Slashdot Mirror
Neighborhood Area Networks?
schmaltz writes: "Recent discussions about long-haul wireless on Slashdot seem geared mostly to benefit institutions, really, until this post on the peer-to-peer-oriented Decentralized list opened my eyes: "What will society do, when there are kits in every computer store and mall, for 802.11a neighborhood routers? What if you could buy a kit with four pole-mounting 15DB directional antennas, and a router in a sealed case that maintains mesh networks? ... There will be a great blooming of local gaming, IM, and voice/video telephony ... a lot of sharing of music and video on these NANs (neighborhood area networks) ... share a 2nd phone line ... we will all realize pretty quickly this is NOT the Internet ..." Maybe NANs could put the telephone company out of business. Seems like the equipment and software are either available or nearly so -can this be done today? I want to build the first NAN AP on my block!!"
I've actually started doing something similar, but with 802.11b. I'll be putting out a flyer in my neighborhood rounding up people to participate. Since just about everyone there are old hippies, the communal thing will probably work.
--- Think of it as evolution in action ---
... a lot of sharing of music and video
Don't let the RIAA/MPAA find out about this. Then your NAN will be sued and shut down!
Hm. Did someone think their webcam was off? Oops, forgot to secure that Windows box? Wow, nice porn horde you've got there...
This all works well, but the FCC, RIAA, et al will all step in and crush those networks. Don't ya think? They have vested interests in local bell cos and won't let them die because we have taken matters into our own hands. Phil
my apartment and the one nextdoor are already doing this with cables. we share a cable modem and have 8 boxes linked together. no complaints so far. wireless would be great for us, but catV is quite a bit cheaper when you work for someone who installs it. heh.
I'd love to find out more about this. My biggest concern is about latency: Ricochet's network was a 'web network' where APs would route packets amongst themselves until they got to a downlink router that had a hardline to the net. This resulted in 500ms latency, making it unusable for gaming and other realtime activities.
I'm also really curious how lookup tables and downlink load balancing would work in a system like this. If we have 50 AP transcievers, 10 of which are hooked to people's home DSL or cable modem hardlines, how will a router know which one to use to access the greater net? What's to stop one from getting flooded while another goes unused?
I'm sure these answers are easy for those folks with extensive TCP/IP networking experience, but it would be great to filter this knowledge down as the technologies and responsibilities for managing routers is falling into the end user's hands.
Kevin Fox
"He was so nice when I couldn't get xyz on the NAN all I had to do was message him and he solved my problem..." : Common comment by Annyonomous Grandma in the NAN era?
Or will this just pull the geeks out of their comfortable corners into social realms they don't want to be in? Will it force the geeks of the world to be more social?
What I'd love to see though... is Annyonomous Grandma taking a hand to the backside of the neightbor hood Script Kiddies... or better yet, DoSing them of the NAN herself!
If I can't see it in Lynx I'm not interested.
A few friends are considering doing something similar in a guerrilla internet kinda way:
We're looking at creating a bandwidth co-op where we lease a small amount of office space, and a symmetric DSL line (total cost, sub-$500 a month, closer to $200). The members get a single UPS'd power plug and an ethernet drop.
You end up with whatever services you want to provide for a nominal startup cost and $15 a month...in whatever OS you want.
"Draco dormiens nunquam titillandus."
This certainly is a terrific idea! However, working as a network engineer at one of the nation's first widely deployed consumer wireless access companies, I'm all too familiar with the expenses involved in building such a solution.
:(
Ultimately, while playing with the technology and the design of your "NaN" would be fun, we live in a world where bandwidth providers will not accept make-believe money for the pipe(s) to your "NaN" router(s). The issue would quickly become:
1. Who among your neighbors would be willing to shoulder the cost of the bandwidth, AP's, router(s), switch(es), and lend the time and expertise in the installation and configuration of same, and
2. Who is willing to face the inevitable slew of legal and/or licensing challenges in reselling or providing bandwidth for free to the neighbors on your "NaN". Are you willing to pay for a T1 out of your own pocket to feed the bandwidth need? If not, and your neighbors throw monthly contributions into the hat, you face a host of very real-world, non technical legal, tax and business issues.
Please don't misinterpret these points! I think it's a neat idea. However we must remember, regardless of the technology available, ultimately the twin evils of Money and Regulations drive the market; free or otherwise.
If an "Internet Bandwidth Commune" is your goal, don't lose sight of the inescapable truth that somewhere, sometime, eventually SOMEONE will have to pay for it.
Scot
THIS SPACE INTENTIONALLY LEFT BLANK.
There's really nothing to setting up a (open/insecure) NAN, provided it's just linked to itself.
Sure, you could have problems with overlapping NANs, with frequency fights, but that's mostly handled silently by the hardware.
Inter-NAN is a little thornier, especially if the hardware becomes commodity items installed by Joe Average. I can easily forsee accidental broadcast loops due to misconfigurations.
The hardest one, however, is actually linking up the NAN to the 'net to get somewhere else, as has been mentioned in every other "setting up a wireless network" article. It's against just about every TOS. Sure, you could try buying a T1 lease, and charging for that... but now you have to track who has paid, keep people from hooking up others on the sly, provide support... in other words, become an ISP.
Now, if we all said "the hell with it, we'll ditch the Internet", and built our own from the ground up (possibly with NAPs at universities, those pesky academics are always giving stuff away for free) with long-run links between towns in a kind of wireless fidonet, then you're on to something. The infrastructure costs on that though... yeesh.
Maybe NANs could put the telephone company out of business.
Doubtful it will put the telco's out of business, but I've always thought that cheap (over public spectrum) wireless is eventually going to be very, very big, and I think 802.XXX is where its going to start.
Like the beginning of ISPs, radio, TV and cable I bet that small, regional Wireless ISPs are going to spring up all over the place (because it doesn't cost all that much money to set up). Some will be non-profit, some will be small for-profit. I think we've just started entering that phase. Eventually the market will mature and there's going to be consolidation as companies start buying each other out, including the big telcos.
This is still good though, because as the tech becomes more refined its going to give traditional telcos, Cell phone companies (who are trying to push G3) a run for their money, and keep them on thier toes.
The real problem that may be encountered with NAN's is when a neighborhood group gets together a NAN (and assuming there are no technical/ fiscal problems) and wants a connection to the Internet.
Okay, fine, it either gets routed through an ADSL, cable or wireless broadband connection, which works great... until the provider figures out that behind their ONE, $50/month connection lurks 5-10 households. I think they (AT&T, LEC's or Sprint Broadband, etc.) will react negativly to this; it violates the Terms of Use agreements I have seen for any of these services, plus they will be losing upwards of $500+/month in revenue.
Boom, the connection is turned down, and/or several nasty letters threating legal action are sent out. Maybe they insist on an upgrade to a costlier business class service, for a significantly higher rate (around $250/month minimum).
Never mind that one person is responsible for the connection to the Internet may move, or have a disagreement with another neighbor, and pull the plug. One could go on about the multitude of non-technical problems that could occur with this type of setup for days.
It would be interesting to see how all this will pan out. But, I do not think it will be more than a niche product until the bigger aspects of this (connection to the Internet, or other NAN's, can be worked out).
Gerald Roebke
Not a Number?
Not a (real) Network?
Nanites against Nanotechnology?
NAK ACK NAK?
news.admin.net-abuse?
Nethack all night?
Nontrivial Address Nodes?
NAN is already overused, and might lead to confusion. I could go on for some time. Why don't we instead call these something like 'Residental Area Networks' or 'LANS for Blocks'.
The next Slashdot story will be ready soon, but subscribers can beat the rush and slashdot the links early!
As someone who's been involved with three different wireless networks, I just don't see this being a threat to the ILECs like you're thinking it will be.
First, I had Ricochet. For what I wanted, it was great. I wasn't tying up my phone line, I could take my portable anywhere on campus and get a connection, and hell, it was a PPP connection, as opposed to the terminal connections my school was offering, for not much more than what a dialup through an ISP would cost. And it wasn't significantly slower, as it was in the days before kFlex/x2/v.90. It had its problems, however, as there were times when I would get some massive latency. The worst location I could keep the modem was in my apartment...other places were great, but the fact that I was in a concrete building sucked ass.
My next network was a bunch of us from my work living in an appartment complex. We had some ISA wavelan cards that one of the guys had, and it connected up three of our apartments, to an ISDN line outbound. For what we wanted, it worked fine. When there was snow, rain, whatever, the connections would flake out, and you'd have to find a new 'optimal' place for the antenna. Of course, then the person with the ISDN line bought a house, and moved out. The new uplink was put in my place, but GTE had changed their tarrifing for residential ISDN lines [and our complex was 19k feet from the switch, so no DSL, which had just come into the area, and they weren't on the public cable system, so that was out, too]. For what we needed, it worked, but it wasn't a sort of 'set it up and forget it' situation.
Now, I've set the wavelan cards up, and I've got line of sight to my neighbor, who's sharing my DSL [business class, not residential] line. It works better having line of sight, but it still flakes out randomly once in a while, and you have to nudge the antennas a little bit 'till you find a good connection [that whole problem with nodes and antinodes in wavelength].
Wireless may solve problems that you have, and you may be willing to deal with issues, but I don't see people setting up a network, and watching other people start hogging their uplink, or probing their boxes, dealing with the support issues, etc. I do think that wireless has good potential in many, many areas, but I don't see it being anything for the telcos to worry about any time soon.
Build it, and they will come^Hplain.
Speaking of 802.11 antenna's, my friends and i at one point had a plan to get an uncapped ("business class") cable modem, and to share it out w/ wireless
The antenna we decided on was the SMCANT-DI135 (warning PDF). It has a 4.5 mile signal thru a 45 degree arc, 7 mile point to point, is 10 inches long, and weighs less than 20oz. We figured it could be put on the side of a house and hidden from view fairly easily, and with 3 of them, we could have wireless access throughout most of our city (it wasn't that big)
Course we never did it, i moved to college, and we're lacking money, but...
~z
sig?
Let's see:
For peanuts, we can set up a NAN on our own block. We could link these NANs gradually, using directional 80211.x (Pringlenet), or even ruby lasers on rooftops.
Why hook this up to the Internet at all?
The Internet is going to be regulated and policed. Hysteria and business interests are gutting the thing 'til it dies and is reborn as a fancier cable TV network.
Build a new network on poles on rooftops. It's cheap, it's fun, it's not subject to regulation (YET).
Eventually repeaters are going to be tiny things you plug into wall outlets, so relaying the signals into the house past the chickenwire/plaster barrier is not a prolem.
Bandwidth? 802.11a has plenty for our needs at the moment, and higher frequencies will give even more capacity. Latency? Well. that's important for web sites and gaming, but guerilla Pringlenets really should be used as a simpler WWW (Neighborhood Wide Web? NNW?) or even a BBS and newsgroup connection.
Why in the world do this? Because newsgroups and web sites are getting censored preemptively by threatened lawsuits; anonymous posting is becoming impossible; EVERYone seems to want to know what we're doing and who we are.
Don't connect your NAN to the Internet. Connect to other NANs... they'll connect to others... and freedom comes back, at least until the FCC and DOJ enforcers come tearing the poles down.
But the DOJ and the various IP owners have already "torn down the poles" on the Internet as it is, so the Pringlenets give a little more time to think of something else (lasers? power lines? quantum encryption over the regular net?).
Someone here mentioned that someone has to pay for all of this, and I say: why? It costs money for the PC cards and for the wireless routers, but not much. And when you buy a can of Pringles, you get not only a directional focus for 802.11, but also yummy remanufactured potato chips.
Eventually the hardware itself will be regulated, maybe, but we get years of grace from the jackboots, and get to have fun at the same time.
As things have gone right for the open source free software community,the free net community should be just fine.
Things have to be setup in a way which promotes freedom, which does NOT allow a company to monopolize, and keeps the power in the hands of US and not companies.
Meaning we must keep this seperate from the internet, and keep it from becomming comercial.
I think its a good idea.
When things become comercial, then all the benifits of this community based shared internet access go down the drain.
I believe if this thing ran via the freenet protocal, it would be revolutionary for communications purposes,
or maybe not the freenet protocal, but it needs a protcal which cannot be censored or stopped by big business in the same way the GPL cannot be stopped by microsoft.
Because believe me, ISPs will fight this. be prepared to face AOL.
If you use Linux, please help development of Autopac
This is the ???th article in ./ in the last couple of months about building a wireless infrastructure on top of 802.xx.
You need to realize that these unlicensed services operate on frequencies that are specifically not guaranteed protection from interference, and which are shared with other users.
The power levels unlicensed systems can legally use are very low, and they are vulnerable to interference from cordless phones, other wireless data users, and other services sharing these unlicensed bands.
All things considered, these systems have worked remarkably well so far, but they are fragile and there's no guarantee they'll continue working.
7. the global Internet will come to the NANs for various confirmations of location, authentication and reputation. A reputation on the Internet will be unthinkable if you don't have a reputation someplace in the NAN environment, Like, who the heck are YOU, if nobody knows you on the NANs?
Which is the way it was, sortof, before AOL discovered the Web, or Microsoft with MSN.
Speaking of which, how much you want to make a bet the Microsoft either tries to take it over, monopolize it, or outlaw it as competitive to them?
"It is a greater offense to steal men's labor, than their clothes"
So, for those looking to do one or more of the above, this probably isn't the route to go. Also, remember Shannon's law -- the smaller the bandwidth, the less the data-carrying capacity of the data channel. So for those frequencies capable of spanning long distances (through skip, moonbounce, or whatever), the data capacity won't come close to what can be had with an 802.11b network.
Still, as strictly a communications medium, data over amateur radio frequencies is more than sufficient. What needs to be done is to discover methods to increase capacity on the available bandwiths by (1) increasing S/N ratios and (2) devising modulation schemes which transmit more than 0 or 1 per frequency cycle. Some of these schemes are very popular, but there's still a lot of work to be done, and a lot of improvements to be made.
Many comments here suggest wireless digital communications over many miles at low cost simply doesn't exist. It does exist, but with restriction. If you can live within the restrictions (and a little imagination might even provide solutions to work around those restrictions), then the low-cost solution is already here. There's no need to simply talk about it as if it doesn't exist.
You have 61 new messages.
Beeeeep!-
"Hello this is Ms. Johnson from down the street. The computer is broken and you have to fix it now. I didn't do ANYTHING to it it just stopped working. Don't ask me anything about it cause I don't understand computers. Just come over my house and fix it. Thanks."
Beeeeep!
"Yo you luser my skilz rule and I will h4xor your shit everyday!"
Beeeeep!
"Hi, can you - hey you kids shut the fuck up!!!, put that rock down!" click.
Beeeeep!
"Can I buy a computer from you?"
Having worked at a company designing and building the devices you use to deploy such networks...
This isn't about Internet. You missed the point.
It's about setting up neighborhood networks, so neighbors can communicate without paying anyone.
At some point, this can grow so different neighborhoods can talk to each other, over whatever means are available.
At some point, this can turn into an internet. In fact, tha'ts how the interent worked in the first place....
non-connected sites still used IP addresses, assigned to them to be unique to their organization, so they could someday hook up to others.
I'm not sure you're entirely correct, that 802.11b is only CSMA/CD. You can run 802.11 in CMSA/CA (Collision Avoidance) mode by using Medium Reservation for hidden nodes. The Hidden Nodes problem happens when you have an access point where A can talk to the AP, and B can talk to the AP, but A and B cannot see each other. It's very normal in 802.11b WAN setups to ensure every card has medium reservation turned on, so they can avoid collisions on the access point. Basically, Medium Reservation causes a node to "listen" for a free "line", if it's free it reserves the channel with the AP. There's a chance that the medium reservation request may collide, but by and large it helps avoid collision difficulties.
The downside? You're limited to 20 or 30 nodes per AP channel, so you have a similar problem. Also, in my experience, medium reservation very slightly downgrades performance per machine under heavy loads for the network.
I've been using an 802.11b uplink to my ISP for two years. It's been an on-and-off proposition. We share airspace with a bunch of emergency alert towers since our town is near a chemical weapons depot (Tooele, UT), and the interference from those siren communications is pretty enormous (right in the same band with 2.4GHz). However, within a mile or so you can get really good performance using directional or YAGI antennas for an ad-hoc peer-to-peer net without an access point, or with an amplified AP (but you have to be careful you're staying under the FCC 1-watt limitation if you amp your AP).
A friend of mine and I had a discussion on just this topic not long ago. We called the concept "ghettonet" -- that people shared connections amongst themselves, with gateways out to DSL, modem, or other wireless links. Routing and channel bonding were the thing we figured would stymie the effort, beyond just a few nodes in the NAN.
Thanks for the info about the reduced-preamble mode; I cut my MTU down to 576 to try to get around the issues with some packet loss due to the interference from the aforementioned towers, but the negotiation overhead with my ISP's AP makes the reduced MTU of somewhat dubious value. It's reduced the number of dropped packets at the expense of bandwidth. Eh well. It's a fun game to play!
Matthew P. Barnson
I learn what I think when I read what I write
The halls that i live in at uni (Bangor N. Wales) were all laid with cat 5 for years however the university refused to actually install the routers and switches till this year. Me and some other geeks in the building started planning the following solution to the problem :))
Every 2 floors we bought 1 adsl connection (£40 pm) and connected all the machines to one back bone a squid proxy + linux box connected to eac adsl line + ip masquradeing then connectect each machine to local squid serevr and each squidd server to each other as peers
make the squid server Masquarade but rather than just simple masqurading use a round robin DNS so each connection request went to a different adsl line. Hope fully this would keep down starvation/flooding of links. Using tos for more advanced routing and a couple of patches u can favour local (ie close low number of hops) connections if they are free and use further away connections when localtraffic is high so ping times will vary but bandwidth will remain constant.
This also means u can use a large number of different ISP's so if one goes down every ones access gets slower but doesnt die compleatly
so now you have a system where every x many houses share an adsl line amungst them and the larger nan is only there for local use or when there are a lot of local requests also your squid proxies share amungst themselves (most traffic is web only) and reduce total bandwidth use you can also use this to have ipV6 in the nan and ipv4 on the internet connection (this will help with namespace and ip allocation
This is nothing new. Windows 95 has had Network Neighborhood for years. You stupid linux geeks are so behind the times.