The Internet is About to Get A Lot Safer

MaidSafe has launched the Fleming public shared testnet. This release demonstrates the remaining technical features necessary to prove the Safe Network’s viability. The world’s first autonomous data network, The Safe Network is both a decentralized internet and dApp platform. If you ever use the internet, you’re about to get a serious upgrade. Here’s what you’ll get:

  • Data Ownership! Enjoy private, anonymous, and secure browsing because you must grant express permission for anyone to see, use or transform your data (which is encrypted in transit & at rest by default)
  • Data Permanence! Default pay once; store forever policy means your data will never get lost
  • Rewards! When you provide resources (i.e. unused hard drive space, processing power, and bandwidth) to and develop dApps for the Network, you’ll earn Safe Network Tokens
  • Asset Tokenization! create and issue custom tokens to better control the token economics of your own projects
  • Atomic Swaps! directly exchange tokens peer-to-peer

All these benefits just scratch the surface of why the Safe Network will revolutionize the decentralized landscape. Because the Safe Network does not run on a blockchain, Safe Network avoids many of the capacity and throughput limitations that blockchain-based attempts to deliver decentralized solutions face. Visit the forum to learn more about the Safe Network’s uniquely sustainable and scalable approach to solving many of today’s most pressing technology challenges—from data privacy to cybersecurity.

So, what’s new with this public shared testnet, and why’s it such a big deal? Not only will this testnet showcase the remaining essential functionality to prove out the Safe Network’s viability, it’s also loaded with a ton of innovative tech features. The upcoming release will introduce:

  • Section Splits: Allows the network to grow naturally by automatically splitting nodes and data into new sections, ensuring that the new sections contain enough nodes to handle data requests and adequate storage space.
  • Rewards/Farming: Individuals who supply resources to the Network will be rewarded with Test Safe Network Tokens. (Note - these are only test tokens and will not be tradable for real Safe Network Tokens at launch. The only way to get Safe Network Tokens pre-launch is to get MAID now.)
  • Accepting New Nodes Only When Resources Are Required: Allows existing nodes to decide when they want new nodes to join the network, which helps to prevent Sybil attacks because an unlimited number of nodes cannot join the network at will.
  • Section Chain: (A list of section keys linked together with signatures) proves that a piece of data was signed by a group of nodes that were once valid members of a section, even after those nodes are long gone.
  • Fork Resistant Section chains: If there is a disagreement between sections on which key to append next, section chains can now fork while always presenting a unique and deterministic order of the chain blocks to the outside world (i.e., reach consensus).
  • CRDT (Conflict-free Replicated Data Type): A new implementation of the section chain data-structure that guarantees (eventual) consistency regardless of the order in which the operations are applied, how they are grouped, or even duplicated (i.e., allows everyone to agree without involving any complicated consensus mechanism).

Here are more details on the upcoming testnet. Be sure to follow @Safenetworktech and keep your eye on to find out how to join the testnet when it launches in early Q2 2021. If you’re more technically inclined and familiar with CLI, joining this testnet will give you a deeper understanding of the mechanics on which the Network runs. If you’re less technically inclined, fear not! MaidSafe considers effective UXD a priority, and is refining both the browser and app to deliver an easily accessible frontend experience for any and every one to use. Here’s a sneak peek at that user experience, and stay tuned for more updates on UI/UX developments.

Remember, the only way to get Safe Network Tokens before the beta launch is to obtain MAID. At launch MAID will be exchangeable 1:1 for Safe Network Tokens. While the exchange footprint for MAID is now expanding to include P2PB2B and will continue to expand in the coming months, MAID is also available on:

  • Bittrex Global
  • Changelly
  • HitBTC
  • P2PB2B
  • SimpleSwap

You can find more information on how to buy, move and store MAID on the community forum.

As a 14-year journey to restore privacy, security & freedom to the internet draws towards its destination, now is the time to hop on board! We can’t wait for you to join us.


Did I miss something?


The Safe Network will allow independent projects (e.g. JAMS, PDC, etc.) to mint and swap their own tokens. This asset tokenization could also be extensible to other use cases, like NFTs.


No, it’s not, and it’s extremely confusing when people say that, as well as sounding clueless.

The Internet is fundamentally a point-to-point real-time communication system. Safe is fundamentally a distributed data store. Safe does not provide the functionality of the Internet, and in fact Safe relies on the services of the Internet.

The Internet is not the Worldwide Web, and Safe doesn’t even replace all of the Worldwide Web.

1 Like

I think that’s a superb announcement/into article for early adopters, influencers and traders. Nice work @Sotros25


If you consider the “Internet” as only the hardware infrastructure keeping us all connected, sure. Safe does, somewhat, replace the TCP/IP networking layer above the hardware layer, to fundamentally change the way networking and communication is done.

It’s definitely selling it short to say it is simply a distributed data store.


On the contrary, safe network replaces much more than the Worldwide Web.

Safe Network provides routing, name resolution, encrypted persistent data storage, encrypted data transit, self healing, etc.

The browser element is just an app that sits on top of it, providing WWW like functionality, without much the the pain of it. Many Safe Network apps will not use a browser, including the Safe CLI that is arriving with the testnet.

I suggest reading the primer for more background:

If you want to understand the specific layers that the Safe Network replaces, see here: SAFE Network Primer


Thanks, @happybeing! My goal was to make this as accessible as possible :smile:

Also, thank you to both @wydileie and @Traktion for explaining why Safe very much is a decentralized internet (and more so than the many other projects that want to claim that title/functionality).




Great post! Someone cross posting to Reddit for some additional attention?


Congrats guys, can’t wait to start using this thing for real! Time to ditch the ole spying services!


If you consider the “Internet” as only the hardware infrastructure keeping us all connected, sure.

“The Internet” basically means the IP catenet… and, when speaking more loosely, all of the applications layered over IP. That’s what it’s always meant. It’s not just hardware.

Safe does, somewhat, replace the TCP/IP networking layer above the hardware layer,

No, it really, truly does not.

  1. Safe runs over IP. If you remove IP, Safe completely stops operating. That’s not replacement in any sense or to any degree. Safe is a layered application on top of the Internet, or, if you prefer, a new part of the suite of Internet services. Safe does not run on the “bare hardware” and is dependent on much more than the hardware to keep running.

  2. Whatever point-to-point real-time communication you can do over Safe is qualitatively slower, qualitatively less efficient, and has qualitatively less predictable timing than the point-to-point communication you can do over TCP/IP. There are enormous numbers of applications that can’t tolerate any of those things.

Sure, I’ll agree with this.

I’d agree with this to an extent, which is why I said “somewhat”. From a user and system level, IPs are not part of the equation. Obviously routing will require the IP infrastructure.

P2P, yes, will be qualitatively slower, by necessity. However, “less efficient” is an interesting qualification. There are definitely applications that would be much more efficient in a decentralized network than even P2P communication. Pulling down a hosted website from Japan while I’m in the US, can be faster with a decentralized network. So yes, while there are certainly applications that rely on P2P communications, and low latency (such as video games), there are a number of applications that would see performance increases, as well. Perhaps many of the ones that would be slowed down can be reimagined for a decentralized network, with some limitations.


This is great!

I think the language and everything about this works really well as a press release, and a basis for future things of a similar nature. Thanks @Sotros25, and for whatever work you’ve done getting the exchange listing, which is what I assume this is.

The only thing I wonder if people might want to hear more about is a bit more sense how data ownership is achieved in relation to the current internet, and maybe even more benefits of it. eg. no servers/middlemen, ease of publishing. That’s maybe just the bits of Safe I’m personally looking forward to using though!


On the contrary, safe network replaces much more than the Worldwide Web.

My point is primarily about the Internet, but this is still not true.

The definition of the “Worldwide Web” is pretty nebulous, but I’d say it includes anything that uses HTTP and anything that uses URLs. It definitely includes anything that runs in a browser, but I don’t think a browser is a necessary element.

I agree Safe provides services that the Worldwide Web doesn’t natively provide, mostly in support of assured data availability.

But Safe also lacks services that the Worldwide Web does provide, and those services get a lot of actual use. Therefore Safe, as it stands, cannot entirely replace the Worldwide Web.

You wouldn’t, for example, stream real-time video from your surveillance camera over Safe, or control the PTZ. Maybe you could. I’d want see it demonstrated before I removed the “maybe”. But even if you could, you shouldn’t. It would be insanely wasteful and pointless, and it would make the whole thing work worse. Even HTTP is really a waste for that, but Safe is completely over the top, without adding any value that’s meaningful for that application.

In fact, as far as I can tell, Safe is unsuitable for more or less any real-time or near-real-time application. And I’m not so sure it’s usable for any application where you actively prefer that the data not persist, either.

Safe Network provides routing

… but doesn’t expose it in any way that’s useful for general-purpose communication. It’s basically there to support the data store.

Routing in the Internet is visible to the user and to the apps that represent the user. A user can direct some kind of request to host X, like say trying an SSH login, and the Internet will try find a way to get the underlying messages to that host. And the service can be anything, not just data storage or retrieval

Routing in Safe is not visible to the user, and not really visible to apps, either. The user asks for a data item. Safe will find the data, but it the user has no idea of nor control over what Safe talks to in the process.

There may be a capability buried in Safe somewhere that would enable sending arbitrary messages to specific nodes, or locating and using non-data-storage services involving arbitrary sequences of messages, but I don’t believe that it’s really available for apps to use in any practical way.

name resolution,

There is name resolution in/for the Worldwide Web. You can’t meaningfully use a URL without name resolution. DNS is a service that’s available on the existing Internet and is heavily used within the Worldwide Web, even though DNS itself predates the Web.

Even if you don’t see DNS as part of the Worldwide Web, it’s still universally available anywhere the Worldwide Web is used, so Safe isn’t adding a new service from a Web user’s point of view.

Safe adds decentralization and (one hopes) censorship resistance and permissionlessness, but name resolution in itself isn’t new with Safe.

encrypted persistent data storage,

I think I know what you mean here, but I wouldn’t say those were the most salient words to use to describe the new value that Safe adds. An encrypted hard drive is also “encrypted persistent data storage”.

Again, Safe’s contributions are more in the area of robustness and censorship resistance.

encrypted data transit,

I think TLS is now sufficiently embedded in the Worldwide Web protocol stack that it’s just plain wrong to say the Worldwide Web doesn’t offer this. Few major Internet applications, Worldwide Web or otherwise, can’t provide encryption in transit in 2021.

self healing,

That could mean anything.

I agree that the suite of stuff usually thought of as the “Worldwide Web” doesn’t provide anything you could call “self healing”.

The underlying Internet IP routing infrastructure does, of course, provide a huge amount of self healing, although what’s being “healed” is the ability to provide communications between point A and point B, rather than the availability of information unit C.

Many Safe Network apps will not use a browser,

If you take the Worldwide Web to be roughly HTTP and URLs, then many Worldwide Web applications don’t use browsers, either. It’s not just a hypertext system. It stopped being just that within weeks to months after the first HTTP servers and clients appeared.

And there’ve been CLI clients for HTTP and whatnot forever.

I suggest reading the primer for more background:

That’s kind of presumptuous of you. It turns out I have read it. It told me that Safe does not replace the Internet and does not really serve all of the same purposes as the Internet.


So this is a new listing on an exchange? Great news!!


Will a mechanism for automatically paying devs when people use their apps be in the testnet?

And this asset tokenization and atomic swaps will be in the testnet?

Will the permission system be in the testnet? I thought I read it wasn’t.

In official communication I think it’s very important to not imply any more functionality than there is, to try as well as possible to convey reality. To me, having a list of features where some of them are in the testnet and others are not, is not a good representation of reality.

If you want to include potential future features I think it would be a good idea to separate those from the features that have been implemented by placing them under a different heading for instance.

I don’t mean to come across as dismissive, really nice work writing this! I care, so I give feedback. :slight_smile:


The article is structured to speak to key Safe Network Tech features up top:

The high-lighted features are indicative of the experience after the beta launch when real Safe Network Tokens actually exist. For example, Asset Tokenization and real Rewards (to farmers or developers) cannot exist without real Safe Network Tokens.

In the second half, the article covers testnet features:


It’s odd that I wouldn’t understand that after reading through it several times. I guess maybe it’s wording such as “If you ever use the internet, you’re about to get a serious upgrade.” that makes it sound to me like the coming list is regarding the testnet. I don’t view 1.0 to be about to be released, but that’s subjective of course.

Headings might be a good idea to make it really clear what is what, and to not leave nice wording such as what I quoted above up for interpretation. Or maybe just more explicit wording around the feature lists?

That’s the feedback I got, thank you for accepting it and nice work once again! :smiley:


If you have read the primer, you will know the layers that Safe Network claims to complement or replace. You likened it to not even being a WWW replacement, which grossly simplified what this technology achieves.

Moreover, I did not say that Safe Network is the first to do name resolution or encryption. However, DNS is not WWW. It is external to it, much an NRS is on Safe Network. Therefore, Safe Network provides more than just a subset of WWW by definition.

Regardless, you seem keen to minimise the achievements, without any real reason. The technology is clearly pushing many boundaries, yet you seem more interested in poking holes in a layman’s statement of what it does.

Do you think there is an easier way to describe how replacing/complementing 3 OSI model layers of the Internet will feel like for users? Do you think they will get sniffy if too much or too little is changed to suit your specific categorisation? Do you think they need to have such excruciating details explained in a press release?

If they are technical, they will come here, read the primer, grok it, then either like it or pass it by.