I have come across the Blockstack project, and I wonder how different it is from SAFE. As they state:
Blockstack is a new decentralized internet where you own your data and apps run locally without remote servers.
With Blockstack, users control their data and apps run on their devices. There are no middlemen, no passwords, no massive data silos to breach, and no services tracking us around the internet.
Sound just like SAFE!
A clear difference however is that it is based on a blockchain. But, is this it?
Sounds very much like they donât actually secure the data itself, just identifiers. That is not really the same thing as âowning your own dataâ. Happy to be corrected by someone more technically literate.
âThe top-most layer (layer-3) is the storage layer, which hosts the actual data values and is part of the data plane. All stored data values are signed by an owner key deďŹned in the control plane. By storing data values outside of the blockchain, Blockstack allows values of arbitrary size and allows for a variety of storage backends. Users do not need to trust the storage layer and can verify their integrity in the control plane. Our design beneďŹts from the performance and reliability of the backend cloud storage systems.â
SAFE is pretty radical, I donât think youâll find anything like SAFE despite hearing very similar sounding features touted in a lot of blockchain projects.
Iâd need to read more about Gaia, but it also doesnât sound like the aim is to release spare capacity to run the network. The incentives model is quite different at first glance.
It doesnât sound like much of a viral formula to me tbh. Itâs just more of the same, blockchains seeing the use-cases and desperately trying to fit the square pegs through the round hole.
Not sure, but point 1 means the games bust. The user should not know where the data is geographically and worse on a specific computer as that is actually bad. Interesting to have these discussions though. It is why we publish network fundamentals. They are vital otherwise bad stuff can be seen as features or good things. To me data locations is a central issue to defeat.
I was just reading about blockstack and their whitepaper as well some tonight. One of the first things I noticed was the predictability of the location of user data. I was pondering if SAFE could go even further than just unpredictable storage on a first PUT is possible.
Was thinking it may be possible to even implement a âdata in motionâ approach where encrypted stored data is actually in motion, never hosted on a single vault for extreme amounts of time(regardless of vault uptime/health), say days or so. And could move in an unpredictable way different location at a different interval of time for its ttl on that node. Could be beneficial to the network too by letting lots of vaults do the lifting than just the ones that first received the data in the first place. Also with how complex analysis tool and packet capturing and tracing gets especially in the future, it adds extra complexity on establishing packet flow captures if nodes get compromised(some sort of tracing injectors dropped in, that canât see the data but are trying to establish probable pathing from say a source to destination for some safe X site) when the environment constantly seemingly changes around it. I think some optimizations around data that is not accessed very often moves less, and the most popular data moves more frequently. Obviously not anything to be done in an mvp mode but maybe something to ponder for the backlog if there could be security or network value there.
I think SAFEnet is already like this. Vaults may get disconnected, then to keep data redundant, it is duplicated to another vault.
I wonder, whoâs paying for all Blockstackâs S3/Azure/Google servers? And is it possible to move your data around once itâs been created on some hub? Is it automatic, or you have to copy it by yourself?