31 votes

The Server Needs To Die To Save The Internet - Article on MaidSafe



Tip of the hat to Ernest Hancock, and the DP's Joη for this great article.

by Natasha Lomas | TechCrunch
July 23, 2014

Do we have the Internet we deserve? There’s an argument to say that yes, we absolutely do. Given web users’ general reluctance to pay for content. We are of course, paying. Just not with cold hard cash, but with our privacy — as digital business models rely on gathering and selling intel on their users to make the money to pay (the investors who paid) for the free service.

Users are also increasingly paying with time and attention, as more ad content — and more adverts masquerading as, infiltrating and degrading content — thrusts its way in front of our eyeballs in ever more insidious ways. Whether it’s repurposing our friends’ photos and endorsements to socially engineer selling us stuff, or resorting to other background tracking and targeting tricks to divert our attention from whatever it was we were actually trying to do online.

The commercialization of the web is the ugly reality of the hidden cost of all the datacenters and servers required to power the Internet. And that commercialization is compounded by the power of the big digital platforms that dominate the web we have today: Google, Facebook, Amazon. Increasingly we’re forced to play by their rules if we want to participate in the digital space where most of our friends are.

But perhaps there is another, far better way — that benefits individual web users and startup developers alike.

Continue reading:

Trending on the Web

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

As far as I can tell, this

As far as I can tell, this system is a lot like Bitcoin... a distributed system for security and redundancy.

this is the future, no doubt

Hope this succeeds.

“The welfare of the people in particular has always been the alibi of tyrants.” — Albert Camus


I don't know how to assess the likelihood that two hackers in a garage just obsoleted the internet ... but I guess that is what skype did once upon a time to telephony, and google did to advertising, napster did for music and gnutella did for file sharing (https://en.wikipedia.org/wiki/File_sharing). It is also where Apple and HP and Microsoft originated. So...

Fascinating! I have my fingers crossed! Getting my privacy back would be wonderful. Few things would give me more of a thrill than watching google and the NSA get a comeuppance.

Bill of Rights /Amendment X: "The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people."

Do you need a politician or judge to "interpret" those 28

I don't know about you, but

I don't know about you, but for me to have internet at home I have to pay for an Internet Service Provider and it costs me quite a bit of money. So no, my internet is not free. And still I have to look at all of those ads.

It is better to look dumb and not be, than to look smart and not be.

The 'Free' in MaidSafe means

'Free' as in 'Freedom', not 'Free' as in 'No cost'. I'm surprised you didn't pick that up, given that you're posting on this particular forum?!?

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

The video claims that the

The video claims that the user would not have to use any third party to use the network, only the software, without realizing that the software is the third party. While this tech sure sound nice there's just too much perfectness for me to buy this whole heartedly. I like the idea to distribute and encrypt data but...

I dunno.. something tells be this wont work because of unstable connections that might contain important data on the specific file you're looking for. Or, say, for example, that a country decides to cut off the communications at the border. Something that most definitely has happened before, actually just this last year. If that happened a lot of data would disappear making the rest of the data totally useless.

It is a nice idea but I would like it better if you could get this combined with a server-system that works as a back-up. Of Course this would have to be a pay-to-use corporate system. Which causes other issues, of course.

"3 continents would have to go dark!"

"3 continents would have to go dark!" according to David Irvine founder of MAIDsafe

To add to jonat3's comment below

Or, say, for example, that a country decides to cut off the communications at the border.

If they did do this, then the rest of the internet is gone, whether it runs on MaidSafe or the current internet. With MaidSafe, though, the cutting off would create a new subset of MaidSafe, where there might be a chance that your files are still within the subset. The network is programmed to automatically reconfigure, and so you may not lose your data.

Your point amounts to:
"Freedom could never work because someone could just shoot you dead" or
"Let's not start a business because the government might decide to make it illegal".

Yeah, they might cut the internet off at the border. But then huge amounts of commerce now relies on the internet, so it would burn the system, too.

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

Eh, calling the software a

Eh, calling the software a third party is true in a sense, yet it misses the point entirely. The problem with third parties have always been the human aspect and that they are middle men. Middlemen increase the cost of the system, because they demand fees. But they are also incentivized to make people reliant on middlemen in pursuit of those fees, basically striving for monopoly status. In other words, where you have middlemen (or just plain humans for that matter), you have sources of corruption. With software you don't have that. The behaviour is predictable, works for free and even if it fails, it fails in a deterministic manner, which is infinitely more solvable than the unpredictable nature of how humans fail. Since the software is opensource, there's also no lock in and force involved. Even if the software forced you to pay up, eventually there will be modified versions of it that will strip any nefarious aspect the software might have. It's opensource afterall.

Your other concerns about reliability may prove correct, but brings up a solution that's even worse IMO. If the connection between regular computers can be easily interrupted, then it would be even easier to interrupt the connection between servers, seeing that servers are centralized in nature and thus SINGLE POINTS OF FAILURES. In a battle between centralization and decentralization, the decentralized system wins every time, because the entire point of decentralized systems is that they have no single point of failure. The question is if the system proposed here has proper decentralization. If it can fail in the manner you just described with connections being easily interrupted, then it would mean that the system is not properly decentralized, because those attack vectors are single points of failures.

Anyways, I do agree however that the idea is an ambitious project and ideas as ambitious as this one tend to be harder to put in practice then one might expect. The best ideas in the world are worthless without the proper execution. I think the idea itself is solid and they are certainly on the right path, but it might be that this project will not amount to anything anyways, due to unforeseen problems that might occur in the future. I'm fairly confident though that the solution of the future is a solution that doesn't involve servers.

forgot one thing about the

forgot one thing about the maidsafe+server issue. The point is that the regular service would continue, the server would act as a backup if there is no fast enough node to deliver the data the server could respond and replace the other source of data. The server would not be a service used much. But it would be there as redundancy if the data required is either unavailable, slow or with a choppy error-prone connection, and there are quite a few of those (even on optic-fiber connections).

Somehow.. I get the feeling that the hardware on the consumer-level (The computer, router, wifi, shitty network provider, shitty isp's), the same target audience of maidsafe, is not of enough quality at this time to handle this kind of service, maybe in 10-15 years when we have a 95% GigabitEthernet connections this could be a reality...

Now I guess it is up to the devs to prove me wrong :)

You bring up some valid

You bring up some valid points. And I guess it all boggles down to the amount of stored data on the system and the overhead of the system. If the overhead is enormous there will probably not be an issue with lost data. Then the data will be available on 1000's of "mini-servers" at the same time. Like someone said below the required size of broken connections to cause a fatal failure is large. But if the overhead is small.. well.. I don't think a have to explain what will happen.

But.. I would like to be proven wrong. The only thing I am afraid of is that the software would be bloated and hackable, or maybe even controlled in the dark... think NSA CIA osv osv.. I know nothing about code so I can't really speak about it but it is a concern.

Another point is that the system will constantly send and receive data to update itself.. This WILL PUT A LOT MORE STRESS ON THE NETWORK('s). A stress that ISP and fiber-owners don't necessarily can handle at this moment in time. Maybe in 10-15 years when we have mostly gigabit-ethernet then it could work. But, like today, with a heap of Xdsl connections then this wont work, not even a little.. Actually I challenge the creators of this to show how it would work on a low-upload-bandwidth connection. Something tells me it wont be compatible

Garan's picture

I See Two Aspects of This, that should not be combined.

The first one is the overall scope and impact of using this type of technology: individual privacy, power, and control. This point should not be confused with any individual attempt or implementation, or the main point may be lost.

The second one is the details, caveats, concerns and arguments of the specific implementation. Nearly any critical thinker can come up with potential counter examples to a solution, yet any individual concern does not invalidate the entire solution. That's throwing the baby out with the bath water. Also, any individual concern can be met with a solution to either prevent or minimize the likelihood of a problem.

That being said, from my view, some of the biggest promises of this video look a like over-statements to me. I'm guessing some of the claims are a bit over simplified and come across promising too much.

However, this type of system has got to be the way of the future for private communications (in my view). Also, this type of technology will probably render any political views and government actions against privacy irrelevant.

Ultimately, data privacy cannot be controlled. The right tools are coming.


I have spent time on the forums over at MaidSafe, where the majority of people are liberty-minded. Some, however, have their niche concerns and try to influence the development of the network to add features or remove features. One guy's over there trying to convince us all that all advertising is spam.

I am reassured by the fact that David Irvine (the founder of MaidSafe and the CEO (not for profit)) has repeatedly asserted that the network design will be built to be completely anonymous, distributed and uncensored (or rather un-censorable).

Which points do you believe are over-simplified, and which are promising too much?

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

Garan's picture


I have to admit to a fair amount of ignorance,
yet I have from time to time pondered this type of system
and occasionally read some of the technical literature
on similar information networking systems.

Maidsafe is new to me, so I apparently haven't been paying enough attention.

So, I can only talk with fairly broad strokes and probably can't offer irrefutable arguments. I'll just have to point and question.

Overall, the video on Maidsafe appears to be what I've been looking forward to. That being said.

1. "..you are completely anonymous"
The first one that comes to mind is that there are counter-efforts to the anonymizing feature of tor networks. How does Maidsafe deal with potential counter-efforts? (I should read this for myself, yet that is my first question/concern).

2. "You will never loose your data again."
The following counter-idea is a little fantastic, yet feasible with huge computing resources:
Could a massive amount of clients be hosted to first participate cooperatively with a Maidsafe network, then later be shut-down in an effort to discount Maidsafe's reputation through (effectively) deleting the distributed data (for those who happen to network exclusively with the nefarious servers, and not store their own copy ..cloud computing! :)

A lesser point is that (I assume) people can loose access to an unencrypted version of their data if they loose their private encryption key. I imagine this is a caveat to 'never' loosing data that would probably detract from the video presentation, so was omitted.

Most of my other questions are compounded by follow-up questions due to the fact that I haven't studied the details of Maidsafe.

For instance, when data is shared, are the patterns of sharing deducible?
Can the identity of data be tracked, even if one may not know what the data is?
If shared data is exchanged with a public key (again I have to read about Maidsafe), can a snooping party then generate documents of interest and monitor their transfer?

All these questions simply show that I am interested in learning more.
I'll have to study it.

Opening up the patents to the world is noble and necessary.
Maidsafe has definitely grabbed my attention.

No probs

All of my previous comment was essentially in agreement with the majority of what you said about implementation.

1) All IP info is stripped from any sent network information. Only the immediate connecting node knows from what MaidSafe node the comms are coming from, but it doesn't know the IP address or geo location of the node. They had to use the RUDP protocol to achieve this, and this is currently being rewritten, in order to better secure it and to improve efficiency.

Furthermore, there is much talk about counter-efforts, particularly here
This is being added to as they turn up any potential vulnerabilities.

2) I believe that this attack has been proposed on the forums, and that (don't quote me) the answer was that more than 70% of nodes would need to be taken over. The network nodes carefully police each other, and will isolate and remove any nodes that do not 'toe the line' IE they act outside the network parameters. Also, a node cannot read the data that it stores, and so even if I controlled 50% or more of all the nodes on the network, I couldn't choose which data would be stored there. So shutting them all off would cause the remaining nodes to reallocate the data, and because there are 4 copies stored for each encrypted chunk, these would be automatically copied to remaining nodes.

Aside from that, they have discussed this question on the forums (dev and maidsafe.org), and, assuming a large network user base, this would be prohibitively expensive.

if they loose their private encryption key

Absolutely. This needs to be stressed. If you lose your encryption key, the data is gone. At this point, that is the state of affairs (but we're not live yet, of course).

Also, bear in mind that the network would not lose this data. That would be the responsibility of the user. However, as people are realistic, there are discussions ongoing about this. Try this thread:

For instance, when data is shared, are the patterns of sharing deducible?

No. Intermediate nodes and storing nodes have no way of telling who data belongs to, and where it came from.

Can the identity of data be tracked, even if one may not know what the data is?

No. Storing nodes and intermediate nodes also have no idea where the data has even come from.
Quote from CEO, David Irvine:
"It would be pretty bad if we could identify where data is, as this would be a fundamental security leak. The network scrubs IP addresses immediately and data coming to you does not even contain the XOR address of where it came from. So you have no idea how it got to you or if it will ever come from there again (it very likely won't)."

If shared data is exchanged with a public key

Check out Fraser Hutchins reply in this thread:

Although this is not limited only to MaidSafe, it really is designed as a system, and can only properly be understood when taken together. I haven't quite finished that process myself. My biggest concern will be people's confidence in the network, and our ability to explain it to people!

I'm with you on the patents!

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

Garan's picture

Many Thanks

Thanks for your reply and points.

I just googled "fsf maidsafe" to try and see what Stallman thinks.
It looks like Maidsafe is going all-the-way with freedom.
No disappointments so far.

I'll study this further.

Competing currencies...

Competing currencies and Competing concurrency.

The path to the future.

This looks awesome

The battle for the future is between centralization and decentralization. Solutions like this are exactly what we need. I would like to see more content like this on the DP. I prefer solutions over politics.

Make sure I got this straight...

So I save a document i typed up in word. I save it using this system, it gets chopped up and sent to thousands of other computers. Now a few days later, I want to work on the document again...but half the computers my document was stored on are off. Only half the document gets recovered right? Now what?

If ignorance is bliss, Washington DC must be heaven.

There are two main types of computers

connected to the MaidSafe network, Users and Farmers.

Farmers are analogous to Bitcoin miners, in that they offer storage to the network, and are paid SafeCoin for successfully serving data retrieval requests. They are paid according to a reliability measure, which takes into account downtimes, speed, latency etc. So it is in Farmer's best interests to always have their computers running.

Also, as jonat3 already pointed out below, the network automatically reconfigures and copies any data that goes offline from its remaining storage nodes, ensuring that data is never lost.

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

The article addresses this.

The article addresses this. According to the article, there are four copies of your data floating around. If one computer gets shut down, there will be another node that will copy the data to a new online node. I assume they arrived at four copies through some probability calculation.

Thanks for the explanation

I watched the video, but didn't have time to read the article (didn't even notice there was an article).

Still a little leery about it though. I don't even use 'cloud' apps now, at least not to my knowledge. The idea of my stuff floating around is a little unnerving, even if it is broken up and 'impossible' for someone to steal.

If ignorance is bliss, Washington DC must be heaven.

No prob. And indeed, there's

No prob. And indeed, there's no such thing as "impossible'. But if it's highly unlikely, that's good enough for me. Will just wait and see how it performs first.


This is how SkyNet begins. :-)

Definitely keeping an eye on

Definitely keeping an eye on maidsafe. I think the basic premise behind maidsafe is solid and will certainly prove to be superior to current models. The question is if the execution can be done correctly. Ideas are important, but so is execution.

it will be interesting to see where this goes

The Internet has indeed outgrown itself so a change will be happening one way or another. The "cloud" was just a new marketing term for centralization. So this seems a lot like bitorrent...with a charge. So like a cross between bitcoin mining and torrents. There will be duplication like happens to keep torrents available when peoples machines go down. The data has to be somewhere else too.

What has happened with bitcoin mining is it started out very much decentralized but as it got to needing more and more power people grouped together and then the groups got more and more consolidated so now there is a threat of one group taking over the majority of operations. A similar thing is happening in the Internet Infrastructure...the consolidation of power.

There is a constant pendulum/force back and forth with centralization and it will be interesting to see how maidsafe will counter that. Where will it be vulnerable in the future.

Does it remind you of a hive?


Does it remind you of a hive?

MaidSafe is actually modelled upon the concept of natural systems; namely ants.

"Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!" - Patrick Henry

great idea but

i sold my maidsafe shares when the price went up. when they show more progress i will buy back in. there's lots of competition in the space now but none as comprehensive as maidsafe.
bitshares (dns) is moving very fast but they are all friends and helping each other. Storj is very interesting too but not as all encompassing as maidsafe and only focusing on one or two parts of the equation (p2p decentralized cloud storage). http://storj.io/
blockchain technology is a game changer.

Official Daily Paul BTC address: 16oZXSGAcDrSbZeBnSu84w5UWwbLtZsBms
Rand Paul 2016

Good stuff.


ChristianAnarchist's picture

Ok, I'm in...

Ok, I'm in...

Beware the cult of "government"...