TL;DR at the bottom.
To preface this, almost no legwork has been done for this project. I am not a dev and I have a career and a family, however, I've been in crypto for awhile and I am in the information security industry. Right now, I'm just a guy with an idea. TL;DR at the bottom
A global database of interconnected argument maps (https://en.wikipedia.org/wiki/Argument_map) written to a highly available, publicly readable and writable data-structure can allow for a minimally redundant public discussion forum and act as a best-effort unified field of logic and reason, a sort of “best-effort truth database.” A condensed summary of how this is achieved follows. The key to advancing discussion is the reduction of redundant arguments. Much time is wasted repeatedly explaining the same arguments and counter-arguments both in person and in online discussion forums every time a particular issue arises in a conversation or online thread. This is an extremely inefficient way of propagating information, and it is vulnerable to a variety of attacks. This propagation can be improved by treating each specific “thought” as a node, and rather than repeating that thought in each case where it must be used to argue for or against a cause, simply link to that node, increasing the complexity of the best-effort truth database. Imagine harnessing the collective decision making capacity of all other users of the system in an efficient way, that's what this attempts to achieve.
There is already some open-source work done on argument mapping - but it serves a different use case. Maybe some code can be reused - I don't know. http://www.argunet.org/
Here is an example relevant to the BCH community. We will call this system "argumap.it" for now (I bought the domain but there is nothing there yet):
Someone makes the claim "Big blocks can't scale." If that node does not yet exist in the system, create a node declaring that statement. Then, create "refutation" links from nodes that show the statement to be false to the node claiming "big blocks can't scale". The opposition can create nodes that attempt to refute your negations or to support their statement, and the process continues. This "map" grows and becomes tangles with every other debatable topic. The original node "big blocks can't scale" could be used by the opposition to support or refute another node, and all the logical leg work about the statement can be viewed branching off that node.
I've thought about spam protection already. Essentially, users select trusted (don't crucify me yet) moderators who take argumap.it posts, quality control them (remove spam, fix grammar, prevent "text walls" from showing up as each statement should be separated out into a node, etc) and sign the QA'ed posts with their digital signature. Users then "filter" the map by posts with a digital signature from their selected moderators. Anyone can become a moderator, anyone can filter by any moderator's digital signatures, or they can filter by no signatures at all but that is vulnerable to spam attacks. I would call this a "federated trust model".
There's more to this for sure, and it's success is going to be based on how well we can create a front-end that shows users as much information as possible in an easy to understand way. There are other challenges too. There may be attack vectors not yet discovered. People might not even use the system. I imagine this could end up being a sort of social media for debating, if it is successful. I met a dev at defcon who was interested and we've been in light communication since then and he is still interested in working on the project (yeah, we move slow, very slow, but I really think this can help the community, and the world).
TL;DR: Use argument maps written to a blockchain or other public data structure to create a web of debates, spam protection via user selected moderator cryptographic signatures.
[link] [comments]
source https://www.reddit.com/r/btc/comments/a30nyn/i_have_an_idea_for_a_peertopeer/
No comments:
Post a Comment