Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

A blockchain, when applied properly, can provide levels of trust in a way that is not directly associated with the proof of work they run. Bitcoin uses SHA256 for POW, which is pretty well understood by now to be secure and hovers around cost efficiency.

Using the trust embodied in the blockchain allows for the mapping of comment, author, topic commented on, participants in the topic (a sub-aggregate) and a general context in which the topic is shared (web link and/or IPFS link). This can be done without making any of those things part of the POW.

Yes, anonymous comments and topic submissions can be made, but those may have a lower associated karma attached to them, similar to what we have here in HN. With a bit more code (Script), one could add voting based on value (i.e. I could vote you down with $20 if I had over $20 of karma associated with my account), transparency in voting (you could see it was someone with high karma commenting on x/y/z articles voting you down), and any other analytics anyone thought would be valuable in determining trust levels.

BTW, I am not indicating karma is a one-to-one mapping with value injected into the blockchain by market growth or trades. For example, I couldn't vote you down with $200 of fiat just because I didn't like what you said. I would have to "earn" the karma from upvotes from other users over time, which would be tied to my source address used to pay for voting on other users. When I actually went to vote, a portion of the revenue would go to the platform operator (HN equivalent), a portion to the submitter of the topic, and a portion to the person I voted up (or down).



So what you're saying is that basically you have verification that the conversation you see is one that took place and that there are a fixed set of identities (not people) who had that conversation?

Can't we just schlep a merkle tree over the top of any forum software and get similar guarantees? And isn't this sort of false due the lack of convergence of the block chain over long partitioning events?


You could totally do that with IPFS today, but I think the problem is that social commentary occurs in a centralized location. If you play Rust, the analogy would be being able to talk to someone while within a given zone.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: