Could This App Create A Free, Secret Web? | Hacker News Reader

Could This App Create A Free, Secret Web?

forbes.com

71 pointschippy26 comments

Just a caution because it's not clear from the article or app, Open Garden volunteers your mobile data connection. That seems to be the intended purpose for meshing your personal devices, but it volunteers your connection to the entire mesh, ie everyone nearby[0].

That'd be fine for eg my home wired connection (especially if bandwidth throttled so mostly for the use of firechat), but metered cell data could potentially be blown just by sitting near someone with a laptop that wants to do an update... it seems like something you'd want to warn people of and not hide in a forum post.

0: http://forum.opengarden.com/discussion/345/i-want-to-know-is...

I'd like to hear more of the story of how mesh supposedly avoids surveillance. I talked to a few people working on mesh projects (not Open Garden) who readily admitted that a spy node could join the mesh and monitor user traffic (even apart from the possibility of passive wireless monitoring).

Does "secret web" here just mean "normally within your own city and not using someone else's faraway data center as an intermediary"? Or is there a stronger privacy claim? I don't see one in the Forbes story.

Depending on project, some aims for obscuring or hiding the ultimate source/destination from intermediary nodes, as well as ensuring end-to-end encryption of all data.

Generally the idea seems to be that the mesh functionality is there to prevent blackouts, while the surveillance avoidance is down to layering encryption to various extents.

Key to surveillance avoidance then becomes to get enough "legitimate" traffic over such a network that just monitoring traffic amounts isn't enough to look for potential suspects.

But you still need to get out to the main internet somehow right? And then whichever poor sap has an outgoing connection is going to end up sharing it with everyone else and get cops coming to his door about the child porn.

http://opengarden.net/2013/07/using-tor-over-open-garden/ might be a solution to that.

https://play.google.com/store/apps/details?id=org.torproject...

Maybe just get over caring about the "main" internet, and leave it to be the ad ridden, surveillance clusterf*, old and rusty one?

Or you can use ham radio.

Encryption is outlawed on public radio bands.

He says, posting on a website he visits daily on the "main" internet...

I had a lot more insightful discussion on now-defunct .onion forums and IRC channels than I ever had on hacker news.

Much more interesting links, as well.

> now-defunct

To be fair, that's kind of his point. The relative advantage of the Internet in this case is already having critical mass.

Arguably, achieving critical mass is one of the failure modes for forums. :)

Open Garden: Not open source. Nice one guys.

Serval Mesh is an open alternative.

This is really cool. Combine this with IPV6 and decentralized DNS and it will be very hard to censor the internerd.

> There’s been similar interest for FireChat in Iran. Users in the country have started 1,800 FireChat groups, according to Open Garden, making Iran the second biggest user of the app after the United States. India, Brazil and Mexico follow close behind, and Open Garden says people in Cairo and on the outskirts of Baghdad, Iraq are using the app too.

Oh god. What have they done? Please tell me they have encrypted the communication between users by now.

Wouldn't it be illegal for an American company to export strong encryption to Iran? At least, that's the gist I get from the JCE unlimited strength policy file. If not illegal, it would definitely go against the agreement for distribution of that.

Historically, clearly so. But since 2010 there has been a "General License" for exporting "Personal Communication Services" technology (essentially based on the idea that individuals' communication weakens the government's hold on society, or, as OFAC put it, "are a vital tool for change").

http://www.treasury.gov/resource-center/sanctions/Programs/D...

I don't know how the General License applies to cryptographic technology in particular. Paging Collin Anderson...

I do know there are requirements that need to be met for exportation of cryptography [0]. Interesting fact related to this: Strong crypto used to be considered a weapon. This is why the DVD standard is 40 bits (iirc, it's possibly 64 bits).

0: http://en.wikipedia.org/wiki/Export_of_cryptography_from_the...

Interesting concept. A year or two ago I was pretty interested in the various dark/mesh-net projects that kept popping up, but there were a few problems. Namely the fact that they were somewhat difficult to set up, and the fact that they were somewhat like TOR in the sense that accessing the "regular" internet required someone to tunnel traffic through.

Packaging this up into an app pretty much solves any difficulty of setup, making it accessible nearly to anyone.

I have no idea about how secure/sound current implementations are, but this looks like a cool first step.

Combined with something like TOR to bridge local communities, it certainly has the potential to become a truly useful technology.

Tested just now between an iPad3 and an iPhone5. Breathtakingly unusable and flaky as hell.

Requires you to pick a bunch of chat categories at startup, which is ridiculous if you only want to use it in nearby mode.

Could not see the other device unless both were logged into the same WiFi network, which is going to be totally useless in the middle of a field. Even with both devices on the same WiFi, only about half the messages showed up.

Made no difference if BT/WiFi were on. Completely unreliable. Deleted the app. Nice try. Better luck next time.

> A note on the technology behind this: For iPhones, Open Garden harnesses Apple’s Multi-peer Connectivity Framework

While it sounds like the iPhone has a secret new radio transceiver inside, it's actually based on existing WiFi and Bluetooth[1].

[1] https://developer.apple.com/library/ios/documentation/Multip...

The Serval Project is already working on this: http://www.servalproject.org/

Unfortunately, they are limited by Google's very long standing lack of response to requests to support peer to peer WiFi on Android. See https://code.google.com/p/android/issues/detail?id=82

“This can be a billion dollar company if we can get to the tipping point,” says Benoliel, referring to the 7% uptake in urban areas.

It might hard to monetize an app up to a $B where you have no real insight into/control over the audience… which is kind of the point. The company sounds super-cool but it should probably be operating more as a non-profit or lifestyle business, not as a hyper-growth startup company.

This is pretty cool.

I had no idea a smartphone was able to communicate with another smartphone 70 meters away. I had always thought that the reason a smartphone can communicate with a distant WiFi network, or a cellular tower, is because the recipient has a large antenna, which means it can send data over long distances, and pick up relatively weak signals up over long distances.

The wonders of this world. An app that that is hardly ever working is getting all this coverage on Forbes et al.

Just shaking my head in disbelief...

Reply on news.ycombinator.com