Eben Moglen Calls To Free the Cloud

paxcoder writes "You have been informed about Diaspora, a (to-be) distributed free social network. What you may not have known is that it was inspired by an excellent talk by Eben Moglen called 'Freedom in the Cloud.' But it doesn't stop there. At Debconf 10 this month, Moglen went further, and shared his vision of a free, private, and secure Net architecture relying on ('for lack of a better term') freedom boxes — low-price, ultra-small, plug it into the wall personal servers. He believes they will catch on since they will eventually cost less than a router, provide more functionality and freedom to the user, and even help your friends bypass any censorship by encrypting and routing their traffic. Since hardware is being taken care of, we are called to assemble the software stack. The title of this sequel talk is How We Can Be the Silver Lining of the Cloud."

Comments

All they need to do is everything

[Anonymous Coward]

Hardware that no one has adopted with software which no one has written is not a replacement for social networking sites.

Re:All they need to do is everything

[Unoti]

Hardware that no one has adopted with software which no one has written is not a replacement for social networking sites.

You raise a good point, but this is a chicken and egg issue. Back in the day, near the dawn of the personal computer, user's personal machines were generally not networked. You could get a network card, but there wasn't much point for most users. This is because there were not generally useful network-aware applications, there was a lack of lots of other machines to communicate with, and a lack of generally useful information to share on the network. Each of those kinds of problems posed a barrier to solving the others.

Facebook, dating sites, and other social network sites in general have the same kind of chicken and egg problem when starting up-- there is no real value for the early adopters because nobody else is there yet.

So your statement that hardware that no one has adopted with software which no one has written is not a replacement for social networking sites is completely true, obviously. But at the same time, there has to be a way to make the statement false. Otherwise, we must say that today's existing social networking sites can never be replaced. Because whatever replaces them will, at the time of their birth, have zero people using them.

It may well not work out or not catch on, but somehow, some day, today's existing status quo will fall and be replaced by something else. And something else has to be built before it can be used.

Re:All they need to do is everything

[mrogers]

It may well not work out or not catch on, but somehow, some day, today's existing status quo will fall and be replaced by something else.

I can't help thinking this is how the Communist Manifesto would have sounded if it had been written by Marvin the Paranoid Android. ;-)

Re:All they need to do is everything

[Xamusk]

I think there's more trouble facing the early adopters. For example, even the hardware isn't all that good to start with. The "modern replacement" of SheevaPlug (mentioned in "hardware is being taken care of") isn't all that good. In fact, this new version, the GuruPlug, suffers greatly of an lack of thermal design. This causes the plug to overheat and start rebooting, until the embedded power supply fails (also because of heat dissipation problems). As a result, to use one of those, the user must also mod the hardware, which creates all sorts of trouble. The manufacturer doesn't even care about it, and keep selling it for those naive enough (like me) to think that the manufacturer should take care of those problems before even starting to sell a product.

Re:All they need to do is everything

[Anonymous Coward]

They've announced on their website that they have, in fact, fixed it. New models ship with an internal fan, and older model owners can get a "free fix (for a nominal shipping fee)".

As software engineers, the EFF are good lawyers...

[dominion]

I see where he's going with this, and while I expect that certain aspects of the concepts will eventually be implemented in different ways, we have to be clear that the idea of everyday people administering their own servers is just not practical. I realize everyone here sees it as something we're willing to invest our time in, but most people don't. Servers exist for a reason, there are people (called system administrators) who can specialize in making sure the server software you're accessing, your data, etc. all are secure and have 99% uptime.

I'm not the kind of person who thinks that there is a divide between a sort of tech elite and the unwashed masses who will never understand this stuff. I'm one of those people who thinks that even your grandmother can learn how to recompile Apache given enough time, interest and dedication. The problem is that doctors are busy being doctors, plumbers are busy being plumbers, parents are busy being parents, and so on an so on. Even as a software developer, I prefer to not administer my own servers if I don't have to. I have friends who are very intelligent people who are very accomplished in non-computing fields who use virus and adware-ridden Windows machines. I don't suspect they're interested in taking the time necessary to fully secure a server that holds a digital representation of their life.

So this idea of a total peer-to-peer networking is not an approach I think we should pursue, not because it's not technically achievable (it totally is), but because it's not practical on a social level. This is reflected in the difference between Appleseed's approach to open source social networking and Diaspora's: Appleseed uses a federated node structure, and Diaspora claims to use a P2P, although we haven't seen the code yet, this was the original promise, and since the EFF is backing the project, it fits in with what Moglen is suggesting here.

We'll see where we end up, but I worry that if we push for Moglen's approach, we may see a small ghetto of tech savvy users who adopt it, while everyone else chooses to remain with the proprietary systems, because they're just that much less hassle. It makes much more sense to me to push for federated, hosted solutions, so that an ecosystem of servers (administered by professionals) can exist, and users can move freely between them.

Michael Chisari
http://opensource.appleseedproject.org/

Transcript

[PrecambrianRabbit]

For people who hate watching video as much as I do, here's a transcript: http://www.softwarefreedom.org/events/2010/isoc-ny/FreedomInTheCloud-transcript.html

Re:Transcript

[paxcoder]

I guess that Tor or Freenet are two of the things that would be run on these. Then there's your mail which you don't let Google read, there is social networking secure with PGP (and so is your mail) - so under your control. The main thing is it all runs 24/7, comes pretty much preconfigured, and as said, is more convenient than a dumb router. Then there is telephony which I ommitted - who gives you encryption for your calls? Well now you can. There is also absolutely no reason why one should pay so much for a simple thing as sending an SMS. Your own web server if you want, torrent, versioning system I don't know... You've got CPU time to spare so BOINC perhaps.
In short, you have a simple to use server of your own and don't need to use loads of third party web services anymore. It's you and perhaps your friends - the *real* trusted computing. Think of your own application for this. Federated things are a way to go, lest we want to loose our freedom.

Re:Transcript

[Lennie]

The idea is to have a small box, which does not use a lot of power. Which you can use to securely communicate with your friends in a distributed fashion, without someone else having the logs they can analyze and sell to companies, like Facebook is doing.

A small server which is simple to use, easy to update (most people shouldn't need to admin their own box) and backup. It will hold your data, and possible your friends (you keep my backup, I keep yours, encrypted ofcourse, think: duplicity ).

Re:Any tech specs yet?

[mrogers]

Someone please just tell me: are they nailing down a protocol spec first so that we can all do our own interoperable implementations, or at least all contribute code, and so not have the time wasting nightmare that was the Freenet project?

They've done better than that: they've written the code, bundled it into a convenient cross-platform installer, documented everything, and ported a ton of apps to run on top of it, including BitTorrent clients, web servers, anonymous email and IRC. It's all free as in speech and free as in beer, and there's a supportive community of developers and users.

Yeah, I know, I couldn't believe it either. It's called I2P.

An excuse

[ksandom]

I'm split on this. Mostly I think it's excellent because it sounds feasible to get a lot of people behind it, which would then make it quite effective. It'll bring back a level of "privacy" that we took for granted not many years ago. It will also open up the connotations that come with that, although I'm sure that has/will be discussed to tiring length.

But where my concern really is, is the trend that those in power see something like this as if it's only purpose is crime. They will be scared of this, because it will undermine their ability to do their job. When there's something they are scared of, they clamp down on it and make an example of someone. If you're that person it doesn't matter if you've done anything wrong, because they will find something, and bend it to the context that allows them to say you've broken a law. eg It could be an image sitting in your browser cache that they can object to based on someones' religion, that came in an ad on a page.

Early adopters will face significantly higher risk than those adopting once the project is well established. In this countext I see three distinct routes:

1. Manage the athorities' and public view: Ideally sell the idea to them that this is a good thing for them. I can't think what angle that would be, but it would be worth it. Convince them that this isn't the evil devil they will otherwise assume it to be.
2. Ignore the authorities: Take a chance and go for it. Don't rub it in their faces. Just get on with it and try not to make a scene.
3. Rub it in their faces: Highlight that this is going to let people bypass their precious proxies that combat terrorism.

At one end of the scale, you may even get buy in, but hopefully won't attract too much negative attention. Potentially, you may have a more "legit" user base who have positive community concerns. At the other end of the scale, things could get rather ugly. The authorities will. not. like. you. They will do everything in their power to shut you down, and there will be significant risk to innocent people who had good intentions at heart. This is also very likely to attract the people who the authorities will have a legitimate concern over. You're going to get those in any scenario, but the proportions will make a big difference.

Take care. I really do believe this has a legitimate positive place in modern society.

Here's the thing...

[fyngyrz]

It's fine to build a better server. But a network is not just the nodes; a network is also the paths, and the paths, my friends, are not anything either the telecomm concerns or the government are going to allow us to control, or have any of our own. And this gives them, if they think they need it, complete control over these new systems. If traffic passes over their paths that concerns them, they'll just shut it down.

So while I appreciate the idea, it's literally only half-baked. Wake me up when someone builds an inexpensive network in unregulated RF space. Until then, control, and therefore freedom, is unattainable.

Re:Here's the thing...

[fyngyrz]

I know. I'm an extra-class HRO.

The FCC has long been complicit in awarding, maintaining and ensuring that access to the RF spectrum is thoroughly monopolized through them. That's why they never allowed local AM and FM stations at the citizen level worth a damn; that's why even the ridiculously expensive "low power FM" stations were only drizzled out, and even then, incompetently and mega-slowly; that's why HRO's are restricted from playing music, "broadcasting" (meaning, transmitting to a general listening audience, like SWLs, rather than just to other HROs); that's why any number of restrictions exist. The government protects and serves the corporations before the citizens get even one moment of consideration. And that in turn is part of why we'll *never* have access to a citizen's "network band" or anything like it. The other part is the government's perceived need to monitor us. That's only getting more intense as well.

What we actually need is a new *method* of communication, and worse, we need to get a jump on it before the government does. What? I don't know. But as different from RF as laser links are, and as non-interfering as they are as well. Quantum coupling or something. I don't know. All I know is that the Internet as it exists now is more locked-down and regulated every day. The odds of actually increasing freedom within its bounds... pitiful.