[tsunamifury]

I applaud the author and thinker for taking on this timely and hard topic.

I struggle with this question in the same way I think as the author, but in technology we are afforded less time to ponder if we are wrong and more time to test if we are wrong.

However the author points out, even in testing as he does with his students, we can be wrong in a fundamental way that all the branches of my iterations stem from the wrong source.

So I’m left with: who thinks I’m wrong and why does that matter.

I’m finding that outside of reddit, very very few people will tell me im wrong and this is deeply frustrating. Really only my wife who is tired of my pondering fully engages in what might be wrong with what I’m working on and I’m thankful for that.

But I wish more people would help me be “constructively wrong” which means they understand the goal but want to correct the approach.

Most online merely want to point out irrelevant wrongness for sport.

[randomdata]

> I wish more people would help me be “constructively wrong”

What's the value proposition? As you noted, not even your own wife will help you until she sees some kind of return for herself (abating her tiredness). Online actors pointing out irrelevant wrongness get to laugh at the meltdown of the maladjusted "intellectual" that usually follows.

This is what consultancy is for. You pay someone to look at what you are doing and tell you where you are going wrong. The pay offers the incentive. Most people are quite happy to offer consultant services for pay. But presumably you are having this wish because you want it for free?

[tsunamifury]

I get your point, but to be genuine with you I have paid.

Very very few people you are paying with genuinely critique or disagree with you for very obvious reasons.

However I can’t help but point out that you’re being exactly as I’ve described redditors…argue an adjacent point to just say the original point is ‘stupid’. It’s a waste of good brain cells.

[randomdata]

As before, I am humoured by the meltdown, which has not failed to disappoint. There is no other value proposition online.

[tsunamifury]

I’m merely saying I also have done that.

I think you’re a troll and looking at your comments you seem to engage totally disingenuously with HN.

[randomdata]

> I’m merely saying I also have done that.

It seems you haven't. Apparently you've exchanged money for something, but apparently not what you were actually looking for. It seems, based on the description provided, that you just wanted to give away free money to whomever was willing to take it. What value proposition have you offered to get someone to actually point out where you are wrong?

> I think you’re a troll

I cannot be a troll. If I were, you would not feed me. It is likely I am disingenuous – but that's the whole charm here. One comes here because they want to interact with screen names, not people. A screen name implies nothing of what lies behind it. It could be an LLM for all anyone cares. Decidedly, it is not intended to be a real person. If a real person is pulling the knobs and levers behind the scenes, that's merely an implementation detail of the software.

> and looking at your comments

Well now I wonder what I wrote that made you think my other comments would be something worth reading?

[passion__desire]

Which is why we need chatgpt as a thought buddy! Chatgpt can route insights it gained from one chat and insert into another chat with a different person. That is learning afterall. Most of insights are memoized in some sense, we don't derive them from first principles again and again.

[tsunamifury]

I built allofus.ai for that! Try it! Long term memory and 8 different perspectives based on real people.