- gdhkgdhkvffThose darn sneakers are just too delicious!
- The difference is ISPs usually have monopoly/duopoly pricing power and LLMs already have freely available open source models. If one AI company decides they want to start gouging, they have to compete with other providers AND open source. And if all of the ai companies start colluding on price gouging, there’s always the option of new competitors cloud hosting open source models.
That said I do think eventually prices will increase somewhat, unless SOTA models start becoming profitable at current prices (my knowledge is at least 6 months old on this so maybe they have already become profitable?)
- Only ~14% of respondents had <=5 years of dev experience. And 55% were >10 years dev experience.
One thing I did find funny though was this survey found that devs overwhelmingly visit stackoverflow aka the site that puts the survey out found that so many people that use the site, use the site.
- I was confused when reading the abstract so I ChatGPT’d a summary and it just so happened to include this exact explanation:
1. What does a “critical regime” mean?
In neuroscience, a “critical regime” is like a sweet spot between too much order (where the brain is slow and rigid) and too much chaos (where it’s noisy and erratic). In this state:
* The brain is highly sensitive to inputs.
* It’s capable of flexible responses.
* Some researchers think this is ideal for things like learning, memory, and information processing.
BUT — that’s during waking states.
During sleep, especially NREM sleep, the brain is supposed to be less active so it can:
* Consolidate memories,
* Clear out waste (literally),
* Reset emotional balance,
* Rest and repair.
- Made a “machine learning movie recommendation” website. It went through some relevant questions and had you select some of your favorite movies to base the recommendation off of. Then it “calculated” and finally recommended Weekend at Bernie’s to everyone regardless of what you answered. Then gave you an option of another movie, which was always Weekend at Bernie’s 2.
I had no specific reason to build it other than I thought it would be funny.
- And for reference, it takes around $10/year to run a single efficient indoor LED lightbulb. So charging a cell phone for a years-worth of usage costs less than 1/10th of running an efficient LED lightbulb bulb for the full year.
Again, cell phones are just confusingly not energy intensive.
- One point missing from this comparison is that cell phones just don’t take all that much electricity to begin with. A very rough calculation is that it takes around 0.2 cents to fully charge a cell phone. You spend maybe around $1 PER YEAR on cell phone charging per phone. Cell phones are just confusingly not energy intensive.
- I really appreciate you taking the time to write this all out for me. I’ve read through it several times now with a couple breaks between.
If I were to summarize, it’s about enjoying the process rather than the result. Much easier said than done, but the ideal goal. The thing that keeps me and people like me from enjoying the process is the constant background of “is this going to be worth it”. Perhaps just the awareness of seeing this happening in real time could help in cutting short that enjoyment-blocker. To call it out and label it as such.
I’m not looking for a response from you, but wanted to take the time to thank you for writing this all out.
- This comment really resonates with me. The annoyance at the label, the realization of internet addictions being a coping mechanism for uncontrolled stress (subconsciously finding myself scrolling hackernews or Reddit on my phone when I run into a problem that doesn’t have an immediate solution), the feeling of failure from not being able to be productive 100% of the day.
I’m glad that you were able to find a solution, and I’ve heard some others(and ChatGPT) say similar things, but I never understand what it means to “solve stress”. Like what does that mean? To me, stress isn’t a singular task that can be killed/solved, it’s just a long running background task that takes up more resources than it should. Likely you won’t want to get into your personal life here, but can you give an example(even if you have to make it up) of what it meant to remove the stress from your life?
- Googling a question and finding an incorrect answer every now and then doesn’t mean that googling is useless. It means that you need to learn how to use google. Trust but verify. Use it for scenarios where you aren’t looking for it to be the trusted fact checker. It excels at brainstorming, not at fact giving.
- Most people do see productivity gains from using LLMs correctly. Myself included. Just because some people don’t learn how to use them correctly doesn’t mean LLMs aren’t helpful. It’s like when internet search came out and a handful of laggards tried it once, failed to get the exact perfect result, and declared “internet search is useless”. Using a tool wrong is not evidence of the tool being useless, it’s evidence that you need to learn how to use the tool.
- Why is that a guideline?
- Some quick searching says openai had a headcount of 3500 employees in 2024. It also had a valuation of $150billion. Another search says $150 billion companies tend to average 10000-50000 employees. So it does seem like they’re doing more with less. A different search says average company with $3 billion revenue has 5000-10000 employees. Either way OpenAI seems to be running leaner than average.
- I’m not sure what you’re actually trying to ask here.
Are ai advancements being OVERLY hyped? Yes. That doesn’t mean there’s nothing to them, though.
Are ai advancements being hyped by paid ai hype shills? My guess would be yes. Again, that doesn’t mean there’s nothing to them.
Will the max usage end up being as high as current hypers are hyping? Unknowable. There is SOME amount of false hype, but also SOME amount of meat to the hype, and SOME amount of continued improvement in ai abilities. If ai abilities (not just underlying model strength, but also the ability to harness models in real use cases) stopped increasing today, then the total realized gain from AI would fall short of existing hype. If ai abilities continue to increase, then who knows when the hype bubble will pop.
Either way, when the hype bubble DOES pop, it will fail to reach the usefulness of peak hype, but that “when” could be tomorrow or could be after 10x improvement in actual usefulness.
- For one, you can’t debug c# code in cursor without using a hacky third party extension. Because the c# debugger is only licensed to run in official vscode instances. And only way you find out is you try to run c# and get a runtime error saying that it can’t run for that reason, you google/chatgpt the issue, find your way to some old GitHub Issues threads where someone mentioned that’s a possible solution.
- I don’t want to pretend I know how bank analysts work, but at the very least I would assume that 4 bank analysts with reasoning models would outperform 5 bank analysts without.
- Actually, foreign companies are banned from owning more than 25% ownership of a tv and radio broadcast licenses in the US. And that law exists for this exact same reason, to avoid a situation where another country has the ability to control media consumed by the US.
- “That fact that it was a drop in the bucket for them makes it that much more outrageous, not less”
I disagree. I get the point that you’re making. That they could have more easily NOT done it. But I would be a lot more ensconced if these people were putting up 50% of their net worth on these bets.
And again, I fully agree that they shouldn’t be able to trade individual stocks. In my past I was a dev at a private wealth management company. While working there I was completely barred from trading individual stocks because it’s possible that I could have come across nonpublic info in the company because they would do internal audits for some entities. It made sense. Congress is an even bigger deal because they literally write the rules of companies that can affect stock prices. I was barred because I could have passively found nonpublic info, but they can actively cause the situations that cause price movement.
- Younger generations being anti-war is a tale as old as time. Unrelated to the current “TikTok-fed generation”. Meaning TikTok isn’t making them enlightened, that’s just how younger people always are.
- And in fact, currently 2.5% of the sp500 is meta. So if these guys just have 100% of their net worth in the sp500, they’d have more META than these two transactions.