I am confused on one thing with HDD mining. What is the disk space being used for? (what service is it providing and to who)?
Only for data generated by you, that is needed to mine burst.
Let me try again. I don't understand. If I mine Burst with my harddrive space, who is using my harddrive?
The whole community I think.
incorrect. "No one" is the right answer.
As of yet, the space dedicated to BURST mining, is solely for mining. it is not "made use of" in any other way.
However, down the line (it's in the roadmap) we will be adding filestorage to the coin. THEN you will be sharing your drive with others. At that point I will be able to explain to you how it works better.
I think the idea is that you will tell your burst system that it can use, say, 1TB on your drive C and 2TB on your drive D (or whatever) and then burst will make a directory on those two drives that is then used by the burst system. Yor harddisk is not shared as such, only the contents inside those two directories is accessed by the burst system. Much the same as today, but instead of plot files, the miners will create files with information in them, some of it hopefully used for an anonymouns distributed file system.
i think the features and capabilities are still out in the open, but as burst already has users with thousands of terabytes of storage dedicated to burst, that userbase can be turned into a giant cloud service pretty easily.
i imagine the following features would be practical :
anonymous decentralized (government intervention) safe file system :
- user cannot deduct which miners hold parts of his files
- miner cannot deduct which users files he holds
- miner cannot deduct what is in the files he holds
- intermediaries between someone requesting a file, and someone storing a part of it, cannot see what file is being transferred or what the contents are
- files are stored in a tree like structure, perhaps with an addition of a tag cloud per file
- files and directories can be password protected with some sort of key (not transferred, but needed to reveal contents)
- files can be stored such that neither tree placement or filename or contents can be known by other than the person storing
- people storing files, and people reading files pay some transaction fee, and people mining collect this fee as payment for the space they rent out
- files are stored in many places in many small blocks, so a lot of nodes have to drop out of the network at once before a file gets lost. The individual seeder himself of course always has his own local copy too.
- the network somehow detects if a block is getting scarce and then creates more copies of it.
Whoever creates this, will have to read up on a lot of research into distributed file systems, read up on how torrents work, read up on basic cryptology, and be a bit smart and creative on top of that.
I think it should be optional if a user wants to offer filesystem resources with his harddrive space or just plain mine for the blockchain.
the filesystem as described above will do to storage what bitcoin did to finance.
Ohhhhh ok. Thanks for straightening this out . It sure seemed like there was no mention of providing utility (yet?)
Anyway, I was daydreaming about this in class the other day as we we're learning about different storage protocols and HDD arrangements (various RAID configurations) and realized that you could create a distributed file storage systems where people's hosted information is striped across all disks/hosts within a particular RAID set. With enough redundant parity bits, any file could be reconstructed with a bit of computation whenever the file is needed. Of coarse you would use some sort of cryptographic privkey to gain access to the location of each bit you need when you ask the network to access your file. In this way, nobody can reconstruct anyone else's data because each host doesn't have a single file in its entirely. I think what I just described is maidsafe
Show Posts
Anyway, I was daydreaming about this in class the other day as we we're learning about different storage protocols and HDD arrangements (various RAID configurations) and realized that you could create a distributed file storage systems where people's hosted information is striped across all disks/hosts within a particular RAID set. With enough redundant parity bits, any file could be reconstructed with a bit of computation whenever the file is needed. Of coarse you would use some sort of cryptographic privkey to gain access to the location of each bit you need when you ask the network to access your file. In this way, nobody can reconstruct anyone else's data because each host doesn't have a single file in its entirely. I think what I just described is maidsafe 
