I think it's not as much about how right or wrong or interesting or not the output was, for me anyway, the concern is that I got a bit... lost in myself, I have real things to do that are important to people around me, they do not involve spending hours with an LLM trying to understand the universe. I'm not a physicist, I have a family to provide for, and I suppose someone less lucky than myself could go down a terrible path.

▲

johnisgood 8 days ago | parent [-]

Okay, but like I said before in another comment, I have spent 3 days straight coding, neglecting myself and everything around me in the process. I was learning a lot, coding a lot. I was productive. Of course I should have had some breaks (for my legs and mind, and my body). Just make sure to have breaks. I did not have breaks because I was completely zoned in. I set up a timer by then that remind me to take a break.

I checked the content, I do not think that it is useless, and I am sure you have learnt a lot. Perhaps get in a rabbit hole about http://CharlieLabs.ai (your project, before people think I am advertising). :P

▲

roywiggins 8 days ago | parent [-]

Lengthy ChatGPT rabbit holes are kind of a simulacrum of productivity, they keep you in a flow state but it's liable to be pure cotton candy, not actual productivity.

Spending all weekend on a puzzle or a project at least keeps you in a tight feedback loop with something outside your own skull. ChatGPT offers you a perfect mirror of the inside of your own skull while pretending to be a separate entity. I think this is one reason why it can be both compelling and risky to engage deeply with them: it feels like more than it is. It eliminates a lot of the friction that might take you out of a flow state, but without that friction you can just spin out.

▲

johnisgood 8 days ago | parent [-]

It depends. Do not pursue pure cotton candy. :P

▲

roywiggins 8 days ago | parent [-]

Put it this way: at least with vibe coding you'll eventually hit something where you realize that it's produced crappy, useless code that you need to throw out.

With extended philosophical conversations there is nothing grounding the conversation, nothing to force you to come up short and realize when you've spent hours pursuing something mistaken. It's intellectual empty calories.

	▲	bonoboTP 8 days ago \| parent [-]
		Depends on how you use it. You can "ground" it by asking what authors have explored this or ask for book recommendations, then read the wiki page of the author, read some texts by them etc. You can explore the history as well, like what was happening at that time, who were important contemporaries or influences, people who thought the opposite etc. I've found interesting books (that are somewhat niche but fairly well known in the field, non-fringe) this way.