3 ms·
Yes, that is a possible application of "blockchain technology", given a broad-enough definition of what constitutes "blockchain technology". But the consensus-
by Kadin 8y ago
Yes, that is a possible application of "blockchain technology", given a broad-enough definition of what constitutes "blockchain technology".
But the consensus-based proof-of-work (or proof of storage, or proof of memory, etc.) models aren't really necessary or even clearly desirable.
There are timestamping systems dating from long before "blockchain" (at least 1990, probably before) which involve sequential hashes, with each subsequent hash also including the previous one (in some respects this is a very simple blockchain), and then if desired you can roll them up through a directed graph to one master hash produced periodically and published immutably -- printed in a newspaper was one approach, or posted to Usenet, but you could think of others -- such that it validates all previous hashes as having been created before a particular time.
Guardtime was doing this back in 2007 or 2008 and printing the top-level daily or weekly hashes in the newspaper; I noticed they are now marketing themselves as a "blockchain" company, which is astute marketing (and I'd do the same if I were them, probably), but the ability to do secure timestamping existed in several forms without blockchain.
The reasons these systems weren't used have less to do with technical feasibility than lack of pressure to implement them.
- rusk 8y ago> sequential hashes, with each subsequent hash also including the previous one Isn't this "in principal" what a blockchain is? Just because they didn't call it a blockchain back then doesn't mean it isn't. You could also consider Git to be a blockchain of sorts. What distinguishes Bitcoin in particular is proof of work which provides a means by which distributed untrusted actors can transact in a trusted way. The actual chain of hashes just records these transactions.
- munchbunny 8y agoThe answer to that is tricky because it depends on intent. Taking an expansive definition of blockchain as a log of sequential, cryptographically verifiable transactions seems okay enough. At the point where we're nitpicking what is/isn't blockchain, that's not a useful exercise, because the real question is whether this is the right solution for the right problem. On the other hand, "blockchain" is a highly loaded term, and choosing to label something on the margins "blockchain" means diffusing the legitimacy of application onto the marketing term, or vice versa. In that sense, I'd be hesitant to lump this marginal example into "blockchain" specifically instead of just "cryptography stuff".