Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

A subset of the authors point is the "truth collapse".

It's increasingly difficult to differentiate truth from falsehood. Also increasingly difficult to differentiate between truth and 'almost' truth.

A further problem, likely in the future will be 'source collapse'.

It will be increasingly difficult to determine the actual source of news, with unlimited sharing and distribution. Add in to this mix of chaos AI generated news articles, it may well be possible that people will shun internet news altogether.



Today, a common "good" chain of custody of facts is a broken game of telephone that's hard to unwind anyways. How often on social media do you see something like someone on reddit/fb/wherever posting an article reporting on a mainstream publications's article reporting on a scientific study of some phenomenon. And then most people will still only read the headline, which is probably going to turn "Effects of blahblazine on rat tissue in petri dishes" into "Scientists say coffee makes you live longer."

If that's the best the ecosystem can do to report on good science, what hope is there? Reality is almost always nuanced, but nuance doesn't go viral, and everything has to go viral to bubble up to visibility.


>It's increasingly difficult to differentiate truth from falsehood. Also increasingly difficult to differentiate between truth and 'almost' truth.

Our decreasing ability to recognize truth is merely a consequence - and we've never been great at it anyway. It is the concept of consideration that is crumbling.

A lot of people either don't take the time (wilfully or not) or simply don't have the ability to reason out their interaction with a sample of information before they accept and internalize that event as is. Especially the appeal of social exchange pulls in even the sturdiest critical thinkers to make an immediate judgment and an emotional investment. That is then the trap which forms our digital and eventually real life bubbles.

The Information Age demands a strong familiarity with our own flaws, our behavioural, cognitive psychology, and sociological tendencies in particular. Since that is by far not the case, we've become a lot more vulnerable to self- and external manipulation. That is why content collapse has had such a negative impact on society.


> A subset of the authors point is the "truth collapse". It's increasingly difficult to differentiate truth from falsehood. Also increasingly difficult to differentiate between truth and 'almost' truth.

It's kinda hard for there to be, in actuality and in an objective sense, a destruction of truth when, really, truth never existed in the first place except as a perspective-based pragmatic tool and even a self-actualized Other-being. Really, when one person claims that truth is becoming destroyed or something, such occurence is actually an instance of a contest of truths, all equally valid, in essence, with the one that has the most ambition able to enjoy the fruit of victory. That's why "fake news" is often touted: a transgressive action on its antithesis and opponent that is the "false."


If truth didn't exist, we wouldn't be able to communicate with each other. Language would be meaningless like a randomly generated sequence of words conforming to a grammar.

There would be no measure of consistency between statements, and there would be no predictable link to empirical observations. Without truth, our species would never have developed language in the first place, because what's the point in evolutionary terms?

Speaking always means to make a commitment at least to some degree. That doesn't mean it's always possible to establish the truth value of everything that's being said or that someone must always be right and someone else must always be wrong.


How do you know that there is a communication going on and in what absolute sense does it occur, without resorting to any certain sense as certainty always comes up short, never converging to actual truth and is, therefore, not suitable for matters of truth. Any meaning given to any instance of language depends on subjective based grammars that give life to words, indeed conforming to a regime of, ultimately synthesized, grammars which don't exist without a subject of self-governing origination. Every measure of consistency requires a necessary foundation, with the foundation always being some imagined standard based on certainty, that commonly found tool of the mind that never really tells you anything. The emprical, material, and even biology and its evolutionary theories? Knowledge of their objects always depends on interpretation and interpretation is always not objective but instead subjective. And for there to be a commitment, there must be a personality. But personality is, logically, dependent on a regime of truth and it is codified by the user of this regime—a mess of endless self-reference, as truth is wont to do. Really, at this rate, I'll eventually converge to the truth that there is no truth, that real truth is not, instead of being is. Indeed, I'm going one truth further.


Perhaps it comes down to how we define truth. Truth, for me, is not some sort of ontological object as the word "exists" may imply, and it requires nothing absolute.

Truth is an artifact of communication as experienced by those who are communicating. It's the semantic non-randomness in language. We share expectations regarding the consistency between statements and the degree to which language predicts empirical observations.

If none of those expectations are met, i.e when all truth is gone, then language loses its function.


Indeed, since truth does not exist, language and communication is impossible. But the truth predicate for propositions isn't necessary, as truth doesn't exist to legitimately create a legitimate need for any sort of necessity. So, instead of searching for the truth predicate, and as grammar and syntax, those subjective and arbitrary creators of languages and communications, are the ultimate sources of meaning rather than any inherent meaning itself, why not look for something else? Or establish something else? Like untruth and uncertainty?


I think it's a mistake to assume that two people are communicating with each other just because they're saying the same words. Some of the worst things that have ever happened to me have been the result of people thinking they are in agreement despite having totally different ideas behind those words in their heads, and in retrospect I would have rather had arguments.


I'm sorry to hear about these bad outcomes from misunderstandings. One useful tool I picked up from Nonviolent Communication (NVC) that might help you is to request the other person repeat back in their own words what they heard you say/request. It definitely feels clunky because we don't typically ask this of people. Proactively as the listener you can offer to summarize the thoughts or request of the speaker to make sure you really received what they were saying, and ask if there's any part of it they want to clarify. "Let me make sure I understood you. What I heard you say was..." Maybe a useful analogy is to think of it as the md5sum of the exchange. I don't tend to find arguments clarifying because true listening breaks down (further).


Contest of truths sounds an awful lot like alternative facts.

I disagree with you.

There may be many perspectives to a truth, but there is always a singular absolute truth. Something that can be objectively stated, without judgement.


> There may be many perspectives to a truth, but there is always a singular absolute truth. Something that can be objectively stated, without judgement.

I can't imagine how one could find these truths without using judgment. That's like trying to figure out how long something is without a ruler.

edit: And there's no way of being sure which of the things you found were true, or which which were just artifacts of the measuring device one used to find them. One would have to pay a visit to the ruler factory and start the process all over again.

The things one can say about reality don't actually describe reality, they are a subjective human narrative created as a tool to meet human needs that any singular absolute truth ignores without comment.


Interesting. But I'm skeptical about the existence of an absolute truth. Because, ultimately, what would be the essence of what it is, truth? And why would it exist? What source would be the source of this truth, ultimately? And how would we know, especially as truth implies untruth and therefore the possibly of cognitive error?


Is there a thermodynamic truth, and where does it break down at the quantum level?


Thermodynamics and the quantum most certainly would not exist if you did not bring them up. You are now the master of existence. I dub thee Molecule Man.


> there is always a singular absolute truth. Something that can be objectively stated...

In incredibly simple scenarios perhaps, but for most things it seems to me that there are simply too many interrelated variables and perspectives involved for a "single truth" (that is perceivable by a human mind) to exist, let alone be communicated with high fidelity. And this is even if one assumes that we are aware of and able to accurately measure all the relevant variables, which is not even close to true.


A perspective is something that is biilt upon a truth, not something that is part of a truth.

What you are talking about variables seems to be about partially revealed truth.

Even in that case, a partially revealed truth can be stated as such,showing what variables are fully known and what are unknown.

Curious about your idea of supposed complex examples where its not objectively possible to state the truth.


> A perspective is something that is biilt upon a truth, not something that is part of a truth.

Says who? https://dictionary.cambridge.org/dictionary/english/perspect... doesn't seem to support that assertion.

> What you are talking about variables seems to be about partially revealed truth. Even in that case, a partially revealed truth can be stated as such, showing what variables are fully known and what are unknown.

For simple scenarios, perhaps.

> Curious about your idea of supposed complex examples where its not objectively possible to state the truth.

The United States is systemically racist.

We should (or should not) extend the coronavirus lockdown in the US.

Donald Trump is a white supremacist.

Joe Biden is attracted to young girls.

Decoupling from the Chinese economy is good for India in the long run.

Global warming is real and primarily man-made.

We should get our news from respectable media outlets.

Democratic capitalism is the best system for managing a society.

We should trust scientists and ignore theologians.

So many examples, so little time.


>Something that can be objectively stated, without judgement.

You can't always find the objective truth. Was Trump _really_ joking when he suggested people drink bleach? You can conclude with reasonable certainty that he was or wasn't based on his previous patterns of action or whatever, but the only way to truly know what he meant is to be him. And your interpretation of his previous actions will necessarily be incomplete: there's always going to be something that you don't know about that could potentially change your interpretation, and each event in his history that you base your interpretation on will have to be analyzed with the same amount of rigor. It's possibly for a person to perform this analysis, that's a biographer's entire job, but to do it for every person in current events, for every situation, for every conflict and incident is impossible.

So yes, there is an objective truth, but you can't always be certain that you know it. If you're not dedicating your entire career to analyzing one thing then you'll have to rely on heuristics, and those heuristics will be built on top of your knowledge and analysis of past things which are also based on heuristics, which are based on more heuristics, etc. Those heuristics, the way you analyze and interpret events, are only as important as the information available to you for you to build upon. If your primary source of information is Fox News then your heuristics will be that of course the President was joking, but if your heuristics are based on information from HuffPo then you'll come to the opposite conclusion. If you want to reject your heuristics and find the truth based on first principles, or at least come as close to the truth as possible, you'll need to do quite a bit of research into the issue, and by the time you've finished there'll be another five things the President has said for you to find the truth of.

It's simply impossible to determine the truth of current events without using heuristics and avoid falling behind, but those heuristics are based on known truth only indirectly. Does there exist an objective truth? Yes, but it can't be found in a reasonable amount of time. It's much more important to focus on the heuristics people use to determine truth for themselves.

Postscript: algorithmic newsfeeds that optimize for engagement are optimizing for inevitable unconcious biases, confirmation bias in particular, and are great for biasing your heuristics unfairly. I'm not saying the truth is always somewhere in the middle, but I am saying that if your heuristics are always skewed to one side you won't be able to recognize when your heuristics are leading you astray and when the truth is in fact in the middle or even the other side. If you get anything out of this, at least try to figure out what echo chambers you're in and look outside of them.


Neal Stephenson's latest book ("Fall, or Dodge in Hell") has a subplot around this that I really wish he'd developed further - I think it's the best part of the book.


Is the book any good? I loved Snow Crash and especially Cryptonomicon.


It was alright. I’m a big fan of Stephenson, and I can see what he was trying to do with it, but the main plot line of the book just wasn’t that interesting to me. Parts of the side plot felt almost prophetic, though - he draws out some contemporary trends in ways that feel frighteningly plausible.


You could simply use cryptography to sign all messages sent by sources. Problem solved.


That will not solve the problem to the average user.

Also, even if all sources are verified, it is already very trivial to create a new publication or a "news" source and make it look legitimate in a few months.

It's a very messed up situation.


Agreed on the modal user, but two quibbles: (a) how often does anyone follow secondary sources that are only a few months old? (b) it is less trivial to create new primary sources. (not that publishers are in the habit of giving easily followed references)


"You can't sign a quote", said a spokesperson for the Misuse Of Crypto Bureau today.


Cryptography can be used to verify random person as authentic random person, yeah


Just remember to say nothing that the .Gov doesn't approve of.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: