Hi. I was wondering how maidsafe encrypts public data

As I understand let’s say when someone publishes a site with the last episode of game of thrones, when I access it I am anonymous. But how? Isn’t data public so cant HBO access safe network download that file and search who downloads that file through ısp network couldn’t they find it like they find who torrents?(totally hypothetical)
I remember reading somewhere we don’t want data from the network directly but parts of data from our close group.
So do we disrupt the searches by getting different parts from different vaults? İs public data encrypted, if it is how a usual user decrypts it
Sorry if I sound dumb. But this question has been bugging me for couple of days

Public data is still self encrypted. So its encrypted.

The data is encrypted in route so no noone can listen at the ISP. The same as the ISP cannot read your banking pages.

user [quote=“Seneca, post:2, topic:6728, full:true”]
I understood from that previous discussion that the relation between the client, relay and client manager group is onion-like, though only with one layer. The relay doesn’t understand the payload (presumably encrypted), the client manager group doesn’t know the IP address of the client. The communication from relay to client managers is routed through the DHT, they’re not directly connected. Would love to get some additional clarity though.

How does relay doesnt understand it, but user does. that is my exact question actually.
or was that topic was too old and public data protocols changed?

One possibility is that the node that forwards it to the relay first encrypts it with the public key of the recipient (user).

1 Like

so if I understand it right

for every piece of data
user sends public key with data request to relay by encrypting it with relays public key
relay decrypts request and sends it to the network
Network doesnt know whos public key requested the file ( ip ) but encrypts it with it end sends it to relay
relay passes the file directly to user
user decrypts it it with his private key

I guess if we divide the data into pieces since relay doesnt know what user request from other relays it cant piece together data neither isps

so we know no matter how small, files are always divided?

Here’s an old post that represents what you’re wanting to know quite well.

Some precise details may have changed, but the post is very good at giving you the high-level picture.


Some thing I have always thought about was if you could probe or observe the network via a known chunk to yourself by trying to upload it or simply upload it into the network and see the results, replies, response time, and thus deduct if the network already posesses your chunk (read: your whole file, your precious data, your leaked stuff, your copyrighted whatever, your insert here).

does the network take precautions against such observation or timing attacks, response time probing or similar? will the network always take in the to-be uploaded data no matter what and consume it for nobody to be able to draw any conclusions from insertion behavior?

will the network be insert agnostic so to speak? thanks and cheers.

Read the post I inserted above and see if it doesn’t answer your question. I think it does.

None of this, the vaults and how they operate, has started testing yet. Not publiclly anyway. So we dont know how well it will work, or if it will work at all to be honest. We all hope for this to be smooth and quick in its implementation but if that is going to be the case we dont know. Worst case, it will be years and years with problems before this will work in the real world with errounous hardware and non cooperating and even disruptive parties also taking part in the network.

Its the most important aspect right now that this part goes to testing so any problems can be worked out. I expect changes to functionality here once the theories meet the real world.


Yes it has. I ran a vault in the early tests on my rubbish connection. I then missed out on running vault tests in a later version because my UL speed wasn’t fast enough.

We have seen them work and they did work. Vaults needed optimisation, but they certainly do work. I’ve seen it with my own eyes. :slight_smile:

Vaults are not at all ‘theoretical’.


As @Jabba says

AND what do you think the alpha 2 network is running on? Its running on Vaults. Just those vaults are operated by Maidsafe to allow them to contain errors and measure them thoroughly. Once vaults are at home again the bugs then have to be reported by the public then duplicated by Maidsafe etc etc. Very long process compared to being able to see the bugs on your own system. So Maidsafe are running them themselves at the moment. There are other reasons because they want to implement a little more into vaults before releasing them for home vaults again.


Mhh, this is not correct.

I did run several vault instances during public tests last winter and spring, together with many community members.
Anyone interested could check for instance : How to set up a vault from home
or type “vault”, or “community network” in this forum search.
Plenty of vault experiment returns on these treads.
Also, about “how they operate”, one can check : https://github.com/maidsafe/safe_vault, which is public.