6 ms·
Exactly. Where is the line between "curation" (or "moderation" in discussions) and "censorship"? If the line is subjective, how do we handle disagreements on th
by brightlancer 3y ago
Exactly. Where is the line between "curation" (or "moderation" in discussions) and "censorship"? If the line is subjective, how do we handle disagreements on that?
Even supposed objective measurements such as "Is this factually true?" can fail because we aren't omniscient, we learn new things and realize what was "true" yesterday is actually false, and sometimes an authority keeps telling us something is "true" and anyone who dissents is "moderated" and "curated" into effective silence.
I would trust Reporters Without Borders more than most, but maybe that's because their biases align with mine and I don't object to their censorship.
- pixl97 3y agoReality requires bias. Without bias and interpretation omniscients is required. A human mind does not have the luxury of knowing everything. We have limited compute power, and limited time. Furthermore our systems only have capabilities for limited data. We have to filter against junk or our systems get overwhelmed, but deciding what is junk is also a bias. The best we can do is state our biases, and design our systems to reflect or biases properly. If others don't like those biases they are free to create their own systems that reflect their biases.
- vorpalhex 3y agoI agree with your statement in the theoretical. However I disagree in practice because of this: > If others don't like those biases they are free to create their own systems that reflect their biases. Those who create alternatives are attacked, chased, banned, deplatformed. Even if a webhost is happy to work with you, the credit card network may not be or their bank may ban them. Not that the signal doesn't find a way - the good work of Anna's archive and all that - but it's not just "go build your own platform". It becomes "fight at every turn for your basic existence".
- pessimizer 3y agoThe market of ideas fails under capitalism. The market of ideas is denominated in votes, but capitalism requires it to be denominated in dollars. The market of ideas also fails under governments and corporations who pay thousands of people to work covertly and overtly to eradicate ideas.
- vorpalhex 3y agoThe problem isn't Capitalism. In Capitalism we can simply buy our services and anyone who can fund can play. What is happening when a bank refuses to do business with someone, when App stores ban a twitter clone and prevent sideloading, is not Capitalism. It's ideological. It's about service to an end outside just capital. Capitalism is the only system that freedom of speech can exist under. No Communist or Socialist author has ever entertained a freedom of speech concept - it is contrary to a centralized government. Such systems can not even handle mild political dissent and must (both in the theoretical and practical) literally kill or exile anyone pushing against the central authority. Capitalism can always tolerate disagreement because disagreement is profitable.
- hutzlibu 3y ago"No Communist or Socialist author has ever entertained a freedom of speech concept " Maybe you have heard of Animal Farm or 1984, written by George Orwell? Or do you think, he wrote those books to endorse the concept?
- pixl97 3y agoNo authoritarian has entertained a concept of freedom of speech. Just so you're aware, yes, capitalist can be authoritarian too.
- pixl97 3y ago>Capitalism can always tolerate disagreement because disagreement is profitable. Eh, that depends. The US in particular like to 'two party' problems. Capitalism loves two sides in problems, but that is very problematic if a problem is multi-polar and not bi-polar. You're also confusing socialist/communist with authoritarian. Capitalist systems are completely fine with being authoritarian if its profitable.
- pixl97 3y agoThe existence of society demands this in practice. This is how humanity has worked at least since the beginning of agriculture, and likely long before that. Step out of line too far, and someone caves your head in. At best we can hope for is that the rock holder is accepting, and not a fascist. You will find that it is impossible to create a society that does not exist in this manner. Since a society that accepts everybody also means it accepts people that don't accept everyone by means of violence, hence destroying a society that accepts everyone.
- vorpalhex 3y agoYou're conflating actual violence and "viewpoints I dislike". Yes every society will resort to vigilantism to protect itself in the absence of a policing power. But death threats to someone because they made a tweet you disagree with or abusing your power as a bank to control speech isn't about protecting yourself from violence - though that excuse is often used. It's a form of denial, of shutting down uncomfortable debate. It's an ideological sort of end, like a dictator who has a pathological need to execute his critics. The "inner ring" of acceptable opinions grows ever smaller - it has to. It's not about making the world better, it's about attacking outsiders and the beast always needs new outsiders.
- pixl97 3y ago>But death threats to someone because they made a tweet you disagree with or abusing your power as a bank to control speech isn't about protecting yourself from violence - It's about causing violence. It's no different than "You happened to say something against our religious book, don't do it again or you'll get hurt". Outsiders are a useful target because sometimes they are also the monster. This is the effect of existing in a reality with uncertainty. Accept everyone uncritically and you may be attacked by those you accept, this leads to fear, that fear is then manipulated by those that want power. There is no solution here, there is only the attempt to balance between malicious outsiders and malicious insiders.
- guy98238710 3y ago> Reality requires bias. There's only one reality. > We have limited compute power, and limited time. Simplified models of reality don't need to be biased. Unless you consider every viewpoint to be a kind of bias, which is really stretching the term.
- pixl97 3y ago>There's only one reality. You're almost, but not quite there.... That one reality is what I call thermodynamic truth. Now any time someone brings up thermodynamics other statements like arrow of time show up and other issues with informational incompleteness become problems. Simplified models of reality can quickly collapse in uncertainty in complex situations. Lets say an explosion and subsequent fire at a factory. The people working on the device that exploded where killed, so we only have second hand information on what they where doing. The fire was especially intense so the device expected of causing the explosion was melted completely and only mixed slag remains. The machine was made in the 1950s so other forms of entropy have been involved on information on the metals used in the machine. There is no simple model of reality that can tell you what occurred with certainty in situations like this. The additional entropy from the fire creates a situation where many possible input situations lead to the same output situation. We see this kind of entropy in social situations. The game of telephone is a good example of this. You start with "X5W1" and end up with "EXU1" after a few steps and everyone along the way would tell you thats exactly what they heard. >Unless you consider every viewpoint to be a kind of bias, which is really stretching the term Not stretching the term at all. Biases exist at all levels, physical processes and mental processes, human and inhuman.
- guy98238710 3y ago> Simplified models of reality can quickly collapse in uncertainty in complex situations. Lets say an explosion and subsequent fire at a factory. The people working on the device that exploded where killed, so we only have second hand information on what they where doing. The fire was especially intense so the device expected of causing the explosion was melted completely and only mixed slag remains. The machine was made in the 1950s so other forms of entropy have been involved on information on the metals used in the machine. If I cover an apple with a cup before you have time to look at it, the apple does not disappear nor is there any alternate reality with pear under the cup. It's just a blank space in your knowledge, which you are free to fill with any bias-free probabilistic model.
- tracker1 3y agoBias is only necessary for editorialization or conjecture regarding another's intent. Facts are facts, though omission of facts is also an issue..
- pixl97 3y ago>Facts are facts, though omission of facts is also an issue. Because you are a human, the 'facts' you acknowledge are anthropomorphic. We align our filtering of facts to the scales at which humans and society operates. When we change the scale of what we consider to be fact, then quite often what we consider to be a fact is really just an interpretation of events based on the perspective of the observer. Reality exists in a state of thermodynamic truth, that is if we could reverse the arrow of time you arrive at exactly one state at whichever slice of time. "Facts" as told by humans do not work this way the vast majority of the time. They are incomplete observations of a system using incomplete information. Time reversal of human facts can lead to situations where multiple starting states can lead to the same factual finding. Causality is uncertain without searching for even more facts, those facts which have been lost to thermodynamic scattering, leading us back to interpretation. Bias is absolutely necessary to operate on a human scale.
- ElectricalUnion 3y ago> Bias is only necessary for editorialization or conjecture regarding another's intent. I'm gonna assume you're taking a hard science approach to bias, then: It is statistically non-plausible to observe all facts in the universe, so the facts you can observe suffer from the selection bias fact sampling error. Bias is required in any useful model of the universe to explain the difference between the expected and the observed. Even assuming a non-plausible, accurate model with no biases whatsoever, such model is very likely to require high variance to encompass all fact observations and models with too much variance are useless.
- tracker1 3y agoSo the observed never happened? If the observed was recorded from many angles, did it happen then?
- itronitron 3y ago>> Where is the line between "curation" (or "moderation" in discussions) and "censorship"? They don't have authority to 'censor' anything, so what they choose to include is their choice as curators. If I have a bookshelf in my office that is full of banned books, then any banned book not on my bookshelf isn't banned by me.