Hacker News Reader

I over post this, and am turning into a broken record, but this is my favorite of all time article about predictive ai dystopia.

https://www.bbc.co.uk/blogs/adamcurtis/entries/78691781-c9b7...

my fear of everything being decided by algorithms that exist to please us, is us being trapped in the past and creativity being extinguished. if netflix makes shows based on what it knows will be popular, what happens to visionaries that have an idea nobody else can comprehend. risk, exploration, adventure, new, is replaced by nostalgia, familiarity, comfort.

we can debate fighting google and facebook, or we can come up with a "Human Best Practices" behavior guide for how to best consume content, and spread it. There will always be temptation, moderation is either taught or learned. If not facebook, there will be another mind drug to over indulge. Spend effort countering the concept of unlimited "picked-just-for-me feed" or "whats popular with everyone today" intake, not attacking this specific villain. We need new ways to surface and propagate novel, intelligent, thought provoking content, AND we need to teach people to resist human hedonistic urges and impulses, and offer social support and encouragement to deviate from consuming new and fresh news. Watching the news (on tv anyway) is, and shouldnt be, looked up to.

If I were making a tldr of guide v.01, it would start: "Supplant your facebook, google, and reddit intake with https://aldaily.com/ https://redef.com/ https://longform.org/ https://longreads.com/ "

and here i am now moving on to the next n.yc post.

It is not possible to over-post Adam Curtis. The range of ideas in his videos is fantastic.

I believe it's an economic problem. There is no reason all of these comfort services cant exist to help keep the peace among people uninterested in anything else. If they make up the majority of the economy, then we have other problems to solve though.

After all, we do need comfort for sleep - so it could be .. maybe the role of BIG companies should be comfort focused.

Visionaries can still exist - we would only need to improve their discoverability and ensure they are not at risk of financial crisis as a result of going outside the line

discoverability is the end issue.

facebook, reddit, google amplify what is popular, not necessarily ideas with depth. they in particular incentive things that can be consumed quickly and moved on from. a person consuming 100 links per hour upvotes at 10x the rate of someone consuming 10 links per hour. those shorter posts, bits, soundbytes, images, memes then gain 10x the amplification of longform content.

what is popular is not necessarily without depth either.

Also consider propagation of danger alerts. the longer time it takes to consume the average piece of media, the longer it takes for people to become aware of dangers. There are, of course, shortcuts for authorities to spread information about dangers already identified, but danger is a gradient.

Maybe there is an inherent trade-off between quick, scanning behavior being good for safety and deep, thoughtful consumption being good for progress

We most definitely did not get rid of authority, we got rid of central authority. Every nation used to closely control media so all of their citizens were on the same page. What they believed was biased propaganda with a strong illusion of truth due to dominance in commonly consumed media channels and the echo chamber of talking to others who drank from the same trough. It was far from being high quality information but at least your neighbors had the same beliefs as you so it was less contentious. There never was an authority of truth - only a facade of one. Critical thinking by individuals was discouraged to avoid conflict. Now that people can get their information from anywhere they are completely unprepared to use their minds to separate the good from the bad so they try to find "authoritative sources" instead of using their own judgement. Except their are no authoritative sources, and there never were.

Do we have actual reason to believe that really high adoption rates of people using their own judgement actually is a good thing?

I think there is inherent need for specialization to occur on digesting information and drawing conclusions from it. And when someone is good at it, we should want them to leverage other people parroting what they say to help signal boost

I just don't buy it. Are there any of your beliefs which you'd want to abandon just because someone got on TV and told you they're wrong?

the focus is more that I dont have the time or energy to form personal beliefs on everything talked about on TV (or by news authorities in general). It's easier to adopt someone else's beliefs and entertain that position than talk out of my ass or not be able to talk at all. If that belief is successfully challenged, I change it. It is also easier to change my belief on something if I did not form it myself.

Most people lack strong beliefs on most things. news authorities exist because people are looking for them to help fill that gap and relieve the responsibility of forming beliefs on all the things relevant to the community each person is a part of

That’s fair. I didn’t interpret the original comment that way, but I wouldn’t agree with the strong position that there literally shouldn’t be such a thing as an authoritative source of news.

Ah yeah, I did not want to rule out the possibility that people in general may be shit at forming their own opinions even when they do spend the time and energy. But that also opens the rabbit hole into why that would be the case (low vs high intelligence - but intelligence is hard to generalize), and variance of judgement based on context (e.g. quality soccer judgement vs shit hockey judgement for same person).

All of which kind of comes back to quantitative changes to either time or energy needed for each person to get to an acceptable judgement (even if it's infinite of both for some people-topic pairs), and both of those resources are incredibly finite compared to the volume of topics that are beneficial to the community to have beliefs on.

The big point I want to drive home here, which I have a feeling we agree on, is that we should prioritize the topics people are forming their beliefs on more than what those beliefs are. This is something authoritative figures actually can help with (for better or worse) due to the centralization. So imo, maybe you disagree here, it is not enough to simply allow authoritative sources to exist but we should be more actively promoting sources as authorities for topics.

But you would still require the skill to judge who is and is not a legitimate authority. This sounds more difficult and easily gamed than not bothering with a middle man and judging the facts yourself. Even if there is some perfect way of distinguishing true authorities from frauds, nobody is so authoritative as to never make a mistake, or so clueless as to not occasionally stumble into a correct answer.

You are only thinking about a single belief on a single topic, but we are tasked with a near endless amount of potentially important things to form beliefs on and an incredible sum of time and energy needed to actually do that yourself.

Authorities would be open to judgement by communities. You can better leverage your trust in people to cut down on the amount of energy and time you need to spend to get to a reasonable opinion. There becomes less instances for common miscommunication to go unnoticed. Yes, this comes with a risk of being gamed, which is another interesting conversation, but that's more a human vs human topic than a humans vs reality topic. which is important (economic incentives destroying the world and all) but a bit off-topic from the more general point I am making

basically, looking at all the details yourself is only good if you know what you are looking at and have the time to do so for everything you need beliefs on. The value that authorities provide is in consolidating multiple beliefs into a feed. Secondary to that is getting those beliefs to be high quality. Even if you go do the research yourself for a more quality belief to "fact-check" what you just heard - you have still accepted the value of the authority to tell you the TOPIC on which you should be forming a belief. All you did was locally iterate their work on a single topic to improve the deliverable.

Context matters as well. I dont need to judge the facts myself to figure out how many people died in the california fires if that detail is included in the report from authority. I'm okay with an authority figure telling me that number. Maybe I could arrive at a more accurate number myself, sure - but to what extent does the accuracy matter? It certainly matters if I believe no one died, or if I believe the death toll was orders of magnitude higher than it was.

The only real issue with trusting authority figures is the recursion. Authority figures using their own authority figures.. resulting in an appearance of multiple independent sources arriving at the same conclusion when in reality it is all signal boosting of a single source.

To protect yourself, the simple thing to do is assume any people who say the same thing heard it from the same place until you're given reason to believe otherwise

For a while, I thought decentralization could help with this problem. If people just published websites and there were no central social media, you can better control what content is seen and not seen. But then Google came around and aggregated these websites and made them searchable and now has it's own algorithm for maximum engagement.

Sasha's idea of delaying the spread of information via a timer is very interesting. It can get people a chance to second guess their posts and also give a chance for something in-between to publish or not. If every retweet was delayed then the network spread slows down a lot which might have really interesting consequences.

But ultimately I believe the problem we have is we have chosen the mall over the public park.

An important question is how would it even be possible to have a public park without government intervention?

You have to buy and maintain your own server and pay for uptime. That all costs a continued stream of knowledge and money and it’s very hard. Maybe not for you and I, but for many other people.

Interesting idea, a publicly funded social media site. That would allow the public to have input in how the software works.

I struggle as to how we could regulate massive companies like Facebook and Google from opening up their algorithms in a verifiable way.

Perhaps a better way to attack this is to allow them to continue, but break them up - ie, mandate competition, and admit that monopolies in markets like search and social media (not to mention specific segments of advertising) are as dangerous as monopolies in operating systems or telephone services.

I run an email newsletter[1] that aims to combat exactly this issue. It's a single link every day to something interesting, the kind of blog posts that you come across every once in a while that really make you think. The overwhelming feedback is that being able to read the kind of non-viscerally-targeted news and analyses that aren't easily found online is incredibly valuable.

[1] https://thinking-about-things.com

My solution would be similar, but the user can use a drop down/settings menu to change algorithms rather than install a different algorithm from github.

I frequently ask myself, if Hitler took power today - how many of our government officials, engineers, and soldiers, would get in line behind him. He would have no problem getting the census data and locations of subversives.