dunno, but I pay <10€/month for mine, so it can't be that outrageous

Slow: you can watch HD video ever since 3G came out, and 4G reduced latency a lot

Doesn't scale: that's also a thing 4G apparently addressed. I've never run into capacity issues, just range issues

Depends heavily on weather: never noticed it depending even slightly on weather. I've tested this in heavy rain because I indeed heard that 'straalverbindingen' have issues with this ("beam connection", "directional radio"? Not sure how to translate that word) and because microwave ovens seem to work at 2.4 GHz and water adsorbs this very well, but it appeared to make no difference whatsoever for bandwidth or latency tests on mobile data

Malicious third parties interfering: never heard of this happening to an ordinary citizen, and if you're scared of your government putting up an IMSI catcher, maybe use a VPN


> I pay <10€/month for mine

> you can watch HD video

I wonder... do you primarily watch HD video over this wireless connection, or are you watching it via a more traditional broadband connection? In the US, I pay ~$10 a month for wireless data (across two phones), but these phones are on WiFi 90% of the time and we never watch video on wireless.

Even if you have an unusual plan that allows unlimited streaming at that price (without limiting the video bitrate), it seems like it would probably not scale to most people switching to a wireless plan without a price increase.


Data is ridiculously overpriced in the US. In the UK, I'm currently on a 12GB plan which costs £7.99 month. But you can get completely unlimited data including tethering for £20/month or less from lots of different providers. Never heard of unlimited for less than £15 though (unless its a temporary offer).


“Unlimited” data is not comparable because different subscriber’s data gets assigned different priorities at times of congestion.

In the US, you can get “unlimited” data for as low as $25 per month, or as high as $90 per month, depending on what level priority you want (which you assume might make a difference at peak times).

And then there is a lot of bundling going on like Netflix and Disney or international. For example, I pay ATT $45 per line for 4 lines including all taxes, but they include phone calls and data for almost all of North and South America, minus the Caribbean Islands.


> “Unlimited” data is not comparable because different subscriber’s data gets assigned different priorities at times of congestion.

> In the US, you can get “unlimited” data for as low as $25 per month, or as high as $90 per month, depending on what level priority you want (which you assume might make a difference at peak times).

Yeah I don't think this happens in the UK/EU. The US system seems really weird and complex in comparison. Those prices I quoted above include unlimited calls and texts to anywhere in the UK.


“QCI” was the industry term I was looking for. Here is some unsubstantiated, but true sounding information about the US mobile networks:

https://www.reddit.com/r/NoContract/comments/oaophe/data_pri...

I would be surprised if UK/EU did not also have it. Surely, it exists for certain levels of politicians/government/emergency services. Seems like a short throw to then extending that implementation to capture as much revenue as you can by implementing price segmentation strategies like in the US.


Yeah its illegal to offer plans based on throttling/prioritisation in the EU.

> Under these rules, blocking, throttling and discrimination of internet traffic by Internet Service Providers (ISPs) is not allowed in the EU. There are 3 exceptions: compliance with legal obligations; integrity of the network; congestion management in exceptional and temporary situations.

> All traffic has to be treated equally. For example, there can be no prioritisation of traffic in the Internet access service. Equal treatment still allows reasonable day-to-day traffic management according to objectively justified technical requirements, which must be independent of the origin or destination of the traffic and of any commercial considerations.

https://digital-strategy.ec.europa.eu/en/policies/open-inter...

For the time being the UK is still aligned with EU law on this but there's lobbying to make things more like the US:

https://cei.org/blog/how-the-uk-can-reform-can-improve-inter...


Interesting, but I wonder if this exception would not allow the same:

> congestion management in exceptional and temporary situations.

I presume the US mobile networks are primarily selling higher priority that applies (since it only needs to be applied) in exceptional and temporary situations.


I'd say the EU are generally less tolerant of loophole based exploitation of the law than the US (see for example the potential Apple USB-C speed throttling in the news recently). Anyone who tried this kind of thing would get sanctioned pretty quick.


I didn't know this was a thing. As someone who's on roaming 90% of the time with a MVNO from another country, definitely a good thing to know about! Thanks for linking.

Also relevant https://en.wikipedia.org/wiki/QoS_Class_Identifier (to know what QCI stands for)


> do you primarily watch HD video over this wireless connection

No, I'm just saying it's been possible ever since 3G rolled out, and tech has only gotten better since then. The person I responded to said that mobile data can't be used as a replacement for a landline, but in practice, 99% of us use the airwaves for this already. We call it Wi-Fi. A lot of mobile towers also first beam their traffic to another tower before your packets meet a fiber.

Of course (as a general rule), the sooner your signal gets to a fiber, the better it is for your connection speed and reliability, but mobile data (with good signal strength) is clearly fine for most purposes. I am waiting for the day where people realise this and mobile data scales up and gets too cheap to meter in the same way as today's landline connections don't charge per gigabyte (I'm ignoring the existence of Belgium here; pardon me, belgians!). For now, having a landline dug to your home and work plus paying for mobile connectivity is cheaper than going mobile-only and getting a practically unlimited data bundle. A lot of rooms are also sandwiched in a way that outside signals aren't as reliable as a local WiFi. I'd be surprised if the former is still the case in 15 years, and the latter in 25 (thinking of something like a femtocell here).


> Expensive to maintain: dunno, but I pay <10€/month for mine, so it can't be that outrageous

I agree with this and all your other points. Surely it is cheaper to a put up a few mobile masts than to dig and install cables everywhere. In fact, this is the exact reason the government is using this strategy to get internet to rural locations rather than using cables.


Except rural electrification is a thing and putting a fiber cable on the same poles is not expensive.

The reason you’re seeing that happening is because most states ban electric companies from also providing internet.

I live in a rural area. My state doesn’t ban this. My internet comes from my electric company.

Those “radios” you put up are far more likely to fail than glass.


You’re paying in euros. Which means you’re probably from a small country which also partially subsidizes infrastructure cost.

You’re also, in the context of the macro population, not using as much bandwidth as you would if people stopped relying on physical connections.

The weather dependency depends on your distance from the tower.

That’s not what I’m talking about? I’m talking about a coordinated jamming attack on cell towers in an area. Just because you haven’t heard of it doesn’t mean it’s something that the intelligence community is unaware of as a possibility.

If the majority of the population only had cellular comms, the chance of this attack goes up to devastating effects.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: