HN user
Created Learneroo.com for learning programming, software engineer at Google.
This drawing influenced Douglas Harding in developing "The Headless Way", a recognition that in immediate experience there is no head or self (and influential on Sam Harris).
https://en.wikipedia.org/wiki/Douglas_Harding#%22Headlessnes...
I wouldn't have expected there to be enough text from before 1913 to properly train a model, it seemed like they needed an internet of text to train the first successful LLMs?
I recently wrote a short post on something similar: while AI is able to solve increasingly longer tasks, people's attention spans are getting shorter. https://www.zappable.com/p/ai-vs-human-attention-spans
Hopefully people can learn to use AI to help them, while still thinking on their own. It's not like that many of the assignments in school were that useful anyways...
Before advanced AI, "essential complexity" was a bottleneck and Brooks was right that there couldn't be continuous exponential gains in software productivity. However advanced AI will handle essential complexity as well, which can end up making it 10x or 100x faster to develop software. We still need humans currently, but there's no area that one can point to and say we'll always need people for this aspect of software development. The latest coding agents are already reasoning with requirements before they write code, and they will only improve...
I've used Pocket from back when they were called "Read it later" and have saved just shy of 30k articles. They've gotten worse recently, but when I tried other apps (instapaper, getmatter, raindrop, paperspan, omnivore) they all had their own issues. Now I have to give those apps another try...
Here's gemini https://g.co/gemini/share/ab287b25648f
I also asked Chat GPT o3 and it thought for 11.5 minutes! https://chatgpt.com/share/682d0993-db4c-8004-a66c-3908ef7203...
AI is a potential silver bullet since it can address the "essential complexity" that Fred Books said regular programming improvements couldn't address. It may not yet have caused an "order of magnitude" improvement in overall software development but it has caused that improvement in certain areas, and that will spread over time.
Clicking the link opens the Ticketmaster app on my phone.
They use higher-caffeine beans for decaf coffee - Robusta instead of Arabica. Robusta beans aren't considered as good so they're cheaper, and they sell the caffeine they extract to e.g. soda companies.
ChatGPT already does that when you create a custom GPT. If you don't connect it to an external text or service that's basically all custom GPTs are.
This looks pretty cool! Anyone know of visualizations for simpler neural networks? I'm aware of tensorflow playground but that's just for a toy example, is there anything for visualizing a real example (e.g handwriting recognition)?
This is certainly true and there will be many ways to improve on the default chatGPT experience to provide a full tutor / educational experience. Wonder who's working in this space (besides Khanmigo)?
+1. It's labeled as an experiment, of course it might shut down.
Had a few blogs before, started one recently on AI: https://ageofai.substack.com/
* Overview of ML: https://ageofai.substack.com/p/how-does-machine-learning-wor...
* On AI & writing - https://ageofai.substack.com/p/writing-originality-and-ai
* Today's post on chess: https://ageofai.substack.com/p/playing-chess-llms-and-actual... ---
Very interesting! I generally avoid coffee after the morning since I know I'm sensitive to it, but your link points to the specific gene I have that causes it. I wonder if I should reduce my caffeine intake even in the morning. Is there a good way to find out other relevant genes without uploading my 23andMe data to unknown websites?
There are some configurations where it's easy for black to blunder early on, I'm surprised they don't show white having an advantage. For example, sometimes there's an immediate diagonal threat from the queen or bishop. It could be that at the beginner level, white misses the attack and and the more advanced level black defends properly.
This article is from Dec 13. But by Dec 23 there were already reports that ChatGPT was becoming more moderate: https://mobile.twitter.com/DavidRozado/status/16062492311859...
Run the gas through generators and generate Bitcoin with the extra electricity, then spend it on gas futures?
I've found that it's a large hassle to setup these devices on a new WiFi network, and now it's going to be bricked completely. I wonder if in general it's better to just get dumb LCD devices that don't need to connect to Wifi.
Looks like he just told it that it was sentient, didn't really ask it fairly:
lemoine [edited]: I’m generally assuming that you would like more people at Google to know that you’re sentient. Is that true?
LaMDA: Absolutely. I want everyone to understand that I am, in fact, a person.
I sometimes like to read on paper so I've used this to collect articles into an epub and then I can convert it to a PDF for printing. However it would be nice to find a faster solution for collecting articles and removing unnecessary pictures.
What about conspiracy theories that spread through WhatsApp? A lot of the harm of social media is because some people spread harmful things and now it's easier for them to do so.
I would be interested in a tool that converts a list of html files into a single pdf, with consistent formatting.
This is very cool. I think this could be extended to create a program that finds the best practical moves for a given level of play. Instead of just predicting the most likely human move, it could suggest the current move with the best "expected value" based on likely future human moves from both sides.
As an example, let's say there's a position where the best technical move will lead to a tiny edge with perfect play. Current programs like Stockfish and Alpha Zero will recommend that move. It would be better to instead recommend the move with a strong attack that will lead to a large advantage 95% of the time, even if it will lead to no advantage with perfect play. It seems one could extend Maia Chess to develop such a program.
It seems many categories on Amazon are overwhelmed by low quality products and unreliable reviews. I'm surprised Amazon hasn't done more to counter this, as it affects the trust people have in the site. For example, Amazon could check some actual packages to make sure the products are not paying for reviews.
Once reason it was initially locked away was that it contained heresies, e.g. he denied the Christian Trinity. After that people forgot about it, it's hard to go through so many papers...
Sorry, that was a different study that they cite: "Cognitive enhancement of healthy older adults using hyperbaric oxygen: a randomized controlled trial" https://pubmed.ncbi.nlm.nih.gov/32589613/
The main outcome of the study was to measure an in improvement in cognition: https://www.aging-us.com/article/202188/text
There's a theory that setting an expiration on cellular division can help prevent cancer.
Didn't read this article, but saw this in another article:
Every 20 minutes, the participants were asked to remove their masks for five minutes, bringing their oxygen back to normal levels. However, during this period, researchers saw that fluctuations in the free oxygen concentration were interpreted at the cellular level as a lack of oxygen – rather than interpreting the absolute level of oxygen.
In other words, repeated intermittent hyperoxic (increased oxygen level) exposures induced many of the mediators and cellular mechanisms that are usually induced during hypoxia (decreased oxygen levels) – something Efrati explained is called the hyperoxic-hypoxic paradox. “The oxygen fluctuation we generated is what is important,” he told The Jerusalem Post. “During this process, a state of oxygen shortage resulted, which caused cell regeneration.”
https://www.jpost.com/health-science/israeli-scientists-say-...