Printable View
Free multi-megabit data transfer. This will probably become a reality if the net neutrality bills are shot down in the US.Quote:
Originally Posted by TBOL3
256-bit encryption would mask what file were being transferred through other people's routers, same way bit torrent clients use encryption. The advantages of a Darknet are really limitless, just so long as the user base is large enough.
Dugg
http://en.wikipedia.org/wiki/Net_neutralityQuote:
Originally Posted by schrombot
did you even know what Net Neutrality is?
All I know about Net Neutrality is that the term is misleading, and the real definition seems to take for ever to read (that's a lot of stuff on Wikipedia). I think most people would like to see the internet remain democratic in the way it works, but it's still not free. Not in either sense of the word.
Yes, the FCC is currently in control of how much power ISP's have over QoS. They allow prioritizing things such as VoIP for customer gain, but will not allow blocking or lower prioity of ports or services for gain of the ISP, such as kick backs to keep sites live or fast. If congress were too relieve the FCC of mandate powers over ISPs, they would be able to charge, at will, what they wanted to prioritize sites. On the other hand, if the internet was classified as a utility and the FCC is kept in power, then all ISPs would have to provide equal service for all traffic, unless it was beneficial to the customer.Quote:
Originally Posted by darkhatter
This is my understanding of the whole deal, if I am wrong, let me know! Thanks!
If you find yourself returning to this forum to check on it's progress, I have a request for those who are reading this. Help me try to develop this idea. If you've had this idea before, don't let me take credit, and give credit where credit is due. I don't want to make any money. I just want to find a way to very easily get people into this, ways to make it more accessible, and ways to attract people to it, to participate in a network such as this.
But that's just half of the issue. The other half has to do with software. How would I, or we Ubuntu/Linux/Open Source geeks, go about getting something open source written that could turn already existent wireless hardware into hardware that will behave in this sort of 802.1s way? There's windows software out there that does this. There should be some Linux software out there that does this too, but I don't know what it is, or if it really exists. I've heard that it's being used in the OLPC's, but that's all the info I have to go off of. Is there software for Linux that would allow this to work? And if so, is if open source, and are there developers out there who are willing to invest some time in expanding the software's capabilities?
Anyway, let's just keep the ball rolling. Sounds like a dream, or a waste of time to some, but I call it a hobby, which could become something huge.
I really find this idea very thrilling, and I would be willing to try this whole thing out.
Maybe some background about the OLPC project and how their mesh works. They do not use software for this because if the mesh network runs in software, the first thing to increase battery runtime would be to turn the whole thing down. The OLPC uses a wireless chip that has a separate low-power microcontroller which coordinates the mesh capability. This makes the mesh network also available if the computer itself is turned off. Maybe this method could also work with mobile phones - there are more and more pushing into the market that have wireless lan capability... seems the next one from http://www.openmoko.org/ will also get one.
I also have a question about the mesh network:
If there are bottle necks, let's say a chain of computers that connect two big swarms, how does the connection behave there?
Another small idea for input: Why not put a small computer into your car? You would be able to provide a mesh node to people around and also have net access at the same time. You could for example use http://www.openstreetmap.org/ for gps navigation.
Looking forward to additional ideas about this...
Well, we've got a good p2p software, now we need wireless routers, and good protocols to run them.
I had the same idea (well, on the 'hey, is this possible?' level) for some time after I was inspired by a digg post that showed how to make an antenna out of a Wok. I don't know where that post is, but a search for Wok antenna brought this up:
http://www.usbwifi.orcon.net.nz/
Well, I would assume that it would behave no different than the way any wired network would work where there are two networks seemingly connected to each other via one single route. But that is just sort of a rarity these days, especially on the internet. Now, one path between two swarms wouldn't that bad, considering the available bandwidth between those two links is well in excess of a few hundred Mbps, and I think we've gotten beyond that limitation thanks to standards like 802.1n. So, the short answer is that, traffic might, or might not be, a little bumpy. But over time, multiple paths would appear, convergence would take place on more broad range of geography, and soon, there would be no bottle neck.Quote:
Originally Posted by silwol
Yeah, this isn't much of a new idea. I bought a book about a year ago that goes into good detail about how to build and install you own ITX motherboard based "Carputer" in your auto. You can attach GPS devices, Wireless networking devices, Bluetooth to connect to your cell phone, which would in turn, connect to the internet via a paid subscription, if that's what you're wanting. But we already mentioned a wireless networking device. I've even seen very stylish looking external antennas that you can mount on the roof of the car for added reception and transmission.Quote:
Another small idea for input: Why not put a small computer into your car? You would be able to provide a mesh node to people around and also have net access at the same time. You could for example use http://www.openstreetmap.org/ for gps navigation.
Looking forward to additional ideas about this...
The only problem with this is, let's say you live in a city that is also participating in a thriving mesh network. This same question applies to mobile devices like cell phones, laptops, etc. And the answer is in protocol handling. The question has to do with hop management. Going from one node to the next. How quickly can a laptop, for example, (let's put it inside of a moving car, just for fun)....how fast can it fly to new ad-hoc connections, in the same way that cell phones are able to go from tower to tower on the fly, without interruption (hopefully). This shouldn't take much to work out, but is just a minor concern for now...