I’m able to avoid this basin with a pretty natural baseline professional positivity and frustration management that I would employ with pair-programming. For example, if I just made progress with a human I was guiding through a task, I would be like “Nice, now let’s xyz” (instead of just “now let’s xyz” as if _I_ were the robot lol) or if we had to work for a result I’ll be like “Sweet! Looks good, now let’s xyz” - this is important signal for humans, and the same is true for agents. Also staying emotionally regulated and focused on the goal when things don’t work as expected or when we haven’t made progress after a few tries at something, critical in human interactions :) and even if it’s my job paying for the tokens, the idea of racking up even a microscopic bill for the privilege of having a machine read my insults and then formulate some credible-sounding blob of apology text is belly-laugh absurd to me. I do try to express my genuine feelings during more vision-oriented planning sessions, and just like with a human, you have to maintain the vibes if you want a genuinely collaborative session to go well. If you are toxic people will become either defensive or aggressive in response. From reading the rest of the front page it seems like we are lucky that Claude is the former, and that we especially best maintain a positive atmosphere around Grok.
I didn’t read the author as blaming, and I can relate to what they’re saying. It’s hard to speak about a childhood full of abuses that thought they had your best interests at heart. I was locked in my room for days on end (sometimes back-to-back stints for weeks), deprived of privileges/books/music, berated, demeaned, and ultimately gaslit, throughout my childhood, because my parents didn’t really understand ADHD (or that I had it). It created a pretty big mismatch between expectations and reality. As a kid, I resented them a lot for this. As an adult (and parent), I realize what it must have looked like to them and I can empathize without condoning their behavior.
My performance is heavily bimodal: my adult ADHD assessment included scores in the 98th percentile and the 8th depending on the task. I also have an extremely wandering mind. Taken together, it seemed to my parents that I was a brilliant child who simply “refused to work hard”. This was not only toxic for my relationship with them, it poisoned my self-image and ability to relate to others for the first two decades of my life.
Any resentment I might have over that is ultimately misplaced. They were loving and caring parents who missed the mark in a way that just so happened to fundamentally shape my childhood experience. ADHD wasn’t even understood the way it is now, so the most I could have hoped for in childhood was a prescription.
Instead, I now make a point to be a parent who understands my own ADHD and my kid’s (which thankfully is pretty identical to mine). I help them use alarm clocks and timers like I do. I give them tools to understand and empathize with other people. I don’t treat dragging their feet on a task as disrespect or disinterest. I don’t get hurt when they lash out at me during a hard time. I’m not baffled or disgusted when their words reveal their underdeveloped empathy, and I try to accelerate that development.
I just cannot agree more that it’s a difficult job to get right. All kids need structure and discipline, but the types and amounts can vary widely, even for the same kid across their stages of development. I’m so thankful for the research that has been done and resources that have been created since I was young; without that I would have been a lot more like my parents, and my kid would have barely had a chance. I also appreciate OP for sharing their experience with us. It’s not fun to do but it’s necessary for understanding to grow.
Given the coordination/cooperation aspect of the problem, this isn’t really the “wisdom of the crowd” as I’ve always understood it.
Something like estimating the number of beans in a jar is a good fit, since there is only one layer of perception to agree on and no coordination required.
This experiment as described seems closer to “design by committee” with (predictably) similar results.
This is wrong. You can absolutely dose psylocybin at levels that make a meaningful difference in experience (certainly enough to have an anti-depressant effect) without “tripping” or being in any way impaired. For many adults, this will be around 100-200mg of psilocybin.
That’s fine, but then we need to stop privatizing the AI. If a coalition of companies (and maybe government?) were to train open model weights, and run those models on equal-access hardware (i.e. all residents of the country get N compute credits per year) then we’d be much closer to a system where “intellectual works should automatically benefit society” — however, that is not what we are seeing today.
I read this book in the first few years of programming professionally, and in my naïveté I was so eager to apply the patterns therein I missed out on many opportunities to write simple, straightforward code. In some ways it really hindered my career.
I don’t blame this on the book, of course; ultimately the intuition it helped me build has been very helpful in my work. With that said, as a particular type of feisty and eager young programmer at the time, I can now say I would have benefitted at least as much from a book titled “Designing Data-Unintensive Applications” :)