Is the Safe Network trustless - and if so, could someone please explain how? I know this might be kind of a big question, but I’m just trying to wrap my head around some things.
Trustless is an adjective:
1 Having no trust or confidence; unbelieving; distrustful.
2 Not to be trusted or relied on; unfaithful, unreliable, treacherous, untrustworthy.
Which would likely lead some into a confusion.
You might wonder that SAFE requires no trust but that’s not the inverse of requiring trust.
Trustless is like a negative and the reality of SAFE and other distributed systems is they require zero… trustzero isn’t a word but if it was
As for the how… well you control your data; your system fragments and encrypts your data; and uploads into a shared space that is the network. To be sure of your privacy; security; and freedom, requires zero trust because you are the only one that can access it.
If you want to extend this to faith… then the network requires a bit of faith in the its being stable and that confidence will follow from network size. After a point there is no need for faith and the robust nature of the network will see it defend your data alongside others. Best route to that is to contribute resources to the network and make it stronger for yourself and everyone else.
It is a decentralized network, and you and only you control the credentials to access your account on the network. So, yes!
Oh no I understand the concept of trustless as a network paradigm, I meant the network itself. While my understanding of how DHT works is limited, from what I gather at some point you end up connecting directly to another member of the network. In my mind this potentially opens you up to being identified, or attacked (DoS, etc). I was wondering how the network has the property of being trustless in a real-world manner, if I’m expressing that correctly?
My personal guess was that each of the packets gets routed to different nodes before finally ending up at their destination, but I’m not sure how close that is.
Everything is fragmented, so it’s not a simple one file is over there and you make a request and it gets channeled back to you. It’s a grey goo and no-one can make sense of who is accessing the network for what - even the network in an abstract sense does not know who you are or what data you can access… it’s like Bitcoin in that regard everyone can see all the transactions but only you hold your private key, only you can see your data and make use of it.
It’s a step up from Tor then and it’s built mindful of DDOS and likely able to provide a more robust response to that - DDOS usually requires a target but if you cannot see the target you are attacking the whole… and the larger the network, the harder that becomes.
From the file perspective it totally makes sense to me, but I’m thinking more like, if you wanted to do a voice call like Skype or something. I might not be understanding this fully as well, so apologies if that’s the case.
I don’t know enough how the live streaming will work to be exact but if you fragment the stream into a thousand pieces and throw it to the farthest corners of the globe and back again… who’s to know that you are not just downloading any other file or more simply just supporting the network with your hardware. You are more likely to be caught out by real world efforts to consider that two people go indoors at the same time and then go out and have a beer at the end of their phonecall, than ever see the detail understood well enough internal to the network.
To be seen perhaps if bandwidth flux is a suggestion of network usage or if that is a constant… relative to the user being a resource support or not but then hard perhaps to even know that the traffic is on the network and not just other https traffic etc.
In my opinion the plan should be to make the SAFE network completely trustless. Even more trustless than for example Bitcoin which is susceptible to change through some “politburo”.
Therefore the ability to upgrade the SAFE network core software on the protocol level so to speak should only be a temporary possibility during the Alpha and Beta stages. The production release of the SAFE network should be like the IPv6 specification. Unchangeable.
There’s a lot to be said for that… the politics around Bitcoin is a lot stronger than most anyone expected initially and politics is always manipulated and corrupting over time. Natural systems shrug off politics and just keep going.
I don’t know reasons to be fan of IPv6… I always rather liked Ethernet as an example - it just works.
I only used IPv6 as an example. A standardized protocol. Ethernet is another example of a protocol that has the capacity to become ubiquitous due to the absence of ability to modify it.
Of course the SAFE network is a much more complicated “protocol” yet the idea is the same as for lower level protocols.
Don’t know if it will be what you need, but I recommend looking at this post: All the encryption layers for SAFEnet.
A few aspects may have changed slightly with development details, but the data is effectively still true. It’s part of the underlying design of the network that makes it special on the trust/trustlessness level…
Good point. I came to think of another term: trustfree (similar to your word trustzero). The meaning in a computer context is the same as trustless but perhaps less confusing.
The safe network is also trustless.
… there’s no single point of failure or authority or control. You don’t have to trust anyone. You yourself can join the network any time and leave any time, no lockin, no contract, no bans, no trust.
Yes, trustless, in a way no current dictionary can efficiently deliver. Context is everything in linguistics.