http://www.tempestcom.com/guide/guide.aspx?Id=99&view=3 Bgan is the World's most sophisticated satellite Internet service. Bgan uses a variety of small lightweight portable satellite terminals which allow travelers and remote users to connect to the internet in places like Iran without the need of infrastructure such as phone lines, cellular networks or Wi-Fi hotspots. With Bgan you just take out you small portable modem and point it to the sky to connect at speeds of up to 492 Kbps. This is the ideal way to get connected without getting blocked using Satellite internet. You can also use a mobile device that can work on wifi and hook Satellite internet to a router and that way to distribute to more devices. This way is the best possible and guarantee this can't be block from local authorities but only downside have to make sure you are in there coverage zone. Well this company claims to be able to offer service there in IRAN http://www.tempestcom.com/guide/guide.aspx?Id=99&view=3 So check it out...

Billy Hoffman and Matt Wood are expected to demo a proof-of-concept browser-based 'darknet' called Veiled at Blackhat at the end of the month, which carries many of the features you describe. http://www.blackhat.com/html/bh-usa-09/bh-usa-09-speakers.html#Hoffman The second component, which would decentralize data routing, may require a hardware implementation on devices which can take advantage of mesh networking protocols like 802.11s (which is already being used by the One Laptop Per Child project), enabling users to bypass carriers and ISPs entirely: http://en.wikipedia.org/wiki/IEEE_802.11s A third more optional, but potentially powerful component could be the addition of DIY, small, cheap, unmanned drones to these networks, as demoed by Chris Anderson of Wired at this years Maker Faire. Connected to a mesh darknet, these could be used to further the scope of intelligence gathering and citizen journalism: http://fora.tv/2009/05/30/DIY_Drones_with_Chris_Anderson

hi, i've succesfully ported TOR (precisely, both TOR and PRIVOXY) to windows mobile (tested on 5.0,6.0,6.1,6.5) as a private project. i'm in the process of submitting the port to the tor and privoxy development teams (will do it in the next weeks) to be included in their codebase. The thing is production ready and includes a GUI (similar to VIDALIA on desktop PC) to control TOR (start/stop/show circuit infos/etc...) and a practical CAB installer. I think this will solve your solution. Please contact me by email for further information.

Let me flag this as a "solution in progress" and not quite solved yet. I am excited to the see this challenge posted and am looking forwarding to seeing what others have come up with.

I am the lead on an open-source effort known as the Guardian Project, which is well on the way to implementing this based on the Android platform. Android is not quite there yet in terms of global availability, but we believe that will change in the next year due to the prolieration of low-end Android models being developed and sold by China Mobile and T-Mobile around the world.

Our goal is an anonymous, encrypted, secure handset for browsing, texting, IM and voice, with our intended base of users being humanitarian workers, human rights advocates, journalists and activists. You can view our presentation here: http://openideals.com/guardian

Our first release was a port of the native Tor code to Android with a Java-based Android app on top of it named "Orbot". This effort was done in concert with the Tor Project themselves as well as the University of Cambridge: Readm more here: http://openideals.com/2009/10/22/orbot-proxy/

We are also on our way to providing encrypted SMS (via http://www.cryptosms.org/ ), a Jabber/IM off the record app, and encrypted voice via http://SipDroid.org with VPN and Zfone support.

All code wil be released as open-source under the appropriate licenses.

 We've a functional solution to this challenge - though it requires a bit of fine-tuning to implement fully per the spec. However, before getting into details it's necessary to clarify a subtle but important misconception inherent in the structure of the challenge as written.

 

To accomplish these goals, it's necessary to have more than just a piece of software; more specifically, the software substrate is necessary but not sufficient for a successful implementation. The full implementation requires network service, in addition to the software layer. 

 

Tor is a great example of this objective reality. While Tor has some neat software, and lots of coders really proud of the software they have written for Tor (often, paid for by donations from citizens), it's a complete failure as an on-the-ground implementation. Because it simply "assumes" that some group of entities will donate massive quantities of network bandwidth on an ongoing basis, the Tor project is chronically, permanently, and intrinsically unable to support the demand of its users. That means it comes up with all sorts of rules to try to get people NOT to use the network - exactly the opposite of the plain-language goal of such tools. For 99% of non-geek network users, the hassle of scrambling for Tor nodes that aren't totally overwhelmed with traffic is far more than they'll ever attempt. That is why Tor is a neat proof-of-concept, but completely unsuccessful as a scalable solution for real human beings.

 

To provide that service to people, and not only allow but actually encourage them to use it for ALL of their communications online, all the time, it is necessary to provision real bandwidth. Nowadays, bandwidth isn't terribly expensive - but it's not quite free. Plus, someone must actually run that network and the bandwidth underlying the secure service - not just in their "spare time" when not flying around to conferences and other fun events but actual sysadmin work. A volunteer, donation-only model like Tor simply isn't structurally appropriate for that sort of ongoing service delivery.

 

Our privacy network provides exactly what nonprofit secure networking models lack, and we've been delivering privacy service online for almost 3 years already. We don't limit what our customer can do, the bandwidth they use, or the applications they are "allowed" to run. We simply wrap ALL network traffic in a secure network layer and tunnel it out of insecure jurisdictions and into a cloud of secure gateway servers.

 

To fully deploy our model on mobile "smartphones," we've a bit of tuning to do in optimizing our client software to run on GSM network topologies. We've actually tested those tweaks already, in-house, and it's not terribly complex to do it in a production environment. To scale the model to millions of private network participants requires not much additional coding, but the ability to subsidize the cost of network provisioning as the overall customer traffic scales.

 

Fausty | CTO, Baneki Privacy Computing

You most likely need to think of transmitter and code--using music as the code. It is easily transmitted on most devices (even if by voice). Transmittal will not be blocked (altho maybe in Afghanistan :) ). Familiar tunes are easy to remember and easily re-transmitted. You just need to figure out a code. It could be by song lyric--but i would think notes or chords would be interesting. Didn't they do this by radio last century? Someone could update it. Perhaps playing it back while playing another app would bring the solution to the code. The two by themselves would mean nothing or....(too many possibilites)...