jrh206

> as seen when researchers try to tempt dieters with chocolate and smokers with cigarettes

This sounds so unethical. If I were a dieter or ex-smoker I would definitely not sign up for this sort of study.

If you get bored, skip to slide 28.

I know it's hard to give new ideas a chance in today's internet, where our attention spans are constantly attacked. Despite the hostility of our systems, please try to persevere with this presentation. I believe there are some novel ideas in it which will fundamentally alter how we interact with each other.

Locally-hosted software is definitely important in the short term, but I think that distributed software will be enough in the long term. When everything is public, then there won't be the same risks associated with using networked software. In the long term, I expect the intermediaries to be public too.

But locally-hosted software should always still be possible. I included something similar as one of the core tenets in the repo where I outline an idea for how we get things going: "After cloning this repository, it should be possible to start a development instance of bartok with a single command." Based on your feedback I think we should tweak this to emphasise local use, because there's a distinction between using software locally for development and using it locally because you're using it locally. (https://gitlab.com/bartokio/bartok)

Which part of the argument do you disagree with?

Using Google Docs isn't ideal, and I didn't want to. However, I believe that the time has come that "ordinary people" (i.e. not HN readers) need to get interested in open source. I needed a solution that would be familiar and would definitely scale if the presentation got traffic.

The presentation is public domain.

Let's build a better open source alternative.

I need to be able to make changes quickly in order to get this thing off the ground.

I'm iterating on this document using the God Mode policy: https://gitlab.com/bartokio/bartok/-/blob/main/foundation/Go...

You're welcome to propose to change it.

If you're interested in the ideas behind God Mode, here's a strategy article about gauntlets: https://gitlab.com/bartokio/bartok/-/blob/main/strategy/Gaun...

Once the internet takes over, my powers will be gone. There's no way that I could maintain control versus the combined intellect of 7 billion people.

Is this the best you can do?

Speaking as the initial author: you are correct on all counts (though I think _lying_ is a bit harsh, we've been around the internet).

Despite its appearance, it wasn't actually generated by GPT. I put a lot of work into this. Based on your comment history, I think you might actually come to agree with me.

If you have the time, you are welcome to read it.

You're right. It sucks.

I think I might have stumbled on the impossible miracle passive income technique you're referring to, though. I'm trying to get some traction, but it's hard because of virality filters. Please could you take a look at this and see if it matches your experience?

https://gitlab.com/bartokio/bartok/-/blob/main/StartSomewher...

This is awesome. It terrifies me though.

"On average, probably once a week or so, somebody will go in the lake. Most of the time when they miss, they miss coming back onto the boat. So somebody's outgoing mail might get wet.

We've got a little pool skimmer down below. If we have to, we have somebody lean out the door and scoop stuff back up out of the lake."

  - Ray Ames, Mailboat captain

Which, by the way, seems to be heading towards infinity, which isn't right at all.

Indeed. Unfortunately the article lost me when I saw the graph of "extent to which you consider second order and subsequent consequences" vs "odds of success".

There's balance to be had. Keeping everything inside isn't healthy either.

I usually assume that I can't take sales people at face value. If it's a badge of honour, they're incentivised to say that they don't work hard even if they are in fact actually working hard (this includes pretending to enjoy golf).

Having said that, 30 people is a lot of people, so I'm inclined to accept your assessment at face value.

It's a pretty good first order approximation.

If you've written 400 daily poems and it doesn't seem to be having a positive effect, then it's certainly time to stop. If it's going well, then fair enough, you do you.

This is a small thing, but one of the first sentences in the documentation you linked is "At least I have never thought about API dependencies before building WunderGraph." It makes the problem seem small, as if you can get away with not thinking about it.

You might find Jonathan Blow’s talks interesting. This one springs to mind: https://youtu.be/C5FUtrmO7gI

Also this breakdown by Game Maker’s Toolkit: https://youtu.be/2zK8ItePe3Y

I completed it in 11 moves and assumed that that was the minimum.

11 moves felt like too many. I made my first moves without thinking ahead, and before I knew it, I’d accidentally solved the puzzle on my first try. I felt that the puzzle lacked that elegant “idea” at its core that makes a good puzzle great. (Just my personal opinion)

That said, the visual design is beautiful, and it’s inspiring that somebody was able to create this.

This document puts the sheer magnitude of DALL-E 2's knowledge of images into perspective. The same black box knows how to illustrate The Last Supper in the style of Quentin Blake, paint an ornate Late Baroque cat in sunglasses, compose a candid photo, draw a detailed blueprint... and so much more. DALL-E 2 knows more than a human ever could about imagery.

Whether or not this particular iteration of the model is 'good enough' to be widely applicable, or whether DALL-E 2 is 'creative', it's only a matter of time before the way humans interact with media is changed profoundly.

Morse code translator: https://morsecode.world/international/translator.html

(WHAT HAS GOD WROUGHT?)

Cognitive biases exist in a variety of contexts. The phenomenon of saccadic masking doesn't imply the existence of those other biases logically, but TheOtherHobbes makes a pertinent generalisation, imo.