bgirard

Do you think overall injury rates tell the whole picture when some of them are CTE? It would be more convincing if you were showing data about CTE.

I'd rather my child accept a high risk of sprain, than even a low change of CTE.

I know a ex-professional football player that doesn't let his children play football.

Same. I went to a music festival and their website was hard to use, and the schedule was difficult to understand. And the area had bad connectivity. A few prompts and I was able to build a much better offline site.

That's an example of AI growing the sector that can lead to more jobs. Because suddenly a lot of tasks that weren't economically viable are now going to be in demand. Custom software for small businesses, platform specific optimized code instead of cross platform software, etc...

It's a good nitpick. A hypothetical 1000Hz display will still look smooth to the human eye at 60/120Hz. A 1Hz display will never look smooth.

But it's more about honoring the user's preference. I personally would rather 60Hz, then 120Hz heating up my room unnecessarily. If someone has a display set to 1000Hz for whatever reason, then you should try your best to honor that.

Approximate calculation is putting the cost at ~$5 per million tokens (assuming 42W sustained, 20¢/kWh), and that's excluding hardware and other costs.

What makes you think the unaccelerated path would use the accelerated iGPU path? Typically hardware decoding is disabled because of driver/reliability issues and thus would switch to software decoding to sidestep these concerns.

It's not created out of no where. It's traders saying that the stock is over priced and putting their money where their mouth is and taking on a big risk if they're wrong. They're extracting this money from folks that are putting upwards pressure on the stock saying it should be worth more. They're helping price the stock more efficiency and reducing index trader's expose to an overpriced stock e.g. retirement funds.

It's really neat that the prompt was released!

I'm curious how many unsolved problems are tried against frontier models when they come out. Are we trying every problems against every release? What is the solve success rate? Is there a sub-community within Mathematics that is coordinating this effort? How much untapped opportunity is there here?

> As the person holding the money, it's my job to look at what is effective and what the active ingredients are in any given product.

But I don't have time to do that. I would rather have a retailer do that curation for me and provide me with effective high value products, and stand behind returns when they miss the mark. Then as a customer I can reward them for that value added work.

That's why Costco is great most of the time. Although they sometimes miss the mark with certain products they stock.

I don't recognized the CPU/GPU and PC building isn't my field so I could way off. But here's my honest attempt at it without paying a premium for the form factor which isn't an important feature for me:

PCPartPicker Part List: https://pcpartpicker.com/list/3WkCdq

Price: $1021

SpaceX was said to be subsidized by gov contracts. Look at where that got it...

You're missing the point. There was a lot of debate around if inference was subsidized or not. And that's a huge point to confirm in the public discourse.

Similarly my friend swaps electric cars every couple of years (Volt -> Bolt -> Equinox) bragging about all the discounts and subsidies he's gotten. Maybe it's still beneficial through the used car market but it doesn't feel like an effective subsidy for the government to be handing out.

A ground level fall can be fatal for a senior. Throw in freak accident factors and other vulnerable demographics it's not very surprising to me.

> It’s about bang for buck.

Hard to know when they don't give the price per token. Presumably it will be comparable to a low-mid range model in terms of price. But otherwise their 'Ideal Zone' is meaningless without factoring in the price per token. I don't how much tokens are being used, that's an implementation detail to me. I care about price / performance / latency.

Which ones? Mortgage, real estate costs, repairs, maintenance are all still there with a condo.

My gut feeling is that repairs and maintenance cost more with condos than if you own a home and you're handy to fix minor stuff and know how to find good contractors for bigger jobs. I imagine condo jobs becomes more difficult and contractors charge more for those jobs. But I don't have data to back my hunch. Condo has extra issues in dealing with neighbor problems (issues with garbage, pets, unpaid fees, noise, etc...) and you have to maintain shared spaces (hallways, elevators, etc...) and you end up paying for that via your condo fees.

I struggle with this too. But I remind myself that I'm buying myself one of the most expensive and valuable gift: freedom and independence.

I also continue to work because I enjoy it. And that will let me pass on this gift to my children.

Last week I got together with my math alumni friend. We cracked some beers, we chatted with voice mode ChatGPT and toyed around with Collatz Conjecture and we sent some prompt to a coding agent to build visualizations and simulation. It was a lot of fun directing these agents while we bounced off ideas and the models could explore them.

I think with the right problem and the right agentic loop it’s clear to me improvements will speed up.

I've been using Codex to build a repo that pulls down astronomical datasets and runs simulation to try to find explanation for the hubble tension. Having an agent to do the tedious bits and also having an LLM to bounce ideas has tough me so much about astronomy. I don't have serious hopes of finding anything new and novel but it's still a lot of fun.

I’ve noticed that LLMs can effortlessly read minified JS. How does it do with obfuscated binary code? I wonder if the days of obfuscation are numbered when the tedious job of de-obfuscation can be automated.

Subpoenas and whistleblowing are pretty good tools.

I think they're thinking about this wrong.

I get all my groceries deliver to my doorstep via Walmart delivery pass. The thing I'm really missing is having AI curate meal planning to my family's preferences. I already feed ChatGPT my family' preferences (e.g. Kid A doesn't eat X Y Z and liked meal A B C, kid B likes ...) and ChatGPT is helping me build meal plans. With my preferences we can quickly nail down a meal plan for the week.

The slowest part of my meal planning is going through Walmart's slow site where each page load is 2-3 seconds and it takes several page load per item. Once it can translate my meal plan into a grocery checkout from Walmart I'm all set.

I'd love to see the results of that. I think calling a single prompt iteration lifeless misses the point. It's like looking at a game that has had a few hours of development and saying it's bad. Games need iterations. Seeing your results as the first iteration is impressive. I can see follow-up prompts and custom tweaking get really good results!

Last summer I built a factorio-like automation game with older models and over time the game really started to take life.

It's very useful to understand what you're struggling from even if it's not curable. It explains your symptoms, your experience and help you understand what you're going through. Understanding that you're suffering from something incurable is also helpful in not looking for other ineffective methods to cure a mysterious illness.

> SpaceX has deorbiting assets on top of depreciating ones

The deorbiting part is redundant. Their satellite are just that, a depreciating asset. Their lifetime seem to be 5 to 7 years. The important claim is if the total cost, including the launch, can be recuperate over that lifetime or not.

I've gotten pretty good results by prompting "What did you struggle on? Please update the instructions in <PROMPT/SKILL>" and "Here's your conversation <PASTE>, please see what you struggled with and update <PROMPT/SKILL>".

It's hit or miss, but I've been able to have it self improve on prompts. It can spot mistakes and retain things that didn't work. Similar to how I learned games like Balatro. Playing Balatro blind, you wouldn't know which jokers are coming and have synergy together, or that X strategy is hard to pull off, or that you can retain a card to block it from appearing in shops.

If the LLM can self discover that, and build prompt files that gradually allow it to win at the highest stake, that's an interesting result. And I'd love to know which models do best at that.

Are there benchmarks if we allow the LLM to practice and study the game?

Your experience sounds exactly like mine. My son is very autistic as well. I've had to cut off friends with families because either their didn't understand meltdown and were incredibly judgy because they were blaming my parenting for his ASD meltdowns, or others because my autistic son was a "bad influence". God forbid their (later diagnosed) kid have some exposure to a child with different neurodiversities.

That's not even going into my traumatic health care experience to getting my son help when he needed it.

So now I have all the hardships of raising a family, and I'm restricted friendship within the small ND accepting community of my area. So my support network is incredibly small and I barely get any support. It sucks.

Reading the responses to your story that are nitpicking it over your daycare experience is a perfect representation of the problems that families face.

That's a good question. As someone bootstraping a few projects on Vercel this post has me looking over at the pricing sheet more closely.

That's true and I fully agree. I don't think LLMs' progress in writing a toy C compiler diminishes the achievements that the GCC project did.

But also we've just witnessed LLMs go from being a glorified line auto-complete tool to it writing a C compiler in ~3 years. And I think that's something. And noting how we keep moving the goal post.