Any idea of how it handles context length? I’ve been using VS Code Copilot, and it’s great up until the AI assistant hits its context cap. I know that when it says, “Summarizing conversation,” that the assistant has been lobotomized and replaced with an agent who has all of the confidence of the last assistant, but basically no idea what it’s really doing.
It’s at this point that it’s most likely to screw up the codebase. I can’t imagine how it’s supposed to be autonomous when it suddenly becomes brain dead.
Ustice
Karma: 280
Ustice from the past, I disagree about one point. The preferences you were referring to not being moral questions is wrong. Obviously one could only include humans, or a subset of humans. Including humans in “it’s just nature” is obviously immoral.
The choice of where to draw the line is very much a moral question. At least all humans is a pretty good Schelling point, but after that it there are only a few major points before things get murky, and rational moral minds can disagree.
I don’t believe you meant to be dismissive, but your argument was, and that’s not fair. I apologize on his behalf.
This is really cool analysis, but I think your conclusions are off.
I think this is what happens when you optimize for attention. Especially with user-generated art. I know I’ve watched more “incest” porn in the past few years—because it’s hard to avoid—yet, I’ve contributed to that tend. Gotta give people what they want, right? Bleh. Porn is subject to the same market pressures of enshittification that other businesses on to internet are under. YouTube is a great example, but so is Facebook, Amazon, MySpace, Reddit, and many more.Most of this is fantasy role-play, not real desires. People that like Little kink aren’t pedophiles. Furries don’t want to fuck real animals. Dommes aren’t sadistic assholes.
I’m not saying that there isn’t problematic and porn. There is, and it’s gross. But being squicked out by someone else’s kink doesn’t mean their kink isn’t okay too.
I’m pretty convinced at this point that qualia comes from systems with a certain level of complexity, or maybe it comes from life. I’m very skeptical of the idea that only humans experience things. I think it’s highly likely that ants have inner lives.
Where you lose me is the assumption that all insects are suffering. I expect that, like people, most of them are going about their day to day. Their lives have meaning to them. Yes, there are a lot being injured right now, but the VAST majority are because they were just going about their day. Life is hard.
To me, your argument breaks down when you take it further. What about the single-cell organisms? They also respond to negative stimulus. I don’t know if the question of suffering of single-cell organisms has really been studied, but even multicellular creatures are essentially colonies of cells of multiple types and even species. Wikipedia says there are about 3x10^13 human cells in a human body, with about that many bacterial cells. Should we be horrified at the desiccation and death of skin cells? The countless bacterial cell you are likely murdering every moment of your life as your cells fend off infection, and so too the cells of yours that die in the effort? Of cancer cells, who just want to live?
These things are a part of nature. Intervening is just picking sides out of preference. It’s okay to have preferences, but they aren’t moral questions.
To bring it back around, I’m with you on preserving habitat, and respecting life. When I find insects in my house that I can catch and release outside, I do. I think we need better regulation on wide-scale insecticide use.
When a mosquito bites me, I’m going to kill it. If they’re living in my home, we’re at war.
Hey, thanks for the interesting perspective. 👋
I think it’s easy to lose perspective when you multiply a really big number by a really small number.
Oh man, of people I’ve interviewed, the college graduates are next to useless. There are exceptions, but that’s true of those that have less traditional backgrounds too. There are way more talentless hacks than skilled professionals. Even at the graduate level.
If they’re there because of a paycheck, you can keep them. I want the people on my team that do it because they love it, and they have since they were a kid. They’re the ones that keep up, and improve the fastest—I am certainly biased.
With the new generative AI assistants, we’re going to have way more who are new and dabbling. Hopefully more of them are inspired to go deeper. But you know what, even shitty software that’s e solves a task can be useful.
Assuming AI is aligned, I don’t see how self-replicating machines would be useful. I doubt there’ll be much pressure for that. If AI is malicious, self-replication hardly matters.
The real-world data is likely the real obstacle. With enough, I expect they’ll be able to compensate for manufacturing defects, and have regular maintenance. I assume they’ll be able to self-improve with more experience (and likely pooling experiences)
It’s pure speculation though.
You aren’t looking for professional. That takes systems and time, and frankly, king power. Hackers/Makers are about doing despite not going that route, with a philosophy of learning from failure. Now you may be interested in subjects that are more rare in the community, but your interests will inspire others.
I’m a software engineer by trade. I kind of think of myself as an artificer: taking a boring bit of silicone and enchanting it with special abilities. I always tell people the best way to become a wizard like me is to make shitty software. Make something you know, something small, and that sounds like fun. It’ll be terrible and barely functional, but it’ll be yours, and the next one will be a little less shitty. Keep at that, until you can start thinking at higher levels of abstraction. eventually, your work will be less and less shitty, and before you know it it’ll be good.
Hackers/Makers encourage amateur work, because that’s where most people are, because they’re just starting out. Make no mistake, there are professionals in the community, but no one expects you to be up to their standards. Instead, they’ll help you make things a little less shitty.
Make it. Then make it work. Then make it right.
Failure is your teacher.
he term I tend to think that one of the biggest problems in politics is that words become shibboleths within subcultures, and people in these communities don’t even know it. I recently had that starkly brought to my attention with the word “hacker”. My girlfriend was talking about setting up some home automation idea, and I said, “I can do it. I’m a hacker.”
She was confused. To her hacking means software intrusion through exploitation of security vulnerabilities. To me it means repurposing existing systems to do something not originally intended. I tend to think of her definition as “movie-style hackers”, or the white-hat/black-hat hackers.
I grew up in the 80s and 90s. I was one of the fortunate kid’s that had access to computers, and the ability to learn from reading manuals, and my own experimentation. I naturally fell into the, what we might now call the Maker culture. To people like us, (e.g. Adam Savage, and Corey Doctorow), “hacking” is a creative and explorative exercise.I use like, “I can hack something together,” meaning “I can make it by repurposing some stuff I have laying around.” Or like, “that’s a clever hack,” meaning, “that’s not how things are supposed to work, but you found a clever work-around.” It’s even bled into mainstream US culture with terms like “life-hack” and “hardware-hacker.”
We in the community would call intrusion “cracking.” As in a “safe-cracker” or a “password cracker”. We use other words differently too, like “autodidact” being a term of pride as someone who learns on their own through exploration and research, rather than through formal instruction, which I’ve learned in the research community is a pejorative, ironically in this instance as a synonym for “a hack,” meaning not of professional quality.
So when my girlfriend and I had this disagreement, we went to several dictionaries. All of them supported her definition, and didn’t even list mine. I didn’t see my usage until I saw the “Hacker Culture” on Wikipedia. That’s when I realized what was going on: I had been using the term as a shibboleth.
Once I realized that, tabooing the word in our discussion got us back on track, but it really highlighted how invisible subtle differences in definitions can make communication difficult and cause disagreements. I think that most political disagreements suffer from this. I’m really trying to keep an eye out for that feeling of confusion that comes from talking with someone using the same words differently, and the hubris that comes from insisting that my definition is right.
Well, this comment is long enough. I’m in the weeds with work, and I really need to hack my way out of. 😉
I don’t think that Bayesian reasoning is very popular outside of this community, useful or not. You’re basically describing a directed weighted graph. I wonder if any mind-mapping software would have those added capabilities.
I’d argue that the second result from the values change comic was a positive result, as the character is expressing concern over the chang [sic]. 😆
This shouldn’t affect your conclusions.
I agree.
I’m not a twin, but I am a parent, and I have a a nephew, and my son has a stepsister who has called me Uncle Jason since she could talk.
I don’t feel closer to my nephew than I am with my “niece.” I normally wouldn’t make a distinction based on genetics, except that it is relevant here. I’m not closer with my sister’s kids than I am with the other two.
Also, I’m not sure closeness is really even a good distinction. I’m not generally responsible for my niece or nephew, but if they or my son needed me to travel across the country to rescue them from some bad situation, I’d do it. I love those kids.
Being responsible for a child may present as being closer to them, So does spending a lot of time with a child. One could argue that these are two aspects of closeness. Neither of those things have anything to do with genetics.
Personality can be a huge factor in closeness too, and there is a huge variation in personality, even amongst identical twins.
Genetics seems only tangentially related to closeness, and mostly because the vast majority of children are genetically related to their parents. Family is complex, and often has more to do with shared history than anything else.
I describe myself as a techno-hippy. I was reading Cory Doctorow while waiting for the next chapter of HPMoR. I’ve often wanted a good leftrat community. I feel ya. I have an intuition that Consequentialism is a dangerous philosophy to adopt for optimizers, especially when they get scared. It’s not a huge hop from “saving the world” to “saving people from themselves.”
When I was in my twenties, I’d spend hours a day in forums and on Reddit. I’ve moderated and participated. Now, I’m 47 I just don’t have it in me. Planting a banner works, but it’s a lot of work.This is about as close as I get to social media nowadays. It’s too easy to get people angry online, and to use that anger as a weapon. Right now I’m hoping I can guide my son through his teen-age years without him being weaponized. Hopefully kindness and the virtues of Stoicism stand stronger against inversion than Consequentialism.
It seemed like part of the problem is that Claude can’t think and speak separately. When I’ve been instructed in meditation, they guided me by focusing on my breath, or on some sound or set of syllables: something repetitive, without words.
When using inference time compute, I wonder if it would be possible for the system to use a mantra, and maybe try to reply with a random response, unrelated to the prompt. I’m going to experiment, but I’m a novice myself.
Yeah. I also note down ideas. Most of the time it’s just a part of my habitual writing. Since Notability allows me to search, it’s usually not too hard to find.
Recently, a friend asked me a difficult question, which I needed to consider and process before answering. My journal entry for that day included my musings.
I also realize that use it professionally too, when I’m working out a problem, or when I need to make lists. I’m a software engineer, so that’s not uncommon. For a while I was keeping a work journal, but now that’s sort of been subsumed.
I do make dedicated entries on topics too. I have a whole section for a game I run. I often reference those when we play.
Notability isn’t great for linking between notes. If you need that, I’d find another app for sure. If you’re writing out notes that you want to be able to search for later, I’d recommend it. That’s especially true for handwritten notes.
I’m curious to know what you land on.
Have you tried this in ChatGPT o1? With the chain of thought reasoning built-in, it might be interesting to see how it compares. You can also look at the chain of thought summary too.
It started with RPG notes because I was using them to help me keep track of the details. Hand writing kept me engaged. I generally didn’t have to do much pausing.
Later, as I was reading some of my notes it got me that I had a better record of what my fiction al characters did in than my own life. I realized how useful it would be to have a journal.
My level of detail varies. I tend to be more detailed when there are interesting bits than not. It also depends on how tired I am when writing, and when I’ve done it. Most of the time, I write about my day at the end of the day, but sometimes it’s the next day. When I do that, it rarely has the same level of detail.
I think that I reference my notes about every other month or so. I’m not really sure. Usually I’m looking up what I did on a particular day.
I’m use Notability, but mostly because I prefer to use hand-written notes. What I like is that I can hand write my notes, and then be able to do a text search on them later.
It started with me taking notes while playing RPGs, but turned into a daily journal.
If you don’t care about handwriting, my only real suggestion is go with something that saves files in Markdown format. If the company goes poof some years down the road you want to be able to still access your notes.
My understanding is that IQ tests measure some dimensions of intelligence, but not others. That matches my experience.
Back when I was a kid I was administered an IQ test. I scored in the upper percentile, I’ve always been pretty good at reasoning, pattern recognition, and creativity. Those were qualities that the test I took measured, but I struggled (at least relatively speaking) with social, situational, and spatial awareness.
I think determination would be another trait wasn’t being tested for, but that’s important for hitting targets in a large search space. It’s quite likely that things have changed. It’s been 40 years now.
I totally got carried away. 😅
[Here’s what I did](https://github.com/jeffkaufman/comments-selenium/pull/1). I don’t even know if it is a real suggestion anymore. Maybe you’ll find inspiration from some of it. Maybe not.
I don’t think China is willing to accept yielding. I can’t think of any reason that they would.
This is totally a shower thought, and I don’t trust it, but what about a strategy of semi-cooperation? China has been contributing to open source models. Those models have been keeping up with, and catching up to the capabilities of the closed source models.
I wonder if mutual containment could come through having similar capabilities as we both learn from each other’s research. Then neither side has a gross advantage. Maybe it doesn’t have to be zero-sum.
.