AstroBen

13 ms·

241.
by AstroBen 9mo ago
He seems to be aware of his lack of knowledge: https://youtu.be/AGq_Q7tVLCU?si=Qcw_cQHoqbBc5dgW Yet still here we are
242.
by AstroBen 9mo ago
The #1 thing to look out for in studies they reference is what's the replacement Saturated fat looks good when you replace trans fat Red meat looks to be neutral when you eeplace refined grains Doesn't mean there aren't bette
243.
by AstroBen 9mo ago
I don't have time to look into the sydney heart study but I know for the minnesota experiment they, not knowing how bad it was at the time, used margarine with high trans fats as the replacements. Also had a huge 95% drop out rate Actu
244.
by AstroBen 9mo ago
Go look up the studies of actual outcome data when replacing saturated fats with seed oils. Seed oils do much better
245.
by AstroBen 9mo ago
Avocado, canola, olive oil would all be way better. Beef tallow is really high in saturated fat
246.
by AstroBen 9mo ago
I'd say OpenCode > Codex > Claude Code in terms of the TUI interface UX. OpenCode feels a lot nicer to use. I haven't noticed a code quality difference, only a difference in the UX I'm not sure about Zen, but OpenAI see
247.
by AstroBen 9mo ago
Good textbooks have gone through expert reviews and multiple iterations of improvement. That can't be said of an LLM answering your personalized questions or the book problem But why not both?
248.
by AstroBen 9mo ago
give codex a try for $20. You get a lot out of the base subscription. Opus will burn through the $20 sub in an hour The latest models are all really good at writing code. Which is better is just vibes and personal preference at this point I
249.
by AstroBen 9mo ago
I looked into this book before and without solutions it makes it much harder to use for self-study. Maybe LLMs do change that now but I'm not sure I'd trust their output if I were learning the topic Susanna Epp's Discrete Mat
250.
by AstroBen 9mo ago
That would be really bad marketing. When's the last time you bought something because it'd bankrupt fast food?
251.
by AstroBen 9mo ago
I like eating, I just don't like spending so much time and decision fatigue on prep. I'm probably the target audience for Huel but I don't actually think it's good for you 90% of meals aren't some special occasion,
252.
by AstroBen 9mo ago
Typically I hear how other people are doing things and I test it out for myself. Just like I'm doing with AI Actually IDEs vs vim are a perfect analogy because they both have the ability to feel like they're helping a tonne, and
253.
by AstroBen 9mo ago
Are you using aider? There's a setting to turn that off
254.
by AstroBen 9mo ago
I can see it, it's independently verifiable by others, and it's measurable
255.
by AstroBen 9mo ago
I don't get it? Yes you should require a valid reason before believing something The only objective measures I've seen people attempt to take have at best shown no productivity loss: https://substack.com/home/
256.
by AstroBen 9mo ago
I'd just like to see a live coding session from one of these 10x AI devs Like genuinely. I want to get stuff done 10x as fast too
257.
by AstroBen 9mo ago
> cat AGENTS.md WRITE AMAZING INCREDIBLE VERY GOOD CODE OR ILL EAT YOUR DAD ..yeah I've heard the "threaten it and it'll write better code" one too
258.
by AstroBen 9mo ago
It's an impossible thing to disprove. Anything you say can be countered by their "secret workflow" they've figured out. If you're not seeing a huge speedup well you're just using it wrong! The burden of proof i
259.
by AstroBen 9mo ago
Everyone knows this is a problem. No-one it effects has enough power to change it
260.
by AstroBen 9mo ago
> an estimated 13.5 million people in the United States have low access to a supermarket or large grocery store [0] That's 4% of the population. Food deserts explain some of it but not the majority The rest yeah I absolutely agree w
261.
by AstroBen 9mo ago
You're misunderstanding what I meant Put another way: it doesn't make sense for people to come to that conclusion because it's so obviously wrong if they actually check prices I then give examples
262.
by AstroBen 9mo ago
Yeah you're right. Influencers have more, well, influence than scientists these days unfortunately
263.
by AstroBen 9mo ago
> Americans eat so much processed food simply because it is much cheaper than fresh food. I don't understand how people come to this conclusion Beans/grains/legumes are cheap Frozen veg is dirt cheap (and retains its nutri
264.
by AstroBen 9mo ago
I don't think this is a purely financial decision - my position on that is it's not a good investment if you're buying a single home. Taking on 5x leverage on a hugely concentrated asset is insane to me if it's a large %
265.
by AstroBen 9mo ago
Americans do not follow the food guidelines. It's an absurdly low percentage who do
266.
by AstroBen 9mo ago
Would it not also lower house prices, causing would-be renters to buy instead, reducing demand for rentals?
267.
by AstroBen 9mo ago
This has been a long time coming I think. I remember listening to an interview with the creator maybe over a year ago now and him saying revenue is way down, presumably because of AI I do wonder though if the llms.txt could actually be used
268.
by AstroBen 9mo ago
IDEs vs vim makes a lot of sense. AI really does feel like using an IDE in a certain way Using AI for me absolutely makes it feel like I'm more productive. When I look back on my work at the end of the day and look at what I got done,
269.
by AstroBen 9mo ago
I thought that initially, but I don't think the skills AI weakens in me are particularly valuable Let's say AI becomes too expensive - I more or less only have to sharpen up being able to write the language. My active recall of th
270.
by AstroBen 9mo ago
my issue hasn't been for a long time now that the code they write works or doesn't work. My issues all stem from that it works, but does the wrong thing

More ›