Do We Still Need Telcos (and ISPs)?

eraserewind asks: "Are telecom providers and ISPs going to continue to be necessary in the future? Why are we all paying subscriptions for communicating? What I want is a global extremely-high-speed ad-hoc wireless data & voice network, where the only entry cost is a mobile phone (or newtork card or whatever). Devices communicate peer to peer, or routed via other people's idle devices. Remember there is no subscriptions, so don't expect to piggy-back on someone's paid for DSL bandwidth. What are the technological barriers? What kind of protocols would you need? What hardware advances? How would you solve problems of geographic isolation? Are there theoretical, political or economic reasons it couldn't work?"

Comments

Uh...

[TheShadow]

Yeah, I want everything for free too. Give me a break.

Re:Uh...

[darthtuttle]

1) it's always cheaper to run landline for the highest speeds available.

2) There are great distances between areas where people live. Despite apperances you can't go from DC to Boston through suburbs all the way.

3) Data has to be served from somewhere, and you have to connect that to everyone somehow. Your not going to do multi Gigabit out of a medium sized Data center let alone the big ones.

4) I can count.

6) Even if we got rid of all the companies and did everything as a "community" project people would end up running things and those people would fight for power and the little guy would get charged to much in the end anyway.

7) Whoops, I can't count.
8) Have a nice day

Umm, No Thanks, i like my speed.

[TheCrazyFinn]

So, you want everybody to be restricted by the low-bandwith links common for last-mile today, no fast websites, and non-robust routing?

I don't think you understand the value of redundant OC48 backbones, BGP4 and IS-IS routing, and colocated servers on gigE links.

Your ad-hoc networks would be OK for MAN's (Metropolitan Area Networks), but are simply unusable for anykind of backbone.

2 problems

[jonhuang]

1. freeloader problem--your privately designed cell phones will be replaced with bandwidth suckers that don't do replays. No controlling body, so can't stop it.

2. no "backbone"--hopping accross phones works around the city (maybe), but how many hops will it take to get to.. japan? and don't forget that there's some countable amount of milliseconds per transfer--to get accross the nation is a lot of cell-phone coverage sized hops. Plus, we have to go around the grand canyon.

Re:The tragedy of the commons

[Anonymous Coward]

No, the big problem is that the question is stupid and that it makes no sense.

It's not "the Man" that screws you into paying internet access costs - it costs money to lay wires and run all of the routers on the internet. This is a fact. Wireless infrastructure is stupid on a large area network, as you waste virtually all of your power transmitting to areas where there are no listening machines (or no applicably listening machines).

Why does slashdot continue to let 14-year-olds with dreams of free everything post to Ask Slashdot?

why CS departments teach networking classes

[SuperBanana]

Why are we all paying subscriptions for communicating?

Because infrastructure and reliability costs money(no no, trust me, I get more insightful below. Well, maybe not insightful. It's hard to answer this story insightfully, I just point out the facts.) Communications mediums are WORTHLESS if they are unreliable, which is one of the reasons cell phones took decades to "take off"(realize that it's been at least 3 decades since the cell phone was invented, and only in the last 5-6 has there really been a cell phone boom, at least in the US. Realize that the # complaint with cell phones is still how unreliable they are.)

Devices communicate peer to peer, or routed via other people's idle devices.(snip) What are the technological barriers?

Well, you asked, so here goes:

• Latency- you're multiplying the hop count astronomically.
• Routing- the internet has something of a routing crisis already, with routes being incredibly complex. Now, you've passed the buck to each system or workstation- and it has to know, geographically, where it is and where all the other nodes within range are, so that it knows who to pass a packet to(no sense in passing it to the laptop sitting right next to you, is there?) This might be possible, if the routes were at least semi-permanent, but they're not- they're constantly moving, nodes are going up+down...which brings us to...
• Reliability- systems will crash while handling a packet, or simply never see a packet due to interference- RF or physical(something blocks the signal). That's just on a pure network level. On a higher level, communications are worthless without reliability. You've GOT to be able to pick up the phone and get a dialtone for so many reasons- emergency services, business...
• Speed. Due to extreme unreliability, retransmission will be a severe problem. That means TCP windows won't get very big- and remember how high latency is? That means data transfer rates will be incredibly, incredibly low. Overhead will skyrocket. Even a couple percent packet loss can seriously affect performance.
• Leeching. People will hack their devices to simply refuse to answer routing requests. This is what's happening, basically, on p2p networks...and believe it or not, accounting/policing it is almost impossible without a centralized system.

There are also some hidden consequences, like "everyone's mobile device is no longer idle, it's processing someone else's packets, so its battery life goes into the toilet".

How would you solve problems of geographic isolation?

That's just it- you'd need wires/fiber/something...and that would cost money. But, reliability would be far better- so people would opt for wired connections they had to pay for. Oops, right back where you started.

Also related- the reason high-speed access costs so much money in the US is because of geographic isolation and population density. It's no surprise that several Asian countries have DSL service in the megabyte-per-second range to your door for $10-20/mo; after all, you're probably in a huge apartment complex, in a city.

If the population density isn't high enough to support pricing high speed access low enough, I doubt you'll have enough nodes to even occasionally get any kind of connectivity to anything else- much less guarantee it.

Back to the cell phone example- look at how many billions(if not trillions?) of dollars have been poured into the cellphone network(which in turn is reliant upon a larger wired network.) I don't care what network you're on, soon as you get a little bit beyond the suburbs, off a major highway- forget it, you're screwed.

Are there theoretical, political or economic reasons it couldn't work?

Well, for one, if you did telephone calls over this "system", I'd move to another country. When I pick up the phone, I damn well expect a dialtone, because, oh, say, my house could be on fire. There are no doubt thousands of o

Re:No charge?????????

[no_opinion]

Actually, not that surprising. Everyone here thinks music should be free, so why shouldn't communication infrastructure be free too?

Given the number of "when I download music I'm not stealing because I'm not taking anything physical" I understand why there are people who have trouble grasping the costs associated with non-physical goods (like bandwidth).

An actual ANSWER...

[zachrahan]

The question asked was about (1) possible problems with a totally ad-hoc communication model, and (2) what solutions to those problems may be. Typically, everyone immediately leapt all over the problems, but nobody seems to have any interest in the solutions.

Well, here are some ideas about what you would need to make this work and to deal with the problems.

Problem 1: Freeloaders. Well, you could design a tit-for-tat protocol where you never rebroadcast packets from a freeloader. Think Bittorrent, where if you don't share, you don't get good download badnwidth. The game-theoretic knowledge is there to design an ad hoc protocol where the Nash equilibrium behavior is to not freeload.

Problem 2: Long Hops. OK, so long distance pipes cost money. And they won't go away soon, because some, posibly large, fraction of traffic needs them. So let the operators of the pipes charge tolls. You could have a whole ad hoc marketplace where some people let you use their hardware for free, and others charge. You tell your computer how much money and what QoS you want, and it tries to route effectively.

There are problems here, of course. One is how to establish trust -- how to do billing in an anonymous ad hoc system? Some sort of self-signed certs might be made to work... or maybe we'd need a palladium-ish technology? Either of these solutions can also help with the problem of needing end-to-end encryption on everything.

So there. I've thrown out some solutions. They may have problems, but at least its a start, instead of grousing about the original question.

So I'm a clueless F'in idiot, huh?

[eraserewind]

Well that puts me in my place!

I ask a simple question in the hopes of stimulating some debate. You people are so closed minded. Well, you live an learn. You won't be hearing more from me on slashdot after this post. (are those cheers I hear?!)

Thank you to everyone who answered with reasonable answers either for or against. Before I go I'll answer some of the points people raised.

Land Lines & Infrastructure

I am talking about a wireless network with no central infrastructure, no land lines, just peer devices.

The initial costs of a centralised netowrk are huge. Do you think that operators are going to continue to roll out huge networks after the fiasco that was 3G? (and regular broadband/cable TV in many areas). I think we'll wait a long time before we see those kind of investment by any central organisation again.

The total costs of a distributed network are even more huge. However the cost is spread among whoever wants to pay for their devices. See that FAX story on slashdot from a while ago for an analogy.

Free as in ... peer?

I don't want the infrastructure or services I need for free.

I am not a freeloader or pirate. I am quite willing to pay for my equipment. I just want it to be subscriptionless. The cost of the network is built into the the device and whatever it costs in electricity (at least until I fine tune the cold fusion process & matter replicator that I've been working on that is). If this means $50bn devices as someone mentioned, then so be it ;) Technology prices come down all the time though. How much is an ethernet card today compared to when it first arrived?

Let me ask a question, do you pay a subscription for a bluetooth PAN? For WiFi in your home? Why not? Are you ripping someone off by not doing so? Why not extend the metaphor to communities, or towns, or cities, or the world? I am quite aware that there are problems of scale and many others which was why I asked the question. I wanted to see what you all thought could be potential solutions. Seems you'd mostly prefer to take cheap shots or try to look cool or whatever.

I don't expect to connect to existing networks like the internet, GSM, POTS etc. for free. They are largely owned by private operators and if you want to connect to them they are going to charge you for that privelege. However if you have a no-subscription network out there then maybe web sites, and all those other services that appear on communication networks would start to appear on it, or even migrate exclusively to it.

Spectrum saturation & interference

I don't know enough about about spectrum to answer this myself, so I'll point you at this GnuRadio: MeshNetworks and also this slashdot story The Myth of Radio Spectrum Interference which was featured on slashdot a while ago, and ask it it just BS? They seem to me to be saying that the more nodes in a wireless network, the greater the bandwidth.

Battery life:

This is a problem that is going to take a long time to solve unless there are some major breakthroughs in battery technology. I have no suggestions.

Routing:

Difficult? For sure, but impossible?

You don't have to use IP you know. It's not the internet. I think that it is going to be possible for devices to route to others. I'm not saying it's easy but surely not impossible to at least get a "good enough" algorighm?

I recall reading somewhere about a routing algorithm that was modeled on ant's behaviour to achieve good enough shortest path finding. Is there no scope within this or other areas of research to make advances? Here's a link to one similar paper I found now just to proove I'm not hallucinating: http://www.computer.org/proceedings/icppw/1680/168 00079abs.htm . Use g