cdb's recent activity

  1. Comment on What are you reading these days? in ~books

    cdb
    Link
    Finished Oryx and Crake by Margaret Atwood. I guess I don't have much more to say about this book beyond what I posted last time halfway through. I feel like most of the point of the book was...

    Finished Oryx and Crake by Margaret Atwood. I guess I don't have much more to say about this book beyond what I posted last time halfway through. I feel like most of the point of the book was already made before the ending was revealed, which was not much of a surprise. It seems like the apocalyptic result was due to a society with higher technology but lower morals and lower checks and balances than ours. In the book, most of the damage was done by an individual, but in modern times I think it's hard to get anything significant done without a large team working together in multiple stages. It looks like the current state of tech isn't empowering individuals to do society-level damage, but I wonder if future technology will eventually change that.

    Currently reading The Alignment Problem by Brian Christian, about halfway through. It's about aligning AI systems with human values and brings up some ethical dilemmas introduced by AI models. Although it's from 2020, I think it's modern enough to remain relevant.

  2. Comment on What’s your preferred work monitor setup? in ~comp

    cdb
    Link
    In the past several years I've gone through the following configurations: one 27" 1440p 16:9 one 35" 1440p ultrawide two 35" 1440p ultrawides stacked vertically (probably the best option) one 42"...

    In the past several years I've gone through the following configurations:

    • one 27" 1440p 16:9
    • one 35" 1440p ultrawide
    • two 35" 1440p ultrawides stacked vertically (probably the best option)
    • one 42" oled TV with one 35" 1440p ultrawide on top
    • one 40" 4k ultrawide with one 35" 1440 ultrawide on top

    For productivity moving to two ultrawides was helpful, but going larger on my primary monitor hasn't been that helpful. The oled TV was good for gaming, but not great for text clarity and window management. Even after trying for a while, I haven't managed to make good use of the extra height of the 4k monitor. It's mostly the width that helps in most cases for me. The 4k ultrawide is pretty good on its own in terms of real estate, but there are not many affordable options. So, I think two stacked 1440p ultrawides are probably the sweet spot in terms of having a large amount of area without overspending.

  3. Comment on What are some bands you regret not seeing live (or, just never had the chance to see in the first place)? in ~music

    cdb
    Link Parent
    I saw Hiromi with Stanley Clarke, and I would definitely regret it if I hadn't.

    I saw Hiromi with Stanley Clarke, and I would definitely regret it if I hadn't.

    1 vote

  4. Comment on Is higher education still valuable? in ~life

    cdb
    (edited )
    Link
    As far as I can tell, the advice to just go to college and figure the rest out later has never been correct. There's a lot of discourse about how nowadays a lot of college majors aren't worth it,...

    As far as I can tell, the advice to just go to college and figure the rest out later has never been correct. There's a lot of discourse about how nowadays a lot of college majors aren't worth it, but the narrative was the same 25 years ago for people paying attention. They were publishing stats on average income by major showing that many degrees were not worth the money back then too.

    What has changed is the trend to require more and more education over time, but I guess that's an expected result as society and technology evolve and the complexity of valuable work increases on average.

    I think the best thing to do is the same as before, although even more important to consider as educational requirements increase:

    • Pick a path that seems to interest you.
    • Do some research on what jobs that type of degree moves the needle on, whether they're hiring, and what expected salaries in those positions are.
    • Do the math yourself on whether the degree is worth your time and money.

    So, for your personal situation, what are your plans for that stats degree? I'm in a similar situation myself, getting an MS in CS despite not working in software. For me it's half hobby and half for professional development. Work is paying for it, so there's not much consideration as far as monetary cost for me, just the consumption of my spare time. I'm starting to angle myself as a person with domain knowledge that also understands things about AI and computing in general, which might result in good opportunities as companies sink more investment in use of AI techniques. On the other hand, if it doesn't result in a huge impact to my career immediately, Plan A to get a reasonably valuable degree, then a job in that field, has worked out fine so far. That path was not as linear as I hoped it would be, but that's another story.

    edit: To kind of argue against my own point, the unemployment rate for recent college grads is pretty high relative to historical values, while the unemployment rate for all young workers is historically low. Underemployment seems fairly steady though. So, if you can get a job, the degree is probably still worth it. I think this might support my point that a blanket statement on whether more school is worth it, and a personal analysis is more important than ever, though.

    3 votes

  5. Comment on Anthropic drops flagship safety pledge in ~tech

    cdb
    Link Parent
    From the Axios article linked elsewhere in this thread: I'm not a big fan of the government making threats like this either. I don't know if it's implied or not, but the article makes it seem like...

    From the Axios article linked elsewhere in this thread:

    How it works: The Defense Production Act gives the president the authority to compel private companies to accept and prioritize particular contracts as required for national defense

    The law is rarely used in such a blatantly adversarial way. The idea, the senior Defense official said, would be to force Anthropic to adapt its model to the Pentagon's needs, without any safeguards

    The Pentagon is also considering severing its contract with Anthropic and declaring the company a supply chain risk, which would require a plethora of other companies that work with the Pentagon to certify that Claude isn't used in their workflows.

    I'm not a big fan of the government making threats like this either. I don't know if it's implied or not, but the article makes it seem like it is two clear threats, which are both very serious. Either be compelled to do the DoD's bidding, or get cut off not only from DoD contracts, but from contracts with any company that works with the DoD.

    4 votes

  6. Comment on Anthropic drops flagship safety pledge in ~tech

    cdb
    (edited )
    Link Parent
    I think the issue with type of thinking is trying to apply a human model of cognition to a machine. If we're talking about humans, it makes sense to say that if a human can't even do tasks using...

    I think the issue with type of thinking is trying to apply a human model of cognition to a machine. If we're talking about humans, it makes sense to say that if a human can't even do tasks using elementary school level thinking, they are probably not good at tasks requiring more complex thinking. AI models are different, though. They might have some trouble with arithmetic or counting the number of R's in the word "strawberry," but they can do better than most humans at some more complex tasks, requiring either wide-ranging or esoteric knowledge.

    For instance, this week I asked copilot to put together a study guide to understand the technical and contextual aspects of a specific meeting discussing research based on the accumulated slides and meeting minutes from the past two years. This was to aid onboarding and to help experienced people address gaps in knowledge. This would be an impossible (or very very long) task for someone not knowledgeable about this type of research and a labor-intensive task for an experienced person, but it spit out a really good outline in about a minute. I don't know if it's perfect, but it's really good (I'm the experienced one trying to help a new hire). I haven't used the guide a bunch yet, but asking it to explain the first point also resulted in a good and useful answer. It's possible that it missed a few items here or there, but if you understand 80% of everything happening in this meeting, that's probably higher than average. Point is, there are a lot of questions that humans aren't great at answering that AI models can do as good or better at.

    2 votes

  7. Comment on Anthropic drops flagship safety pledge in ~tech

    cdb
    (edited )
    Link Parent
    Based on the part you quoted, I'm not sure how the conclusion is "money talks." Isn't this more like "do what we say or we'll ignore your policies, regardless of previously negotiated contracts?"...

    Based on the part you quoted, I'm not sure how the conclusion is "money talks." Isn't this more like "do what we say or we'll ignore your policies, regardless of previously negotiated contracts?"

    It seems that based on the Axios article, there were two possible courses of action by the DoD: cut off contracts, or force Anthropic to contract with the DoD on their terms. Even if it's about that contract money, they couldn't avoid the second threat unless they shut down operations.

    2 votes

  8. Comment on What are you reading these days? in ~books

    cdb
    (edited )
    Link
    Finished Food Intelligence by Belluz and Hall a bit ago. I felt like it was a book with good data but definitely had an agenda, so the narrative was slightly disconnected. The main one I have an...

    Finished Food Intelligence by Belluz and Hall a bit ago. I felt like it was a book with good data but definitely had an agenda, so the narrative was slightly disconnected. The main one I have an issue with is the demonization of ultra-processed foods. They suggest that the level of processing might be the issue, present compelling data showing that it's not the level of processing, but the palatability and calorie density, then keep on talking about ultra-processed foods like they're the problem, occasionally calling them "ultra-processed foods that are hyper-palatable and calorically dense" so they won't be technically inaccurate. The section on supplements was a little opinionated too, but seemed like it had less of an agenda, which I liked. Ultimately, I like reading about an interesting topic like this without having to dig into academic research, but I wasn't a fan of the presentation.

    Just finished Failure to Disrupt by Justin Reich. It's basically a summary of recent trends in educational technology such as MOOCs, adaptive tutors, microlearning (like Duolingo), etc. Long story short, they're all helpful, but haven't lived up to the hype, more like smaller step changes in a continuous process of improvement rather than hugely transformative and disruptive changes to how we learn.

    Currently a little past halfway through Oryx and Crake by Margaret Atwood. The setting is that there are a bunch of corporate compounds that separate the haves from the have-nots, and a lot of resources go into making frivolous things for the haves. The frivolous things giving me a bit of an Idiocracy vibe (the book came out first, so....). Seems like it takes the concept of a k-shaped economy to an extreme future with an apocalyptic ending, which is an interesting take given the popular discourse these days, where the top 10% of earners make up about half of consumer spending. Reading the headline about the fart-sensing underwear today also seemed relevant.

    5 votes

  9. Comment on The AI disruption has arrived, and it sure is fun (gifted link) in ~tech

    cdb
    (edited )
    Link Parent
    edit: i had a whole comment here, but I've reflected on the comment on sealioning, and decided that if you dont want to have a debate on this particular point, I wont try to force you or harass...

    edit: i had a whole comment here, but I've reflected on the comment on sealioning, and decided that if you dont want to have a debate on this particular point, I wont try to force you or harass you. Sorry if you felt like I made your day worse in any way. Not my intention.

    4 votes

  10. Comment on The AI disruption has arrived, and it sure is fun (gifted link) in ~tech

    cdb
    (edited )
    Link Parent
    I'd appreciate it if you wouldn't label my comment like that. I'm arguing in good faith. I'm assuming you're arguing in good faith. Let's have a conversation, not a fight. For context, I work in...

    I'd appreciate it if you wouldn't label my comment like that. I'm arguing in good faith. I'm assuming you're arguing in good faith. Let's have a conversation, not a fight.

    For context, I work in drug development. So no, I'm not suggesting we abolish the FDA. People aren't generally doing drug development in their garages, and if they do, the DEA might come knocking. Actual shutdowns are fairly rare, but they do happen. I tried briefly to find a good list but failed, so here's just one example from 2023. The FDA also issues warning letters more frequently, which are considered very serious and effectively hamper progress while the issues are being addressed. This can delay some processes for months, which amounts to many millions of dollars in costs. The last time I personally was involved in supporting an FDA audit, my department basically stopped all development work during that time to work on the documentation.

    So, back to computing, the most problematic issue to me is mention of frontier research and rollout. Lots of research happens at smaller scale and rollout can be just sharing your findings. How can you regulate that without widespread surveillance and enforcement?

    3 votes

  11. Comment on The AI disruption has arrived, and it sure is fun (gifted link) in ~tech

    cdb
    (edited )
    Link Parent
    This sounds like a recommendation for a total surveillance state, or a return to a pre-computer society. The FDA can just come and inspect your facilities any time they want, and if you refuse or...

    we have a highly structured, rigorous, and regimented process for the release of drugs (or we did before the MAGA disease struck). We should, hypothetically, be able to have an AI equivalent.

    This sounds like a recommendation for a total surveillance state, or a return to a pre-computer society.

    The FDA can just come and inspect your facilities any time they want, and if you refuse or they don't like what they see, they can shut down your entire ability to do business. There's a long and involved submission process with many steps that usually has hundreds or even thousands of people involved in working on various parts it. On the enforcement end, the DEA can investigate whomever they want and arrest them if suspected.

    Tying this back to computing, the government would have to be able to regulate the equipment individuals are allowed to have, monitor any capable equipment for misuse, and have a strong enforcement capability to raid anyone's house if suspected of illicit AI development activities. This sounds absolutely horrible to me. You might say that this is just for the big players, but what if people come up with ways to do more with less hardware? You'd have to be constantly checking.

    ETA: I don't think the issue with your idea is feasibility. It's absolutely feasible to do that. I think the main issue is that it runs into a bunch of moral issues.

    5 votes

  12. Comment on Why America is slipping in the Olympics in ~sports

    cdb
    Link Parent
    I guess the economics answer is that things that are limited in supply and high in demand should be expensive. That way those who really want to do it can, and those who don't prioritize it won't...

    I guess the economics answer is that things that are limited in supply and high in demand should be expensive. That way those who really want to do it can, and those who don't prioritize it won't consider it worth it. There is an element of inequality with this thought, but I would say that it's more of a vacation than a right. I'd say the same thing about concert tickets. Limited availability of a recreational activity means it's ok to be niche and expensive. This isn't milk and bread.

    Maybe you missed the last sentence of my previous comment? I'm all for doing sports and subsidizing them for kids. I just don't think they need to involve traveling to the mountains or building a bunch of giant freezers (ice rinks) in cities where it doesn't get very cold in the winter. Might make sense in some areas where you can learn to skate on a local pond in the winter, but I don't know if it's really needed to push winter sport culture if the climate isn't supporting it.

    4 votes

  13. Comment on Why America is slipping in the Olympics in ~sports

    cdb
    Link
    Seems like mostly a geography and culture thing to me. Just focusing on skiing, I feel like maybe it's a good thing that skiing is more expensive. There's a limited number of suitable mountains,...

    Seems like mostly a geography and culture thing to me.

    Just focusing on skiing, I feel like maybe it's a good thing that skiing is more expensive. There's a limited number of suitable mountains, and they're already pretty crowded. Mostly talking downhill skiing. There's a limited number of places that cross-country skiing is even possible. They mention an increase in median age of skiers, but it doesn't seem to be correcting for increase in median age of the population, which seems to make up for the majority of the effect. Looking at the cost, the $30,000 figure quoted for competitive skiers seems like it has to be mostly travel, given that equipment and passes are probably only several thousand. Is that even a problem? Maybe travel should be expensive, given its impacts on the environment. What is even the solution, here, subsidize travel for young skiers?

    Maybe we have bigger problems to use our collective funds on. I'm all for subsidizing local youth sports because of the community aspect, but I'm struggling to see why we should care so much about winter sports specifically.

    5 votes

  14. Comment on What are your food aversions? in ~food

    cdb
    Link Parent
    I'm the same with natto. I like a lot of fermented foods. I like a lot of foods that are slimy. I just cannot agree with both slimy and strong fermented flavor at the same time. It's just too...

    I'm the same with natto. I like a lot of fermented foods. I like a lot of foods that are slimy. I just cannot agree with both slimy and strong fermented flavor at the same time. It's just too strongly indicative of spoiled food, which is exactly what natto is.

    1 vote

  15. Comment on Need a replacement for my old macbook pro, should I just get another one? in ~comp

    cdb
    Link
    I just bought a used M1 macbook pro a few months ago, and I think it's great. For the price ($600) it's probably better than any windows laptop. I bought a MacBook Air m1 when it first came out in...

    I just bought a used M1 macbook pro a few months ago, and I think it's great. For the price ($600) it's probably better than any windows laptop.

    I bought a MacBook Air m1 when it first came out in 2020, kept it for a year, but sold it because I figured that it might have some compatibility issues with some things I was planning to do. I switched to a higher-end Lenovo yoga laptop. The display is great, the touchscreen is nice to have, but the battery life sucks, the daily use performance isn't nearly as good as the m1 air, and the fans frequently spin up while I'm not doing anything taxing. I also tried another Lenovo with a really nice OLED display and ARM chip, but gave that to a family member. I also tried an asus g14, which is a mid-high end compact gaming laptop, but I had a ton of issues with it. The HP I tried was fine, but just ok. After going back to mac, it really seems like if you want a good laptop experience, a macbook is the way to go. The M1 macbook pro in particular can handle all the tasks I need it to, including a lot of multitasking, coding, and some local LLM use. This is all while having a really nice display, great battery life, and reliable sleep. I guess this kind of sounds like an advertisement at this point, but I've tried a bunch of other options, all at the top of "best laptop" lists, and the mac has been the only reliably good one.

    2 votes

  16. Comment on Popcorn lung: how vaping could scar your lungs for life in ~health

    cdb
    Link Parent
    I think the issue is, in part, the inability to talk about things in a complex manner. Any negative news about vaping seems to warn about all vaping, when vaping is just a delivery mechanism. Just...

    I think the issue is, in part, the inability to talk about things in a complex manner. Any negative news about vaping seems to warn about all vaping, when vaping is just a delivery mechanism. Just like how you could smoke a variety of things (tobacco, weed, crack, or other random chemicals from shady sources), you can vape a variety of things too. The source of the vapes doesn't seem like a common thing to include in news articles. As far as I know, the only lung injuries that have been reported and have investigations on the source have been from bootleg weed cartridges that have ingredients that no legitimate manufacturer would include.

    In the case reported here, we don't know what this kid was vaping, but it's possible that a 14-17 year old might be getting vapes cartriges from sketchy sources. Not trying to be a denialist, but the surveys regarding prevalence of teen vaping suggest that over a million teenagers are vaping. If we're talking about a literal one in a million case of lung injury attributed to vaping, it should be worth looking into the details.

    17 votes

  17. Comment on AI hallucination cases - When lawyers use hallucinated legal content in ~tech

    cdb
    (edited )
    Link Parent
    I think it depends on what you mean by "causing a problem." Is it a new potential source of non-existent citations? Yes. Is it a problem in that it's increasing the error rate overall? Need more...

    I think it depends on what you mean by "causing a problem." Is it a new potential source of non-existent citations? Yes. Is it a problem in that it's increasing the error rate overall? Need more data to say.

    The main thing I'm trying to say is that without knowing the base-rate of things, that fact that this list exists or whether this list gets longer doesn't mean the situation is getting worse. Like you say, the overall benefit can't really be determined from looking at this list. I'm trying to say that the way human brains work is if we report on things like "bad list has many items" then there is a narrative and perception that this mean things are getting worse.

    I also see the potential for bias here. This site links to a paid service for verifying citations. There's an incentive to magnify the perception that this is a serious issue.

    2 votes

  18. Comment on AI hallucination cases - When lawyers use hallucinated legal content in ~tech

    cdb
    Link Parent
    I wonder if the narrow scope of focusing only on recording the number of cases where AI is used and was wrong is ripe for a base-rate fallacy interpretation. If we see this database getting...

    I wonder if the narrow scope of focusing only on recording the number of cases where AI is used and was wrong is ripe for a base-rate fallacy interpretation. If we see this database getting larger, one might assume this is an increasing problem. Without comparing it to the prevalence of AI use overall or comparing with the amount of incorrect references and arguments without AI use, it's hard to say if this is a problem that is getting better or worse.

    I mean, I highly suspect that AI use without proper review would result in errors, given that hallucination is a real and common thing, but I can't see this as evidence AI is causing a problem here without more information.

    2 votes

  19. Comment on A case for increasing computer literacy (but also a rant) in ~tech

    cdb
    Link
    What's not clear to me from this suggestion is: what's the benefit to the user? I'll go with the popular car analogy. It's a tool that serves a specific and important function. People need it to...

    What's not clear to me from this suggestion is: what's the benefit to the user?

    I'll go with the popular car analogy. It's a tool that serves a specific and important function. People need it to "just work" to go about their daily lives. Although they use it almost every day most users don't really know or care about its inner workings. Although some level of knowledge and tinkering used to be necessary to keep them operating well on a daily basis, nowadays knowing how to tinker with them is either a profession or a hobby. What's wrong with that? I think it's great that we can derive most of the benefits from computers without forcing some of these time-consuming activities on users, and the door is still open to pursue further understanding if people choose to.

    This is coming from someone who did tinker with computers growing up in the 90's, has some IT experience, has surely built more computers than 99.99% of the population, and has most of a master's in comp sci (I'm gonna finish soon, I swear). I have machines in use running Windows, MacOS, ChromeOS, and other flavors of Linux. I think they are all fine, and people should just use whatever is most convenient for them in the moment.

    7 votes

  20. Comment on I'm annoyed with mundane revisionist history in ~talk

    cdb
    Link Parent
    I was about to say that if people are obviously incorrect, the silver lining is that at least you can be pretty confident you're talking to a person, because AI models most often get things like...

    I was about to say that if people are obviously incorrect, the silver lining is that at least you can be pretty confident you're talking to a person, because AI models most often get things like this right these days, assuming you ask the question correctly. Sometimes I get mad about people posting things that don't survive a simple google check, but I have to remind myself that this is part of interacting with other humans. If I wanted agreeable all the time I could just talk to an LLM. So I did ask an LLM and it told me that DVDs were a major selling point.

    Here's a quote from wikipedia that it referenced:

    Former Worldwide Studios president Shuhei Yoshida acknowledged that Sony had been "awfully unprepared" for the transition from the original PlayStation, noting the company's limited experience in managing generational console launches. Due to a lack of launch software, much of the early consumer interest centred on the console's ability to play DVDs. According to Yoshida, in Japan, the best-selling title during the launch period was not a game, but a DVD of The Matrix (1999).[64] He remarked that while standalone DVD players were still expensive at the time, the PlayStation 2 offered comparable functionality at a significantly lower price point, contributing to its immediate commercial success.

    So, I think for this topic it ends up being a fine-grained detail regarding wording. Was the PS2 popular because it was a DVD player? I feel like there is a spectrum of interpretation for this statement that could put you on either side of it. Did people buy it solely to be a DVD player without caring about gaming? Probably not in most cases. Did the DVD functionality heavily contribute to the purchase decision? Probably in most cases. So for me, I would prefer to move on to the next point, because this one doesn't seem worth it to fight over.

    14 votes