4 ms·
Nothing about "put it on a blockchain" helps that. If you are actually putting the page contents onto the blockchain, you have basically constructed an append-
by Kadin 5y ago
Nothing about "put it on a blockchain" helps that.
If you are actually putting the page contents onto the blockchain, you have basically constructed an append-only, replicated database. Which will almost certainly become unmanageably large over time. (Why would a user want to maintain a local archive of this content? There's no incentive for them to keep it on their machine, aside from wanting a copy of the content for themselves, which they could achieve in any number of other ways. Like just... saving it to disk.)
If you aren't actually putting the content into the blockchain itself, but are really just storing a URL that points to the content, you haven't solved the linkrot problem at all. You've just created a very roundabout way to send people URLs. You could put something like a PURL in there, or a DOI, but that just moves the problem around. Again, the blockchain isn't solving anything.
Here's an easier solution, if you want to try and ensure the availability of a piece of content into the future: strip it down to its most minimal, plaintext version, and post that to a well-known (and subject-appropriate) Usenet group. It will immediately be archived in several places across the world, which have a pretty good track record of keeping ahead of the flow of incoming content and keeping it available. No, you don't get a guaranteed URL, but the Message-ID is a very unique search string that can be used later on to find it.
- thebean11 5y agoYou are arguing against a point I didn't make, but I do want to clarify the problem a little bit more since it is pretty interesting. I agree that storing images or even articles on a blockchain does not scale; that level of data redundancy is a humongous waste for something like a tweet. There's two issues at play here, immutability and availability. Immutability is not too hard to solve with a blockchain. You can hash pretty much any piece of digital content, store the hash on-chain, and prove that the content (hosted elsewhere) is identical to the original content (assuming the chain is not susceptible to massive block re orgs). Availability is trickier. There are blockchain-related projects that claim to solve it in a decentralized manner by incentivizing a group of nodes to store it, and then periodically submit proofs that they have it stored, but I'm pretty fuzzy on the details there. edit: Want to make it clear that I don't think blockchains are the only potential solution to these immutability and availability problems. That being said I'm pretty optimistic that crypto can solve these problems in a decentralized way, sounds like you feel the opposite, but let's try to not let it devolve into the stereotypical HN crypto flame war..
- topspin 5y ago> Why would a user want to maintain a local archive of this content? When I think of all the things that once existed and have now vanished forever for all sorts of reasons it is very appealing to imagine having a local archive.