Hacker News Reader

Comment by briantakita | Hacker News Reader

This sounds like the reaction against "Fake News", "Fake History", etc.

What that really tells me is this is an area of study that the Establishment does not want you to follow so it will label it to apply social pressure to mitigate the effects & discourage people from following it.

The thing is, nature has a way of not caring what the Establishment thinks. It does care on a surface level, but it also conspires to cause the collapse of the socio-information paradigms that the Establishment creates.

Given we have a soup of information & no grounding central authority to give us "objective reality", we ought to utilize other techniques. I don't happen to know what these techniques are, but I suspect that it has to do with network models, cognition models, perspectives (physical & information), attention schemas, faith, complexity, patterns, language, etc.

View on HN

> The thing is, nature has a way of not caring what the Establishment thinks. It does care on a surface level, but it also conspires to cause the collapse of the socio-information paradigms that the Establishment creates.

You're right as far as you go, but actually nature doesn't care what anyone thinks. One thing to question is how the establishment got to be established. If it was primarily through physical strength, heredity, or persuasion (as is more true for the bulk of history), then they are no more likely to be right than a plebe (except that p(persuasive | true) > p(persuasive)). But it's more true than in most of history that the odds of being an elite are much higher if you can consistently make correct predictions about the world.

It's true that we don't have a central authority to give us an objective reality, but we don't need one either. Reality is there if only you choose to observe it closely. I think the best technique for dealing with the soup of information is to find bona fide experts with a track record of providing correct explanations and giving what they say more weight.

> I think the best technique for dealing with the soup of information is to find bona fide experts with a track record of providing correct explanations and giving what they say more weight.

It's up to the experts to make a persuading case & to engage the audience to think for themselves. There are many-a-con-artist who labels oneself as an "expert". That con-artist may even have credentials to make one's "authority" even more convincing.

Labels justify all sorts of things, worst of all, telling one to stop thinking beyond the abstract representation of the label. In the domain that the label casts a perspective shadow upon, the con-artist is free to utilize to take advantage of the audience's ignorance.

Knowledge is seen as dangerous by people holding power over others. It always has been (e.g. the apple thing). Therefore, in an age where knowledge is more readily accessible to a much larger number of people, it behoves the groups in power to somehow thwart its development. A good way of accomplishing this is to use mass media to make the signal to noise ratio as small as possible. Which is itself accomplished nicely by paid, conscious, agents and amplified by the growing hordes of individuals humiliated, rather than motivated, by their lack of knowledge. The true becomes a moment of the false, the real a moment of the fake. I believe that "spaces", real or virtual, rather than techniques will develop for knowledge to flow and grow.

Is there any scientific theory outlandish enough that you are willing to call it nonsense on its face?

A better indicator of quality of a model is to ask, "is this model useful?". At the end of the day, models are representations of another system. Like tools, some are more useful than others.

Ok, so I propose (for the sake of argument) the following model: any message posted on an electronic message board that casts aspersions on "The Establishment" was actually written by an AI running on a computer in Siberia.

Let's suppose that many people find this model useful, because it allows them to dismiss such posts without having to critically engage them. They downvote/flag all such messages, and they feel good for having done so because they believe they are resisting an attempt to manipulate their opinions.

Do you have any problem with this?

Useful in that context is typically used to refer to being able to make accurate predictions about the future.

Does it need to make accurate predictions or just inform good decisions?

I think a model can be bad at the former but good at the latter.

It depends ;-)

Useful to what systems & what aims? What's important to an Entity to engage what systems?

AI needs tuning. Doesn't seem to be reading the posts it's relying to. ("useful, because it allows them to dismiss such posts")

It's useful to think that God will reward you for hard work (or this model gives statistically useful, actionable, and accurate results and makes useful, actionable, statistically accurate predictions) but it is not an accurate model for the mechanism by which external changes in your life occur. I vote against "is this model useful" as an indicator of the quality of a model.

Reply on news.ycombinator.com