309
points
loganfrederick
Joined 2,259 karma
loganfrederick [at] gmail.com
http://loganfrederick.com
http://www.linkedin.com/in/loganfrederick
- A great book on this subject but I didn't see cited in the article is "When Money Dies": https://www.amazon.com/When-Money-Dies-Devaluation-Hyperinfl...
Uses a mix of old economic stats paired with German citizens journals and public newspapers to tell the story of how hyperinflation in Germany started and then ended, but not before creating so much distrust in society that the seeds were sewn for the country to take a militaristic stance to fixing its problems. Highly recommend the book above.
- I dunno, I've met enough people first hand who believe in heavily regulating AI researchers to the extent that private companies shouldn't or couldn't do it without being owned by the government. It would surprise me that a handful of staffers in either the intelligence agencies or Biden executive branch also thought this way.
- The research analyst at SemiAnalysis put out a Twitter thread yesterday explaining why he thinks the impact to the global supply chain is overstated here: https://x.com/SemiAnalysis_/status/1840871017746698617
"In conclusion – existing wafer inventory provides a buffer as mining operations restart, major companies are already using synthetic methods to produce semiconductor-grade quartz crucibles, and there are other sources of HPQ. Spruce Pine FUD is exaggerated."
- 699 points
- Yep, it is definitely worth clarifying that all the code has to be reviewed manually still. An aside is that the LLMs (both Claude and GPT) have common failure modes in using Godot 3 syntax instead of 4, etc.
I meant to make the point that, having worked on mobile game dev recently, if the behavior I'm thinking of seems simple, my workflow instincts have become to go to LLMs first rather than Github. Very possible this is a bad reflex. Just one I've observed directly from my experience implementing similar functionality as the parent library.
It looks like spywaregorilla's comment was downvoted at least once but I think it was making the valid point sarcastically that the idea of "LLM generated code replacing commonly used libraries is a bad one." My original post was positing on whether this might be a future trend though.
Also want to recognize the author of the library (prophesi) commented down below and definitely applaud their work in packaging and open sourcing their work plus making the tutorial video.
- As someone who has recently been doing Godot mobile development with Cursor as my IDE, this repo and the video in another comment gave me some quick thoughts.
1. The main thing the library seems to offer is some pre-packaged libraries for mobile touch controls in GDScript, the custom language for the Godot game dev framework.
2. I have recently come across similar problems to what that this library tries to solve.
3. However, the way I went about solving it was basically using Claude and ChatGPT as my co-developers and asking them to answer coding questions. Here’s a quick example prompt about coding GDScript similar to some of the attributes of the library: https://chatgpt.com/share/c8e713e3-0662-4c7d-8645-9a76b0c032...
4. An optimistic perspective from all the above is that AI coding assistants can make it easier for people to develop libraries like this faster.
5. A negative view would be that AI coding assistants mean people don’t necessarily need “libraries” like this if an LLM can spit out the same results in a couple queries (rather than hunting around Github or the internet for pre-packaged code).
- 2 points
- I think the conclusion is aligned with tournament player intuition as reflected in the official character tier lists.[1]
The conclusion is: "Assuming this conjecture is true, I believe it means you should entirely ignore player strength when picking players, and should only focus on factors that aren’t tied to skill, like character matchups."
I say intuitively because for decades now Smash players have been making tier lists, with "official" ones selected by the world's best players. And the point of the tier list is that "assuming players are as good as humanly possible, which in-game characters are generally superior to others as dictated by the game's implementation."
And the essay does point out the importance of character matchups, though Smash tournament players will often consider that when playing multiple matches against opponents as well.
[1]: https://luminosity.gg/news/smash-ultimates-2nd-official-tier...
- There's actually at least a few cybersecurity related companies in the current batch. Not to mention Reality Defender (https://realitydefender.com/) from a couple years ago who have been doing well. YC is absolutely interested in the space. If anything, they may not have included cybersecurity in the RFS because they're already seeing and funding founders tackling it!
- I have no direct connection to Rob Henderson but he's one of my favorite writers just from his Twitter and Substack because he dives into issues that are very close to my heart and life experience.
I've written past comments on HN about how I was raised by a father who was a professional con man who committed credit card fraud, drove family into bankruptcy, then ghosted on my mother and I. The depths of that process was deeply chaotic and disruptive at a formative time in my life (high school and college).
I've ended up having a solid career, but I have no doubt I'd have been able to make a bigger impact if the family drama hadn't dragged on my time and focus for years (basically in supporting my mother through the experience). Society's answer (at least in America) is basically to say "you could cut out your family if they're destructive" but that goes against one of the strongest aspects of human nature (to have a family). Choosing between chaos and isolation is a terrible dichotomy and a big job of society should be to ensure individuals have other warmer options.
Also identifying earlier and stopping those chaos agents like my father is an area of study that is under-researched compared to the upside for society in solving that problem. There are folks like the Mind Research Network (https://www.mrn.org/) working on this but IMO should be getting 10x the funding they currently do.
- 2 points
- I suspect this was submitted and possibly upvoted because one of the Smothers Brothers died yesterday: https://apnews.com/article/tom-smothers-dies-smothers-brothe...
- I dove into the science of psychopathy a few years ago after a family incident where my father imploded his life and much of the family's through self-destructive activities and there is actually a bunch of smart people studying the neuroscience of pyschopaths.
Two great books to start with:
- The Science of Evil: https://www.amazon.com/Science-Evil-Empathy-Origins-Cruelty/...
- The Psychopath Whisperer: The Science of Those Without Conscious: https://www.amazon.com/Psychopath-Whisperer-Science-Without-...
- A big problem (from my experience) is the cultural shift from high school to university. Our high schools do a terrible job at preparing students for college. Just off the top of my head:
- High school classes are typically too easy
- So kids develop poor study habits which don't serve them well for college material
- And most high school teachers are bad at getting kids excited about the subject because they're exhausted themselves from babysitting and treat the work as a job. College professors can be bad at "teaching" but for different reasons (being researchers first and foremost). This disconnect in the reasons for bad education being different in environments is also not taught well to kids ahead of time (because who in this formula would? Requires good parenting or very self-conscious teachers at all levels).
There are definitely exceptions to this rule, but they are too few to solve the overarching problems.
- A legend of the business and investing world.
Acquired podcast got to do one of (possible the) last interviews with him a month ago for anyone looking for recent content from him: https://www.acquired.fm/episodes/charlie-munger
Also Stripe Press is releasing just next week a book (technically re-release of an older book) on his collected thoughts: https://press.stripe.com/poor-charlies-almanack
- It was officially started in 2005 with a format much closer to traditional magazines. It was a phase when people were making "e-zines", high quality PDFs distributed monthly: https://en.wikipedia.org/wiki/The_Escapist_(magazine)
*Edit: I'm referring to the whole publication, parent comment might've strictly meant the Zero Punctuation show.
- As I posted in the Aftermath/Kotaku thread[1], I worked as a contract writer for The Escapist from 2005-2009, and it was literally my first job ever. Yahtzee resigning is what I consider the end of the publication, as he's been the most consistent contributor for nearly two decades and most likely the primary revenue generator. It's very sad to see a huge part of my past coming to a close.
- I worked as a contract writer for The Escapist from 2005-2009 and consider Yahtzee's resignation the end of the publication. Zero Punctuation was most likely the primary revenue driver as I know how he had an outsized impact even back in the heyday when all of the content was getting accolades.
- I second all the questions here, having worked in consumer lending alongside Risk teams as well. The big questions alongside these that come to my mind are:
1. Who has the lending licenses? Given the website footer disclaimer says Pier isn't the lender. I would guess it's either a banking partnership or a license arrangement carried over from Stilt/JG Wentworth?
2. How does Pier think of itself in relation to someone like LoanPro (who from my industry conversations has had a lot of positive momentum as the best origination software) and the other origination/servicing platforms? I gather the idea is a bundled "origination + decision + servicing" platform.
LoanPro doesn't do the decision engine piece itself, but from what I gathered it was partially due to the precisely the complexity and compliance risks parent commenter noted.
Definitely best of luck to the team, as the space can always use better software than what I've seen at older institutions.
- Uber, Netflix and the online content streaming services. These are probably the most prominent examples from this recent 2010s era.
- 6 points
- 3 points
- 111 points
- I took a philosophy class my freshman year of undergrad where we read his book "On Bullshit". Very memorable experience that made it clear (especially to a young person) that philosophy could be fun, enlightening, written in understandable language, and about "anything" without pretension. It's been years since I last read it and am still inclined to recommend his work based on the positive experience I had. Sad to hear this news.
https://www2.csudh.edu/ccauthen/576f12/frankfurt__harry_-_on...
- 1. When the grandparent post said "corporations", they probably mean something like the Blackstones & Blackstones of the world, who are not small and both mentioned in this article about their residential housing activity: https://slate.com/business/2021/06/blackrock-invitation-hous...
2. Having lived near Ashville, NC for two years, it's a particularly beautiful region and attractive vacation spot, so I especially believe it would attract corporate buyers looking to build an Airbnb vacation portfolio. Also not far from Atlanta, where Invitation Homes (a corporate buyer of residential housing) is active.
- 2 points
- I moved to Greenville, South Carolina when COVID hit and absolutely loved it. Perfect weather, good balance of space, cost, and still enough active cities if one wants to eat at nice restaurants.
Glad I read the post as I hadn't heard of Malloy before. Excuse me if I missed the answer to this, but: How much do you as Findly/Conversion Pattern implement the Semantic Layer on behalf of your users (and if so, I assume you have some process for auto-generating the Malloy models), or do your users have to do something to input the semantics themselves?