It has been about six years for me without any phone, even a "dumb" flip phone. I started a new business recently and it has been really difficult to engage in external outreach without a phone, but I am still managing it. Maybe one day, I have to get a flip phone, but I will try to stay away from smartphone as long as I can.
Does anyone select Codex from the dropdown on the top left corner of ChatGPT desktop app for coding tasks explicitly? Or just use ChatGPT Work for everything from coding to non-coding work? That is the part that really confuses me.
This reminded me one of PG's article about how to get started when you are not really feeling it on a certain day. You just tell yourself that you will go through what you have done yesterday for 10 minutes. That trick works wonderfully in most cases for me. This small trick, to some degree, helps with that activation energy to start working on what I want to work on.
> When I'm reluctant to start work in the morning, I often trick myself by saying "I'll just read over what I've got so far." Five minutes later I've found something that seems mistaken or incomplete, and I'm off.
I have both chatGPT pro and Claude max accounts and I am considering switching Claude for Grok. Despite all those claude hypes, for me personally, claude just does not match codex in almost all tasks, both coding and non-coding.
I have been living without a cell/smart phone for about six years now. I am a runner and I'd like to run at every city I visit. I have done many early morning runs in a strange city without a map or a phone. I lost sometimes but always found my way back.
One thing I notice is that I pay more attention to the road names whenever going somewhere without a map or GPS. I try to memorize road names instead of "turn left" or "turn right" notifications in digital maps these days.
I used to do a very coarse version of this stealing. I ask a question from ChatGPT pro, once it is done, I ask claude chrome add-in to go through all those thinking from the side bar, extract everything along with all the sources used. Then try to reverse engineer the solution it came up with.
For me the bigger value comes from explaining complex stuff. I remember Andrej Karpathy once said (He was quoting someone else actually) "You can outsource your thinking but you cannot outsource your understanding".
I can keep asking LLMs to explain a complex topic until I get it. Ask to explain it 10 different ways, explain it using physical analogies, explain it using visualization. If I don't get it, just say that out loud so that they can keep explaining it to me in different ways. We can keep going that until I really get it. That is the value I get the most using LLMs to learn things, especially complex topics.
> I can keep asking LLMs to explain a complex topic until I get it.
Or you'll be confused by the slop it generates, which is very conversational but may be just false. Or - you could "get" something wrong and believe it to be the truth because the AI said so.
I try not to make any judgement about people, their life and other things based on a few days or at most a few weeks of travel in that place. It is tempting to say things based on a very limited experience on a very limited people in a very limited place.
I can sympathize with the viewpoint in the article. Indeed, I agree with many of the points of the article.
But also, when I didn't speak Spanish very well, Spain felt magical and whimsical and all the people seemed so happy. A few years later I came back and spoke Spanish at a much higher level (pretty easy to improve when your initial level of capability was counting to 100 and ordering beer) and noticed that the situation was much more nuanced. I'd over hear the same sorts of conversations I'd hear at home.
reply