incognito124

It's a bit of a mix between "but wait there's even more details!" and "life will now suck for the foreseeable future".

Thanks, I'll keep that in mind, but don't get your hopes up

mental reminder to finish my history of malloc writeup. It's almost 20 llm-less pages with context and examples. goes through general purpose malloc like dlmalloc, mimalloc, tcmalloc, jemalloc and others, while also mentioning some purpose-specific ones like scudo.

+1. There's a good book "Range: Why Generalists Triumph in a Specialized World" that lays out how it's better to be an all arounder than min-max skills and life areas. The advice actually predates LLMs!

I always felt repulsed by specialization due to diminishing returns and the fact im curious about everything. But I though about it and picking up various experiences to avoid a specialization is, in a sense, itself a specialization: it's unlikely there's another person with a similar set of experiences!

Reminds me of how Python is termed "2nd best pick for every task". It's rarely the best but it covers a lot of ground, so by starting with it, you can uncover what would be a better language. So a generalist can also be a first pick for something, at least until it's more clear about what's a better pick.

Anyways sorry for rambling, I agree with you. If a young person is reading this, the worst thing you can do is specialize too early (read the book I mentioned!); go explore the world first before settling and picking a direction

My favorite Windows alternative is Linux, specifically Manjaro

I am in that minority I guess that obsesses over P&S. If you want a recommendation right now, two most influential books I read are, in order:

* Probability and Statistics for Engineers and Scientists, that showed me how to calculate various percentages and p values and what does statistical testing really means * How to Lie with Statistics, that showed me how easily you can manipulate the numbers to get what you want out of them

Another neat one, albeit niche, is Directional Statistics. Turns out, doing stats on angles and directions is less trivial than it sounds!

Literally. If only elements came already presorted by some kind of a key already...

Tesla was educated in Graz and was only in Budapest for about a year

Serbia and Russia go way back, way before 20th century

That worked so well for copyrighted code and books

How does this differ from Astra's looped transformer? That architecture also reasons in the latent space

I don't even sed anymore. I go directly to ed while characters stream out of an llm

I'm the same as you. I hate asking other people, unless it's for a reference or material to learn. I like to grind to understand myself, and then answer questions and explain things via presentations or a drawing board. I never wonder why I'm like that, tho

That said, hardly anyone asks for explanations now. In my team people default to a chatbot for work questions almost exclusively, and in private life for hobbies I share with other people, they increasingly go for chatbots too.

Turns out, not everyone seems to want a human connection.

I wish Claude's raw output was at least moderately edited. This way it feels like I spent time and energy reading something no human wrote

Qwen 3.8 flash next is way better than 27B. It's so good I dont even use claude anymore

Blake Lemoine, and it wasn't Gemini it was an earlier system called LaMDA

That's exactly the case

PunchTornado was first, fr2029 was second

Love your username

I have a different, more sinister, analogy in mind but yours work as well

bzflag

genuinely amazing

take a look at myticker.com

(previously at https://news.ycombinator.com/item?id=45857053)

That intelligence is "just" retrieval within a huge corpus is an old one, and the one I am thinking about a lot these days: https://news.ycombinator.com/item?id=44060672

- https://github.com/elder-plinius/CL4R1T4S - https://github.com/asgeirtj/system_prompts_leaks

While I can't deny this is a huge technological result, and it's laudable they reduced the price because of it, 80% is really a lot. I can't help but wonder, is this because of the model's capabilities, or was the initial system just really sloppy? The public will probably never know the details

> The true purpose of AI is to allow wealth to access skill without allowing skill to access wealth.

This is such a fire quote

gemma 31b is my goto day-to-day model. It's adequately capable for coding and natural phenomena, and it follows instructions and skills reasonably well. My main negative trait is all the LLM-isms, but I can get past that for the value it brings

Another one I'm familiar with: https://www.verheesengineering.com/d2-twoseater/

Anecdote:

I learned juggling in approximately 12minutes with a friend who knew juggling from before. He spotted me like you would spot someone in a gym: he put the balls in my hands, told me to hit one up, caught it mid flight (effectively freezing time), and told me "ok now hit the other ball up and catch this one". After 6-7 tries my brain was reprogrammed, and by the day's end we did juggling in pairs.

I guess the moral of the story is that a great teacher can push your limits way further than you could yourself? Or that it's easier to learn in a group setting?