Hacker News new | past | comments | ask | show | jobs | submit login

We need to invent the following things so we can be forever truly free:

1) A secure, peer to peer DNS system with no registrars and a dynamic, heterarchical network of trust.

2) An encrypted version of TCP that we can drop in without anyone noticing.

3) A way of distributing keys for our encrypted transport layer without certificate authorities and again with a dynamic, heterarchical network of trust.

4) A secure, peer to peer, federated social network with all the features that we and the rest of the world want and need.

Then we will at least have some breathing space while we work on that whole planetwide wifi mesh thing.




I wouldn't suggest that we need a social network, but your first suggestion isn't too far from the truth.

The exact thing needed is very boring, very low-level, and very unsexy, and I believe can be summarized thusly:

We need a method for locating hosts on a network graph which does not have a central point of failure and which cannot be easily disabled.

We then need a method for authentically routing messages back and forth between these hosts without fear of man-in-the-middle attacks that can change the contents of the messages.

This is pure transport-layer engineering.

As long as we can locate anyone connected to our network, and communicate with them without interference, we can build whatever else we need to on top of that.

We shouldn't confuse our efforts by trying to make a social network, or new hashcoin lottery, or advanced supergovernment, or whatever else. We shouldn't worry about interception of message contents--that way lies madness; as long as I know my message reached somebody in one piece, and as long as they know that a message signed by me is from me, we can fix the rest later.

This is a pure, straightforward, fucking hard engineering problem.


Why wouldn't government simply ban the whole thing? Consider ham radio: my dad used to be really into it, and it doesn't cost much to get into it. But the government can scare away a lot of people from it, simply with the distant threat of some kind of punishment. So most people do it the legal way, at which point it becomes a big hassle - tests and certificates. And as soon as tests and certificates are needed, most people lose interest.

You can set up an illegal radio station for less than $1,000. My friends and I did so in Virginia, in the USA, back in 2002. We were out in a rural area of Virginia. We had great fun playing our favorite music to whoever would get the signal. The FEC is slow to crack down on stuff like that when you are out in the middle of nowhere. The radio station lasted 2 years, and it only got shut down when we moved on to other things. I have fond memories of it.

Running a radio show is great fun and, if you are an extrovert, it can be addicting. So why don't more people do it? Because it is illegal.

Likewise, if you created a protocol so free that government regulation was impossible, then the government could simply make it illegal. You would be "free" to use it, just like I was "free" to setup an illegal radio station, but most people won't go near it if it is illegal.

There are many things that go through our society, and which flow so freely that the government doesn't have the power to stop it, so instead it increases the penalties. Drugs would be an example. In that case, a lot of people get scared away from drugs simply because the government policies are draconian -- small amounts of drugs, found on your person, can lead to years of pain and legal trouble.

I agree with the other comment where someone says that you can not come up with a technology that will solve a policy problem. The ultimate power of the government is that, in the end, to uphold the legitimacy of the law it has the power to kill people. You can't come up with some cool technology that lets you get around the reality of a punishing government, if the government decides that some technology is too dangerous. All you can do is what the people of Syria are doing now -- organize, resist, protest, possibly even fight. There are only political solutions to political problems.


>Running a radio show is great fun and, if you are an extrovert, it can be addicting. So why don't more people do it? Because it is illegal.

Here in the UK, we don't give a fuck. For half a century, not a fuck has been given. All those pulsing repetitive beats you've heard in your pop music for the last 30 years, that stuff that's made people untold billions of dollars? That's largely a result of us over in the UK not giving a fuck about getting in trouble for pirate radio.

http://en.wikipedia.org/wiki/Pirate_radio_in_the_United_King...

http://www.londonpirates.co.uk/stations.htm

http://www.guardian.co.uk/music/2011/sep/08/pirate-radio-rav...

http://www.youtube.com/watch?v=mpS0jR6FG1o (this last link says it all)

Just sayin'.


The catch here is what you're talking about is the ultimate power of a government.

The system described wouldn't stop a government from censoring the internet for its own people. No technology can stop that. It would however prevent a government from becoming the government, which the Americans seem to try endlessly to do.

Once the technology was "in the wild" the US could simply "ban the whole thing"... and join North Korea in the nuthouse of closed internet while the rest of the world passed them by with a curious shrug.


Or, more accurately, which the minimally-accountable American government seems to try endlessly to do.


I used to have an Amateur Radio license. I didn't mind the tests, I enjoyed learning the theory, and I even learned Morse code. I went to hamfests, bought a radio, then suddenly realized "I have nothing I wish to say to these people". It was almost entirely old men telling each other the details of their medical conditions.

I was glad when it lapsed. The penalties for emitting RF that falls outside of the rules somehow tended to be significantly higher if you actually had a some kind of a license than if you had none.


RF is a shared resource. So are roads. Requiring proof of competency to operate nontrivial equipment that uses and can cause DoS or physical harm to people on shared resources is perfectly reasonable.

My dad and several of my friends have HAM licenses. It's not inordinately difficult; it's kind of like getting a driver's license. The structure is not designed to prevent ordinary people from using it, just to keep it usable and reasonably civil.


>Why wouldn't government simply ban the whole thing?

If they ban it but it lets people do something they want to do and can't otherwise as conveniently then they will ignore the ban.

>but most people won't go near it if it is illegal.

Really? I think the media industry would beg to differ. Normal, everyday people have stolen more songs than they have time left in their lives to listen to.

You're taking an obscure example that many people don't even know about, much less want to do and using that to say people won't touch it if it's illegal while missing media piracy; something most people do want and most have had no issue just taking it, laws or no laws.


Jurisdiction. The useable rf spectrum is public and limited by physics, so we need regulation.

Private networks hooked to other private networks are not similar. The internet is not a thing.

They can go after parts they have power over, like .com, but the core concept of simply communicating...... O so easy.


It's funny you mention the illegal radio. All of this end of the internet reading I've been doing had me wondering the other night, if things get really bad, will those savvy enough build their own underground network? Pirate Internet could breed some interesting communities.


How far off is that from the deep net? It's all based on tor (and as such is slow as molasses), but it's navigated by passing links to community members - not searching Google.

That's pretty darn underground.


Coding is a lot of fun. Not many people do it. I think you have mistaken lack of interest with government threats.

However, let's suppose that tons of people wanted to set up their own radio station. What would the result be exactly?


Yeah, I edited my list to include a CA-free encrypted replacement for TCP, just before I saw your comment.

To all the super-geniuses reading this: burn your socialmobilelocal startup to the ground tomorrow and do this instead.

Start by watching this Blackhat talk I linked elsewhere in the thread, the guy has the right kind of ideas: http://www.youtube.com/watch?v=Z7Wl2FW2TcA


ie. if you really want to change the world instead of paying lip service to it and going for the money grab, here's one opportunity.


Indeed. Listen up all you MIT kids: do you want to be cooler than Mark Zuckerberg, or do you want to be cooler than John McCarthy, Tim Berners-Lee and Captain Crunch all rolled into one?


How do you pay for it?

I have a team of capable engineers, and relationships with network/wifi hardware/firmware developers. What I don't have is $2-10M to spend designing open protocols (and building the software), especially with no obvious commercial upside.


No idea, but that doesn't mean it can't be done either non-profit-wise or even with significant upside, it's just your chances of VC-money and selling to a major corporation for fuck-you money are diminished relative to TechCrunch friendly buzzword related startups.


Non-profit might be the best way to go, here. I'm sure you can wrangle up enough in donations and FOSS contributions to make it work. The idea of an Internet free from political and legal malfeasance is one whose time has come, and enough people are realizing that fact that there should be sufficient collective energy to move a well-designed system forward.


Kick-starter? I'll pitch in just to see a decent (and open) white paper come out of it.


Last I heard, Mr "future of authenticity" was just acquired by Twitter. :-)


As I understand it, typing in all-caps or just bashing your face into the keyboard is considered poor form around here, so instead I'll just calmly state that this information displeases me immensely, and evokes a not inconsiderably bitter sense of irony.


:-)

I trust Moxie to be doing the right thing.


Twitter didn't buy Moxie Marlinspike. They bought one of his companies, which makes a variety of encrypted solutions for mobile phones. The Convergence project (which is an alternative to certificate authorities) remains independent of Whisper Systems.


This is a pure, straightforward, fucking hard engineering problem.

Yes, in fact it's one of the famous "only two hard problems in Computer Science": locating hosts on a network graph is just a restatement of "naming things".


If only. Unfortunately, it's not like that. With how well the "internet" is established, this is one case of NOT build it and they will come.

You can build it, but it'll never gain the authority to make everyone use it.

You'll need to be a Google or Microsoft to pull this one off.


Just like bittorrent required Microsoft and Google to become popular? I think you're underestimating the value younger generations hold to free as in speech.


TBH I think it's more like free as in beer. Torrent hasn't become popular by passing censored information, it became popular by passing pirated entertainment. Movies, music and software, especially games. And pr0n, of course -- the best way for a service to become popular is to provide free and easy access to it.


There are some people who consider low-level to be sexy.


But then you'd still be subject to existing laws - it's not like they're going away. With the same amount of effort, you could get the governments to stop making stupid decisions/laws or at least consult someone before they do.

But then again, they have tech consultants who never get a say because ruling in favor of those with money is better for the lawmakers' careers or pockets.

I do believe that most people are pretty comfortable with the current situation... when it'll start affecting more individuals, then the "anti-" movements will ramp up - it's always been that way...


> you could get the governments to stop making stupid decisions/laws

I know we live in an age of hitherto unimagined prosperity and wealth and miracles, but you may be reaching a little too far.


It's not governments that make stupid decisions, it's people; take the decision away from government, and someone else will make those stupid decisions, but stupid they will be.


> It's not governments that make stupid decisions, it's people; take the decision away from government, and someone else will make those stupid decisions, but stupid they will be.

Except that only a government can make me care what they decide.


> Except that only a government can make me care what they decide.

Simply not true; it's just that only they can do it legally, but to even say legally, you need a government to enforce legality.


Or just use a registrar and top level domain that won't respond to the take-down requests. .nz domains, for example, are protected by a rigorous process in a domain (New Zealand) that prevents this sort of thing.


Do you know how far that goes? No doubt it works against any schlub who comes along asking for a domain to be shut down, but would they really stand against it if the US put serious pressure on them for hosting thepiratebay.org.nz?


We've already got Namecoin and Diaspora, but neither appear to be catching on.


Namecoin is pretty great, but as with all these things, there's the problem of the network effect.

Diaspora has all sorts of problems.

With the SSL certificate thing, there's Convergence, which basically solves the whole lot at a stroke. See http://convergence.io/ and the guy's Blackhat talk at http://www.youtube.com/watch?v=Z7Wl2FW2TcA


"problem of network effect" if you accept the premises that: a) the diaspora's of the world are the solution (or at least a step in the right direction) b) the problem is that of a network effect (translation: the cold start problem)

... then one solution is for a well known, highly trafficked brand, that has nothing to lose, to implement "a".

... a.k.a.... what if someone like yahoo which still has vast reach, but is not a player in social, took something like diaspora and promoted heavily. That would solve the cold start problem.


I was thinking about 1) and it seems like a really cool idea. If you could also query the sites trusted by some person, that would be like a HN with domains only to some extent (and if you can have any domain you want, you'd probably use one per site, rather than making a hierarchy again). Realistically I probably need only 20 or so domains on a typical day anyways, so that would only require trusting 5 or so other developers.


The P2P Yacy search engine seems to have been announced very timely.


Sounds like you know exactly what is needed. Why not start the process of making this happen?

Where would one need to begin to have a secure peer to peer DNS system?


If it were more usable, I'd point towards Freenet. They've been at building a system like this since 2000 -- it's just a shame it runs so slowly.


Sounds like a great project for Mozilla to jump on. I'm slightly surprised they don't appear to have worked on this problem before. Or have they?


1) Namecoin - http://dot-bit.org/

2) Tor

3) I think 1 & 2 should solve this.

4) This probably isn't that difficult to implement on top of Tor.


I think that 1-3 are necessities now. I don't know if 4 is.


You can't invent around policy problems.


Sir or madam, with that attitude, I'm surprised you are here, in a place called Hacker News.

A great many of us make good money working around broken systems and architectures. We live for this shit!


And yet when it comes to government enforcement, there seem to be an awful lot of problems we haven't been able to hack around:

- Brokenness of patents causing anti-competitive behavior

- Absurd copyright laws preventing the building of an untold number of innovative products

- The emergence of true competition in the wireless space.

I could go on. We're very good at making end runs around corporations, big and small, but mostly because they can't kick your door down in the middle of the night.


Yes you can, if you make the invention critical before they realize the consequences.


The Internet has become critical and "they" are beginning to realise the consequences, which is leading to increasingly invasive and ill-informed policy.


Yet they're still nowhere near putting the genie back in the bottle, and the internet is so embedded in the economy that they can't, they can only try to minimize the impact.

It's time for the next leap - making the internet properly fault tolerant.


Exactly.


At least not when the policy is enforced by the people with a willingness to use lethal force, ie government. People invent around policy problems, when the policies are business related all the time.

But I wonder how many of the people complaining about this and SOPA were in favor of Obamacare?


http://tcpcrypt.org/

Passive encryption... not the best, it prevents lazy MITM.


Could it be that some or all of these things either already exist or are possible using existing "tools", but are simply not widely known nor in widespread use?

However statements like "all the features that we need or want" are troublesome. How can we possibly agree on those?

What if someone built the low-level, _boring_ "platform", and offered a _simple_ (as little code as possible), _old_, boring but workable conceptual vision, then let all the high-level enthusiasts address (argue about) usability and features?


>However statements like "all the features that we need or want" are troublesome. How can we possibly agree on those?

That last one was the least important and the most flippantly stated (apart from the global mesh network.)

>What if someone built the low-level, _boring_ "platform", and offered a _simple_, _old_, boring but workable conceptual vision, then let all the high-level ethusiasts address usability and features?

That's exactly what's going (think positive!) to happen, having a p2p facebook is just the last piece of the puzzle, the cherry on the cake.


Ah, that's good to know.

Now, sadly, I must ask, is someone going to try to make this proprietary and embedded, contained within hermetically sealed hardware enclosures, complete with convoluted bootloader and behavioural studies rootkit, to try to make billions from it?

To my knowledge the low-level part can happen right now using freely available code. Is it naive to think we could keep this simple and open?

Even the p2p FB piece is possible, assuming you do not need to have 1000's of friends on each of your separate social networks.


  Now, sadly, I must ask, is someone going to try to
  make this proprietary and embedded, contained within
  hermetically sealed hardware enclosures, complete with 
  convoluted bootloader and behavioural studies rootkit,
  to try to make billions from it?
If the problem is submitted to the market, then yes. That's what the market and capital does. So sad.

It'd be great if the world's geeks would stop laboring for the fucking market, increasing the surplus value exacted from their labor, and winding up fucked by their own creations.

EDIT: formatting


"winding up fscked by their own creations"

This would be easier to understand if you could provide an example.

Do you think this is a problem that should be submitted to the market?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: