pillefitz

Are his arguments sound, though?

In terms of bits seen during training, LLM are more akin to a 3 year old. Robots roaming around and learning to interact with the in environment and sharing knowledge might be a game changer, assuming that the current methodologies are sufficient (LLM + RL)

Reading "UFOs and Government: A Historical Inquiry", albeit a bit dry, changed my mind on the topic - I now believe with reasonable confidence alien visitations took place.

Not necessarily creative, but I tend to read a lot on my occasional train rides

Assuming you're a technologist of sorts, did this knowledge impact your life choices? I'm currently diving deeper into philosophy with the aim of gaining a wider perspective on things and maybe, one day, find a more meaningful job than my current one.

It'd argue it's both cause and solution, as it, in the hands of the rich, drives the disparity between poor and rich further and further.

Awesome, came here to mention this important work of Ellul!

Following a parallel theme of labour vs. work vs. action and the dissolution of the public sphere (as enabled by technology), I can also recommend Arendt's The Human Condition for further reading.

Beautiful comment, thanks

Depending on where you live, teaching in part-time?

I live in Germany and spend around 20k a year, despite living in a rather large flat in one of the most expensive German cities. No car ownership though.

Reminds me of the lives of the great philosophers, who, in order to freely contemplate, owned slaves to do the necessary work

Well, Israel does manipulate the media, no need to come up with strawmen

LLMs alone might be powerful enough already, they just need to be hooked up to classic AI systems to enable symbolic reasoning, episodic memory etc.

The problem of overflowing context is solved by RAGs, though.

I built https://readboost.io earlier this year. It adds Q&A to quiz yourself to the end of each chapter and lets you download the annotated ePub. Unfortunately, motivation left me after 3 months and it can't handle all books yet. Anyway, it's free to try if anyone is interested.

Thanks for your insights. So far I've always worked on solo projects, and there's a faint glimmer of hope that having a (accountability-) partner would allow me to finally ship.

For me, a break due to a stressful period at work or long vacation is typically crushing my momentum, and I'm rarely able to continue working on the side project after. How do you deal with it?

How is launching impacting your commitment to your project? Once I built an MVP, I'm typically satisfied and move on (even though I convinced myself for weeks that my ambition would reach well beyond building an MVP)

What made it useful to you - the fact you could use the MVP yourself (internal motivation) or that you had an early user base breathing down your neck (external motivation)?

Did this approach work consistently, or were possibly other factors at play?

To summarize, sequential building didn't work, and parallel building you're currently in the evaluation phase of?

What compells neurotypicals to start projects, then?

Interesting idea, but when looking for candidates, I don't have time for that.

They state themselves it's just system prompts.

Same here. There's no insights to be gained writing code that I couldn't have had iterating on paper (and possibly showing to users).

It's for sure not the "reverse", even if for some people with certain diseases the benefits may outweigh the risks. It's undeniably linked to certain cancers and arteriosclerosis

Aren't you worried about the health impacts of eating steak that frequently?

First I was annoyed by your response, then intrigued: It reads LLM-generated, but is it? If yes, how did it make the link to your app?

By latex he means LaTeX.

LLMs have more knowledge than you'll ever have, while being somewhat worse at reasoning. A friend of mine suffers from severe depression and after trying 5 medications, decades of therapy with different therapists, LLMs are the only thing that give him the feeling of being listened to. Funnily enough, he had therapists berating him in a quiet similar tone as evident in your post, while only getting empathetic and level-headed responses from chatGPT, which have helped him tremendously since.

Anything original, for that matter.