Overview: We want to evaluate and improve how people improve their ideas, currently the system will be built around the philosophy of people evolving their views based upon making observations, interpreting those observations, determine the various places where the interpretation(s) would best fit into a worldview (viewpoint) based upon some metric (eg. how hard it would be to change the worldview + change the viewpoint * sigmoid of the total proportion of the possible evidence has been observed, the qualifications of the person exposing the views, or some other metric), and create hypothetical worldviews based upon one or more core worldviews. As the system grows, the level of automation will also grow from supporthing things like entity detection (Weinstein -> Eric? Other?), contextualized suggestions, abstract paraphrasing/ambiguity detection, etc.) Eg. Interpretations: The corrupt leverage power from exploiting the naive. <- What proportion of the naive are being exploited?; (What constitutes corrupt? <- Not following some philosophies that we've defined and agreed upon.); <- Xyz observed that Person A got conned by Person B; etc.)
The viewpoint evolution system would then feed into a system of determining effects so we could have effective policy management, better testing, etc. Eg, Targetted smoking ads in third world countries is being supported by proxies (mange hedgefunds, and is predicted to cause major health problems for millions of people with no real benefit. As we view this as crimincal negligence supported by regular people through obfuscation, then funds flowing to corportations engaging in this practice should be partially redirected to efforts to report such human rights violation to the regular people that are unbenownst to them, funding it.
More details You register and create persona's (Safety for whisleblowers, journalist's sources, etc.) Henceforth the use of the word "You" will refer to one of your created personas You make observations You extract structure from the observations and use the structure to reference what you actually mean (Links to wikidata/freebase/dbpedia/yago/cyc/etc.) You determine various metadata factors upon the observation which are customizable What is your emotional and/or financial "distance" to this observation. Are there other influences that might distort your observation? Level of abstraction of observation How many levels of interpretation did you have to go through to get this observation? How close are you to fully understanding the entire situtation? (Eg, an accident happens in a warehouse, someone new in the warehouse could see the acident happen, while someone brand new might see some action and notice someone was hurt.) How long ago was the observation? Was this an automated observation made by a machine? You make interpretations It references observations and/or interpretations You determine how those interpretations
Moderating Mechanism, People must be unhindered in networking and communicating.
Looking towards technology to incentives