These are great, thank you so much for sharing the recommendations. I tuned in to NTS and casually just kept on listening for a very long time. If anyone else has good recommendations, I'm all ears. Thank you.
Many built-in functions are open source too. Use the "PrintDefinitions" ResourceFunction to see the code of functions that are implemented in Wolfram Language itself.
With every law we allowed in the name of child safety, we enabled the real goal, control. They barely have to even mention children in the new laws beyond "we have to legally cover ourselves against all the other child safety laws already in place."
Oh, I was not comparing it with out-sourced development and was instead comparing it with developing it oneself.
Sure, outsourcing is similar, but the difference is one uses a process that is inherently probabilistic, while other just depends on the probability of you getting a good team.
You can't literally tax 100%, but this matters little: you just have to tax a sizeable fraction of land rent and this has as an outsized effect on land values due to capitalization, which vastly improves the effectiveness of the real estate market as a whole. You see this today partially, where states with high property tax rates tend to have broadly lower real estate costs.
Except we didn’t even refill it last year when oil prices did fall to rock bottom.
Because the U.S. government (even generally…there’s obviously no hope for this administration) is structurally incapable of making decisions that have a benefit horizon of more than a couple of years.
Assuming that these people (like me?) are "illiterate" and evil and didn't speak out about worse problems because they happen to disagree with this specific (also terrible) development is not a very charitable interpretation.
But I agree it's not worth trying to engage with that kind of rhetoric.
It's a biological computing kernel that:
- Processes 109,352 DNA sequences in parallel
- Tracks 13,271 mutations in real-time
- Uses 8 workers across 4 CPU cores
- Serves a REST API for analysis
- Runs a live terminal dashboard (curses)
- Sips at 40% CPU like it's nothing
That's 5.4x more genes than the human genome.
No bioinformatics background. No tutorials copied. No "hello world" detours. Just Python, FastAPI, and a stubborn refusal to follow instructions.
The monitor updates every second. The genes keep mutating. I just watch and drink coffee.
For x = -i, square(x) = -1, sqrt(square(x)) = i. Meanwhile, abs(x) = 1. You're right that it simplifies to abs(x) for real x, but that no longer holds for arbitrary complex values.
Hey HN — I'm the author. Let me give a proper intro since the original title didn't do a great job explaining what this actually does.
HaiInvestor lets you pick any stock ticker and watch 6 AI personas — each modeled after a legendary investor — debate it in real time:
Warren Buffett (value / moat-focused)
Charlie Munger (mental models / inversion)
Michael Burry (contrarian / macro risk)
Peter Lynch (growth at reasonable price)
Cathie Wood (disruptive tech / long arc)
Bill Ackman (activist / concentrated bets)
Each agent reads live news headlines + financial data, then outputs a BULLISH / BEARISH / NEUTRAL signal with a confidence score and reasoning in that investor's voice. A Portfolio Manager agent then synthesizes a final BUY / SELL / HOLD verdict.
There are three modes:
- Single Stock — debate on one ticker
- Portfolio — analyze your whole holdings mix
- Backtest — test a strategy over historical data
A few honest caveats: this is purely educational/fun, not financial advice. The agents reason from news only (no full fundamentals yet), and LLMs can hallucinate. I'd love feedback on the multi-agent architecture, what data sources to add next, or any other ideas. Happy to answer questions!
Basically it's because what "AI" can do is extremely different from what "AI evangelists" claim it can do.
I haven't seen a single "AI evangelist" address any concerns and limitations, other by than "throw more AI at it" or "it will get better in 5 years, just in time for cold fusion".
> you create a full blown native Mac app, with a single sentence
Like they created a full blown C compiler that "could compile linux" but in reality didn't pass its own tests?
If you gate access behind a Terms of Service, any violation is potentially a felony in USA. Any human who later litigated would have clicked Accept, or subverted your popup like a hacker.
> You can't prove or disprove anything with someone who refuses to comply with the courts.
I don't understand how someone could even think this.
Suppose he stole the loot and refuses to say where it is, but he really put it in a bank safe deposit box. The bank teller remembers him coming into the bank with a big pile of loot and then leaving without it, so you use the teller's statement to get a warrant to search the box, get the camera footage from the bank, etc. There are many ways to prove something without the suspect's cooperation.
How is the alternative supposed to work? The judge tells you to answer a question you'd only know the answer to if you were actually guilty and then you stay in jail for as long as you don't answer it? What are you supposed to do if you're innocent?
Are you sure that these people were "gloating" about private companies moderating arbitrarily? In the US, this is simply an obvious consequence of the first amendment, as many people have indeed pointed out.
The same people may personally hold wildly different beliefs as to whether this legal situation is desirable or not.
And in European countries, where there is no first amendment preventing the government from interfering with social media moderation policies, the situation is often different, and courts have required social media companies to publish speech which they had intended to moderate; see for example:
> the input data is usually pretty terrible. It's based on either similarity
I found luck just using a LLM to chat about my tastes, what I like, what kinds of songs I want to discover ... it does a good job and is able to also give me background.
> The feature doesn't seem to justify the cost. All I can see is added marketing 'value'
You think marketing value does not justify added costs?
> Why not clean the tongue with the bristles!
Same reason you don't use the bristles to brush your hair. It doesn't work as well as a solution made for that purpose.
> Government officials know that children's hormones are being disrupted and they not only let it happen
Government officials define exact safety thresholds and rules for product materials to prevent health problems. That's why today we are debating trace amounts of something in plastic instead of lead in paint causing nationwide crime spikes and cognitive decline.
Everything you described increases the cost of attack (creating a cheat), and as a result, not everyone can afford it, which means anti-cheats work. They don't have to be a panacea. Gameplay analysis will only help against blatant cheaters, but will miss players with simple ESP.
It's almost the same as saying "you don't need a password on your phone" or something like that.
The nice thing about not being in jail is that you have the freedom to choose where and how you live. Feel free to move into a shack in the middle of the woods away from everyone. Plenty of people make the choice to live as hermits or shutins because they don't want to deal with other people or the demands being a part of a community places on them.
The last 3 books I’ve purchased from Amazon (UK) have been of questionable quality. The most recent was Designing Data–Intensive Applications (O’Reilly) and I’m still not sure if it’s print on demand, counterfeit- or just a reject. Roller marks, damaged pages, slightly off print. The returns process is inconvenient, one-offs are okay but on multiple purchases it’s fatiguing and so the book stays.
This isn’t specific to Amazon, I had the same issue with Waterstones in the UK (online)
I now just buy second hand (Abe, WOB) and hope for the best.
