Toggle menu
122
332
11
3.4K
Information Rating System Wiki
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Ratings system

From Information Rating System Wiki
Revision as of 20:05, 25 September 2024 by Dan (talk | contribs)

A ratings system, broadly, is a method by which people can share ratings on things. This includes things like online review platforms, website reputation databases, and the concept of "upvotes," "downvotes," and "likes" on social media platforms.

This page is mostly about the ratings system PeerVerity is building. It is a software solution to the problem of widespread misinformation based on the ability to rate that information and the people who provide it.

What it is

Put simply, the ratings system is a system of software which is intended to help people make better (or at least more informed) decisions. It enables us to ask questions which are answered by the people we trust, the people that they trust, and so on. It also enables us to provide those answers to other people who are asking for them. We can look at several examples of current ratings systems and compare these to our ideas.

For example, you may want to know if Acme Corporation is a reliable supplier of anvils. You don't know, and none of the people you trust directly know either. However, you trust your uncle Larry, who trusts his best friend Bob, whose brother-in-law's nephew is a hobbyist blacksmith and knows all manner of things about anvils, and will give it to you straight: Acme Corporation's anvils are literally made of rubber, and you should shop elsewhere.

However, due to the limitations of humans to connect with one another, it's highly unlikely that this sort of information would naturally propagate to you. Fortunately, just as we have used tools like computers and the Internet to augment our natural capabilities for mathematics, information retention and processing, communication, and so on, we believe that it is possible to use technology to augment our ability to leverage trust, build communities and fight disinformation.

The ratings system has many applications and is seen as a way to transform society for the better. First, it will educate and be a source for self-improvement. It will provide ways to overcome bias vis a vis our current media environment. Expanding on that, it can promote civility and battle entrenched bias held by individuals.

It is natural to think of the ratings system as a type of voting system which would be needed by communities for governance. In this sense the ratings system encompasses a larger sphere of influence, one which takes aim at current institutional failures. Once we begin thinking about governance, we can begin to think about political and economic systems along with the frankly utopian values we would bring to these. These ideas lead us to system modeling in general and how we might try to optimize socio-economic systems within the context of a ratings system. This is an ambitious project, particularly in its larger social implications, but we might remind ourselves of the impact technology has had in the past.

The ratings system will exist at both a personal (subjective) and community level. We distinguish these as the subjective ratings system (SRS) and community ratings system (CRS). The SRS is a tool for individuals to make use of their own network for reliable information. The CRS envisions a voluntary community which establishes a common network for both information and voting on community matters. The CRS has the concept of weight assignment, where members can delegate some of their trust on an issue to others who know the subject better. Both the SRS and CRS will coexist.

In a ratings system, who does the rating? Primarily individuals. But as communities grow up around the system, it is likely that organizations will be tasked with performing ratings on information that requires specialized knowledge. The ratings system will impose a burden on folks to participate and there may be understandable limits to this, both in terms of expertise and time. It will be necessary to calibrate this organizational role carefully so it does not usurp individual rights and responsibilities.

As an information system dedicated to promoting objectivity and honesty, the ratings system will have built into it a number of truth-finding methods, including debate, logic, and mechanisms to promote civility. Collaboration will also be important since we suspect that active partnership with ideological opponents often leads to a messier but more accurate truth than an insistence on agreement. Furthermore, in order to maximize community acceptance, privacy, identity, and fraud, will be taken seriously by the ratings system and designed into it from the beginning.

When we envision a world where an information rating system heavily influences social status and a community’s decision making process, the rating system's security is important, because people will be highly motivated to cheat the system.


How it works

Main article: Technical overview of the ratings system

The ratings system depends heavily on the core concept of a predicate: a statement which can be evaluated for truth. A good way to think about this is that any question which could be answered with "yes" or "no" can be restated as a predicate:

Question Predicate
Is water wet? Water is wet.
Is the cake a lie? The cake is a lie.
Is jazz music enjoyable? Jazz music is enjoyable.


Each person who wants to be a member of the ratings system can install the software and become a peer on the network. They must then seek out other users to connect with directly. Once they have added these peers to their network, they may start to make use of those peers' opinions, and make their own opinions available for others to use.

This sounds simple enough at first. The innovation is what happens when none of your peers have an opinion on a predicate. They will all ask their peers, who will ask their peers, and so on. What you will receive is the computed opinion of your personal network on that predicate.

The ratings system will work to the extent that it properly understands human psychology and social dynamics. After all, individuals themselves will be rated by the rating system in various ways, and experience firsthand what it's like to be judged in this way. Individuals will be able to see where they stand with their peers, especially under the community version of the system where such ratings are public. This is somewhat of a change from our current society where our "rating" with others is often difficult to ascertain. We know other people rate us but we're not sure exactly what they're thinking. It's hard to get honest opinions from people even when they are sincerely requested.

Which properties will we be rated on? Which properties are important for society and simple to grasp? Is individual belief or group behavior more important? What motivates people to cooperate in a system like this? What if someone's rating is unfair or inaccurate? In terms of limits to human potential, can the ratings system break through any of these? What might we do to draw in the ignorant and unengaged person?