Wait, The Safe Network *CAN'T* Be Used As A VPN?

Okay, so I was under the assumption that when using the safe browser or a Safenet extension for firefox , I would be benefiting from the anonymity of the network. The safenet would in that case be operating under a freemium model, where the people uploading personal data or would be fileshare app data would subsidize the people using the network as a free VPN connection to the old internet. However, I read in the “how facebook steals views” thread that it is impossible to view the old internet through the safe network. Is this true? Cause if it is, then holy hell that is a big deal.


Yes that’s true, SAFE is not an alternative to Tor in that respect. I can imagine it’s shocking for you, but as far as I know everyone else here is aware of that.

Tor is the best thing out there for that functionality, SAFE probably couldn’t do it better even if it tried.


On TOR you need to have an exit-node which loads/downloads the data coming from a website. I don’t know about you, but I wouldn’t like the fact that people use my IP as an exit-node. I also don’t really feel the need to use SAFE to surf the normal internet.

Think of SAFE as an alternative to Dropbox, Gmail, servers etc. And to let people get a feeling of it, a browser-plugin was designed. Just to make it easy to go to safe:decentralizedweb and test the technology. In the future we’ll see a SAFE-browser in the launcher. Possible with no connection to the normal web. For that I would just open chrome as I do know.


To my knowledge, Tor is too slow for anything besides whistleblowing and darknet markets, and VPN costs money. If Safenet could be used for VPN/Tor, you’d have two sizable markets using the network who would inevitably use the network’s storage/compute features. Now there’s just using the network as a backend for apps as the lead in to using the network. That, and torrents.

Oh well, I kinda figured there were going to be forks. Now I’m certain there will be.

1 Like

If the safe network is significantly faster than Tor, people making web sites that users will want to access anonymously will put them on the safe network, so when this happens, it shouldn’t be much of an issue.

It’s going to be very easy to put websites onto the Safe network, so there’s little reason this shouldn’t happen.

Perhaps someone could even automate putting web sites that belong to thirds parties onto the safe network, so if there’s demand for something on the safe network, the supply should be forthcoming.


Speed has actually always been an issue for SAFE and it seems so even post Rust. And its apparently not at all due to non optimized code. If SAFE had its own optimized hardware to run on would that make a crucial difference? I asked if the distributed compute capability would help much. We know from stuff like SETI that distributed compute capability isn’t necessarily a distributed computer or necessarily fast for anything but a very limited range of problems. It seems SAFE may become a distributed computer but its speed seems to be challenged beyond local regions for even for things like real time gaming. It seems SAFE is just a lot harder on any underlying hardware network tech and that is part of how it derives its safety so its going to be hard to beat straight line centralized services for a lot of speed tasks. If strange action at distance communication pans out maybe a SAFE net could all act like one coherant q computer with not penalty for wire distance or at least not have a speed problem.

1 Like

Wait, we currently have no figures for speed performance on the Safe Network. In theory it should be fast and even faster compared to today’s technologies. But until we see how the networks goes on a global scale I don’t think there’s any need to call it an “issue” until proven so.


Yes we will have to have some live data.

SAFE is expected to get its speed from parallel access to files. In other words when you access a 1GB file the APP/file system can ask for all 1000 chunks at the same time and the file is retrieved in approximately the time it takes for 1 chunk. Then your connection to the internet’s download speed is the slow part.

For sequential communications the delays may prove that we will need a combination of direct communications & SAFE comms for a few specialised applications.

VPN & Tor replacements are specialised comm apps tailored for the current internet, and to duplicate on a “parallel” access web that has greater lag may prove unbearable.

As to using the SAFE network as a tor system, it is possible, the person setting it up will have the same issues as the current tor network has and that is getting prople to set up relay & exit nodes. Otherwise there is not reason it cannot be done. VPN can too, but it would be very slow since its interfacing two different comms system when leaving it on the traditional internet would be fast.

Another way of considering this is that TOR & VPN are for the traditional internet and they are tools to assist in anonymity/security on it. SAFE network does not need those tools for anonymity/security on SAFE since that is built in. In other words SAFE is its own TOR/VPN built into its infrastructure


This would not be a fork of the safe network, it would be a whole new set of functionality, a new project that would not be competing with Safe as far as I can tell.


Such a project would likely be using code from Safenet’s source code, a modified version of safecoin and the file parsing and distribution code but with physical based addresses instead of XOR addresses come to mind, and would take users away who were using due to a lack of choices away, so that sounds like a competing fork to me.

The XOR space is the universe where all the data of the SAFE network live (including safecoin), without this space nothing works.

Correct. It is required to have IP address to plug your brain into the network.

CJDNS would be a great alternative for VPN service. Build a meshnet community, and have few vpn boxes that allow the meshnet community access to clearnet. These vpn are connected to corporate ISP. Request an IPV6 address for security. If it is IPv4, then I wouldn’t advise to use it since it leaks your ipv6 public key to the world.

Unless there’s some high level techno-stuff I’m missing, the theoretical fork would house data based on the actual physical machines and their locations rather than whatever Safenet does. It’d be less private, but it’d be fine for everyone not needing to hide the very usage of the fork from NSA-level entities, i.e. >99% of the internet.

Then fork freenet for your project as freenet already stores data that way - and be prepared for very slow transfers of large data.

I expect that it is technically possible to make an app that could function as an outproxy for SAFEnet – but the speed would be worse than Tor - and why would you want to use it at all, when we already have Tor?

SAFEnet replaces the existing insecure-by-default internet with a secure-by-default internet. That’s it’s function.


The “whatever Safenet does” is place users in the XOR space as XOR address, store data (both chunks and SD) in specifics XOR spaces, retrieve those data using XOR addresses and send messages and orders in the XOR space so yes, there’s some (not so) high level techno-stuff your missing.

1 Like

You can use an app though within the safenetwork though to view the current internet no? The app developer just need to set up a server to relay whatever you request from his current server, capable of accessing the current internet, to your safebrowser

Only a badly developed app IMHO. In SAFE any links out of SAFE are … well not safe :wink: